main.bib

@Conference{symeonidis_etal_2008,
  author        = {Symeonidis, P. and Nanopoulos, A. and Manolopoulos, Y.},
  title         = {Tag recommendations based on tensor dimensionality reduction},
  year          = {2008},
  pages         = {43-50},
  note          = {cited By},
  __markedentry = {[felipe:1]},
  abstract      = {Social tagging is the process by which many users add metadata in the form of keywords, to annotate and categorize information items (songs, pictures, web links, products etc.). Collaborative tagging systems recommend tags to users based on what tags other users have used for the same items, aiming to develop a common consensus about which tags best describe an item. However, they fail to provide appropriate tag recommendations, because: (i) users may have different interests for an information item and (ii) information items may have multiple facets. In contrast to the current tag recommendation algorithms, our approach develops a unified framework to model the three types of entities that exist in a social tagging system: users, items and tags. These data is represented by a 3-order tensor, on which latent semantic analysis and dimensionality reduction is performed using the Higher Order Singular Value Decomposition (HOSVD) technique. We perform experimental comparison of the proposed method against two state-of-the-art tag recommendations algorithms with two real data sets (Last.fm and BibSonomy). Our results show significant improvements in terms of effectiveness measured through recall/precision. © 2008 ACM.},
  affiliation   = {Department of Informatics, Aristotle University, Thessaloniki 54124, Greece},
  document_type = {Conference Paper},
  doi           = {10.1145/1454008.1454017},
  journal       = {RecSys'08: Proceedings of the 2008 ACM Conference on Recommender Systems},
  keywords      = {rank2},
  source        = {Scopus},
  url           = {http://dl.acm.org/ft_gateway.cfm?id=1454017&type=pdf&CFID=975119194&CFTOKEN=74174692},
}

@Article{li_etal_2009,
  author        = {X. Li and C. G. M. Snoek and M. Worring},
  title         = {Learning Social Tag Relevance by Neighbor Voting},
  journal       = {IEEE Transactions on Multimedia},
  year          = {2009},
  volume        = {11},
  number        = {7},
  pages         = {1310-1322},
  month         = {Nov},
  issn          = {1520-9210},
  __markedentry = {[felipe:1]},
  abstract      = {Social image analysis and retrieval is important for helping people organize and access the increasing amount of user tagged multimedia. Since user tagging is known to be uncontrolled, ambiguous, and overly personalized, a fundamental problem is how to interpret the relevance of a user-contributed tag with respect to the visual content the tag is describing. Intuitively, if different persons label visually similar images using the same tags, these tags are likely to reflect objective aspects of the visual content. Starting from this intuition, we propose in this paper a neighbor voting algorithm which accurately and efficiently learns tag relevance by accumulating votes from visual neighbors. Under a set of well-defined and realistic assumptions, we prove that our algorithm is a good tag relevance measurement for both image ranking and tag ranking. Three experiments on 3.5 million Flickr photos demonstrate the general applicability of our algorithm in both social image retrieval and image tag suggestion. Our tag relevance learning algorithm substantially improves upon baselines for all the experiments. The results suggest that the proposed algorithm is promising for real-world applications.},
  doi           = {10.1109/TMM.2009.2030598},
  keywords      = {image classification;image retrieval;learning (artificial intelligence);Flickr photo;image ranking;learning social tag relevance;neighbor voting algorithm;social image analysis;social image retrieval;tag ranking;tag relevance learning algorithm;user tagged multimedia;visual content;visual neighbor;Multimedia indexing and retrieval;neighbor voting;social tagging;tag relevance learning},
}

@InProceedings{guillaumin_etal_2009,
  author        = {M. Guillaumin and T. Mensink and J. Verbeek and C. Schmid},
  title         = {TagProp: Discriminative metric learning in nearest neighbor models for image auto-annotation},
  booktitle     = {2009 IEEE 12th International Conference on Computer Vision},
  year          = {2009},
  pages         = {309-316},
  month         = {Sept},
  __markedentry = {[felipe:1]},
  abstract      = {Image auto-annotation is an important open problem in computer vision. For this task we propose TagProp, a discriminatively trained nearest neighbor model. Tags of test images are predicted using a weighted nearest-neighbor model to exploit labeled training images. Neighbor weights are based on neighbor rank or distance. TagProp allows the integration of metric learning by directly maximizing the log-likelihood of the tag predictions in the training set. In this manner, we can optimally combine a collection of image similarity metrics that cover different aspects of image content, such as local shape descriptors, or global color histograms. We also introduce a word specific sigmoidal modulation of the weighted neighbor tag predictions to boost the recall of rare words. We investigate the performance of different variants of our model and compare to existing work. We present experimental results for three challenging data sets. On all three, TagProp makes a marked improvement as compared to the current state-of-the-art.},
  doi           = {10.1109/ICCV.2009.5459266},
  issn          = {1550-5499},
  keywords      = {image processing;learning (artificial intelligence);TagProp;computer vision;discriminative metric learning;image auto-annotation;image similarity metrics;tag predictions;weighted nearest-neighbor model;word specific sigmoidal modulation;Computer vision;Content management;Histograms;Large-scale systems;Nearest neighbor searches;Predictive models;Shape;Testing;Video sharing;Vocabulary},
}

@Conference{auyeung_etal_2009,
  author          = {Au Yeung, C.-M. and Gibbins, N. and Shadbolt, N.},
  title           = {User-induced links in collaborative tagging systems},
  year            = {2009},
  pages           = {787-796},
  note            = {cited By},
  __markedentry   = {[felipe:1]},
  abstract        = {Collaborative tagging systems allow users to use tags to describe their favourite online documents. Two documents that are maintained in the collection of the same user and/or assigned similar sets of tags can be considered as related from the perspective of the user, even though they may not be connected by hyperlinks. We call this kind of implicit relations user-induced links between documents. We consider two methods of identifying user-induced links in collaborative tagging, and compare these links with existing hyperlinks on the Web. Our analyses show that user-induced links have great potentials to enrich the existing link structure of the Web. We also propose to use these links as a basis for predicting how documents would be tagged. Our experiments show that they achieve much higher accuracy than existing hyperlinks. This study suggests that by studying the collective behaviour of users we are able to enhance navigation and organisation of Web documents. Copyright 2009 ACM.},
  affiliation     = {University of Southampton, Southampton, SO17 1BJ, United Kingdom},
  author_keywords = {Collaborative tagging; Folksonomy; Hyperlink},
  document_type   = {Conference Paper},
  doi             = {10.1145/1645953.1646053},
  journal         = {International Conference on Information and Knowledge Management, Proceedings},
  source          = {Scopus},
  url             = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-74549153304&doi=10.1145%2f1645953.1646053&partnerID=40&md5=9efd6f75550f47f6f2b07537d6869c2d},
}

@Conference{martinez_etal_2009,
  author        = {Martínez, E. and Celma, O. and Sordo, M. and de Jong, B. and Serra, X.},
  title         = {Extending the folksonomies of freesound.org using content-based audio analysis},
  year          = {2009},
  pages         = {65-70},
  note          = {cited By},
  __markedentry = {[felipe:1]},
  abstract      = {This paper presents an in-depth study of the social tagging mechanisms used in Freesound.org, an online community where users share and browse audio files by means of tags and content-based audio similarity search. We performed two analyses of the sound collection. The first one is related with how the users tag the sounds, and we could detect some well-known problems that occur in collaborative tagging systems (i.e. polysemy, synonymy, and the scarcity of the existing annotations). Moreover, we show that more than 10% of the collection were scarcely annotated with only one or two tags per sound, thus frustrating the retrieval task. In this sense, the second analysis focuses on enhancing the semantic annotations of these sounds, by means of content- based audio similarity (autotagging). In order to "autotag" the sounds, we use a k-NN classifier that selects the available tags from the most similar sounds. Human assessment is performed in order to evaluate the perceived quality of the candidate tags. The results show that, in 77% of the sounds used, the annotations have been correctly extended with the proposed tags derived from audio similarity.},
  affiliation   = {Music Technology Group, Universitat Pompeu Fabra, Barcelona, Spain},
  document_type = {Conference Paper},
  journal       = {Proceedings of the 6th Sound and Music Computing Conference, SMC 2009},
  source        = {Scopus},
  url           = {http://mtg.upf.edu/files/publications/SMC09_emartinez_ocelma_msordo_bdejong_xserra.pdf},
}

@InProceedings{shen_etal_2009,
  author        = {C. Shen and J. Jiao and Y. Yang and B. Wang},
  title         = {Multi-instance multi-label learning for automatic tag recommendation},
  booktitle     = {2009 IEEE International Conference on Systems, Man and Cybernetics},
  year          = {2009},
  pages         = {4910-4914},
  month         = {Oct},
  __markedentry = {[felipe:1]},
  abstract      = {Tag services have recently become one of the most popular Internet services on the World Wide Web. Due to the fact that a Web page can be associate with multiple tags, previous research on tag recommendation mainly focuses on improving its accuracy or efficiency through multi-label learning algorithms. However, as a Web page can also be split into multiple sections and be represented as a bag of instances, multi-instance multi-label learning framework should fit this problem better. In this paper, we improve the performance of tag suggestion by using multi-instance multi-label learning. Each Web page is divided into a bag of instances. The experiments on real-word data from delicious suggest that our framework has better performance than traditional multi-label learning methods on the task of tag recommendation.},
  doi           = {10.1109/ICSMC.2009.5346261},
  issn          = {1062-922X},
  keywords      = {Internet;Web services;information filtering;learning (artificial intelligence);Internet services;Web page;World Wide Web;automatic tag recommendation;multiinstance multilabel learning algorithm;Computer science;Cybernetics;Learning systems;Machine learning;Machine learning algorithms;Tagging;USA Councils;Web and internet services;Web pages;Web sites;machine learning;multi-instance;multi-label;tag recommendation, rank1},
}

@InProceedings{nikolopoulos_etal_2009,
  author        = {S. Nikolopoulos and E. Chatzilari and E. Giannakidou and I. Kompatsiaris},
  title         = {Towards fully un-supervised methods for generating object detection classifiers using social data},
  booktitle     = {2009 10th Workshop on Image Analysis for Multimedia Interactive Services},
  year          = {2009},
  pages         = {230-233},
  month         = {May},
  __markedentry = {[felipe:1]},
  abstract      = {In this work a framework for constructing object detection classifiers using weakly annotated social data is proposed. Social information is combined with computer vision techniques to automatically obtain a set of images annotated at region-detail. All assumptions made to automate the proposed framework are driven by the reasonable expectation that due to the collaborative aspect of social data, linguistic descriptions and visual representations will start to converge on common concepts, as the scale of the analyzed dataset increases. Comparison tests performed against manually trained object detectors showed that comparable performance can be achieved.},
  doi           = {10.1109/WIAMIS.2009.5031475},
  issn          = {2158-5873},
  keywords      = {computer vision;object detection;computer vision techniques;fully unsupervised methods;linguistic descriptions;object detection classifiers;visual representations;Computer vision;Data analysis;Detectors;Feature extraction;Image generation;Image segmentation;Informatics;Object detection;Object recognition;Telematics, rank1},
}

@InProceedings{jin_etal_2010,
  author        = {Y. Jin and R. Li and Z. Lu and K. Wen and X. Gu},
  title         = {Topic-Sensitive Tag Ranking},
  booktitle     = {2010 20th International Conference on Pattern Recognition},
  year          = {2010},
  pages         = {629-632},
  month         = {Aug},
  __markedentry = {[felipe:1]},
  abstract      = {Social tagging is an increasingly popular way to describe and classify documents on the web. However, the quality of the tags varies considerably since the tags are authored freely. How to rate the tags becomes an important issue. In this paper, we propose a topic-sensitive tag ranking (TSTR) approach to rate the tags on the web. We employ a generative probabilistic model to associate each tag with a distribution of topics. Then we construct a tag graph according to the co-tag relationships and perform a topic-level random walk over the graph to suggest a ranking score for each tag at different topics. Experimental results validate the effectiveness of the proposed tag ranking approach.},
  doi           = {10.1109/ICPR.2010.159},
  issn          = {1051-4651},
  keywords      = {Internet;graph theory;probability;cotag relationships;generative probabilistic model;social tagging;tag graph;tag ranking approach;topic-level random walk;topic-sensitive tag ranking;Computer architecture;Graphical user interfaces;Java;Service oriented architecture;Tagging;Unified modeling language;Web search, rank1},
}

@Article{han_etal_2010,
  author        = {Y. Han and F. Wu and Y. Zhuang and X. He},
  title         = {Multi-Label Transfer Learning With Sparse Representation},
  journal       = {IEEE Transactions on Circuits and Systems for Video Technology},
  year          = {2010},
  volume        = {20},
  number        = {8},
  pages         = {1110-1121},
  month         = {Aug},
  issn          = {1051-8215},
  __markedentry = {[felipe:1]},
  abstract      = {Due to the visually polysemous barrier, videos and images may be annotated by multiple tags. Discovering the correlations among different tags can significantly help predicting precise labels for videos and images. Many of recent studies toward multi-label learning construct a linear subspace embedding with encoded multi-label information, such that data points sharing many common labels tend to be close to each other in the embedded subspace. Motivated by the advances of compressive sensing research, a sparse representation that selects a compact subset to describe the input data can be more discriminative. In this paper, we propose a sparse multi-label learning method to circumvent the visually polysemous barrier of multiple tags. Our approach learns a multi-label encoded sparse linear embedding space from a related dataset, and maps the target data into the learned new representation space to achieve better annotation performance. Instead of using l1-norm penalty (lasso) to induce sparse representation, we propose to formulate the multi-label learning as a penalized least squares optimization problem with elastic-net penalty. By casting the video concept detection and image annotation tasks into a sparse multi-label transfer learning framework in this paper, ridge regression, lasso, elastic net, and the multi-label extended sparse discriminant analysis methods are, respectively, well explored and compared.},
  doi           = {10.1109/TCSVT.2010.2057015},
  keywords      = {image retrieval;learning (artificial intelligence);optimisation;sparse matrices;video signal processing;image annotation tasks;least squares optimization problem;linear subspace;multilabel transfer learning;sparse representation;video concept detection;visually polysemous barrier;Image annotation;multi-label learning;sparse representation;transfer learning;video concept detection, rank1},
}

@InProceedings{krestel_fankhauser_2010,
  author        = {R. Krestel and P. Fankhauser},
  title         = {Language Models and Topic Models for Personalizing Tag Recommendation},
  booktitle     = {2010 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology},
  year          = {2010},
  volume        = {1},
  pages         = {82-89},
  month         = {Aug},
  __markedentry = {[felipe:1]},
  abstract      = {More and more content on the Web is generated by users. To organize this information and make it accessible via current search technology, tagging systems have gained tremendous popularity. Especially for multimedia content they allow to annotate resources with keywords (tags) which opens the door for classic text-based information retrieval. To support the user in choosing the right keywords, tag recommendation algorithms have emerged. In this setting, not only the content is decisive for recommending relevant tags but also the user's preferences. In this paper we introduce an approach to personalized tag recommendation that combines a probabilistic model of tags from the resource with tags from the user. As models we investigate simple language models as well as Latent Dirichlet Allocation. Extensive experiments on a real world dataset crawled from a big tagging system show that personalization improves tag recommendation, and our approach significantly outperforms state-of-the-art approaches.},
  doi           = {10.1109/WI-IAT.2010.29},
  keywords      = {Internet;data mining;information retrieval;probability;recommender systems;text analysis;Web;data mining;language models;latent Dirichlet allocation;probabilistic model;search technology;tag recommendation algorithms;tagging system;tagging systems;text-based information retrieval;topic models;Clustering;Data mining;Personalization, rank1},
}

@InProceedings{si_sun_2010,
  author        = {X. Si and M. Sun},
  title         = {Tag Allocation Model: Model Noisy Social Annotations by Reason Finding},
  booktitle     = {2010 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology},
  year          = {2010},
  volume        = {1},
  pages         = {413-416},
  month         = {Aug},
  __markedentry = {[felipe:1]},
  abstract      = {We propose the Tag Allocation Model (TAM) to model social annotation data. TAM is a probabilistic generative model, its key feature is finding the latent reason for each tag. A latent reason can be any discrete features of the document (such as words) or a global noise variable. Inferring the reason for each tag helps TAM reduce the ambiguity of a document with multiple tags. By introducing noise as a reason, TAM can handle noise tags naturally. We perform experiments on three real world data sets. The results show that TAM outperforms state-of-the-art approaches in both held-out perplexity and tag recommendation accuracy.},
  doi           = {10.1109/WI-IAT.2010.85},
  keywords      = {inference mechanisms;information retrieval;portals;probability;TAM;global noise variable;noisy social annotation data model;probabilistic generative model;real world data sets;reason finding;tag allocation model;probabilistic model;recommendation;social annotation;tagging, rank1},
}

@InProceedings{hu_etal_2012,
  author        = {R. Hu and T. He and F. Li and P. Hu},
  title         = {Tag recommendation based on tag-topic model},
  booktitle     = {2012 IEEE 2nd International Conference on Cloud Computing and Intelligence Systems},
  year          = {2012},
  volume        = {03},
  pages         = {1501-1505},
  month         = {Oct},
  __markedentry = {[felipe:1]},
  abstract      = {With the rapid increase of the social websites most social tagging systems are allowing users to share and to label various kinds of resources with their favorite tags. However the uncontrolled use of tags makes the resources attached with some irrelevant even noise tags. To solve the problem this paper proposes a tag-topic model based approach to recommend tags for resources which elicits latent topics from resources and maps new resources to these latent topics so as to recommend the most appropriate tags for the resources. The experimental results show the effectiveness of the proposed approach.},
  doi           = {10.1109/CCIS.2012.6664635},
  issn          = {2376-5933},
  keywords      = {recommender systems;social networking (online);latent topics;social Websites;social tagging systems;tag recommendation;tag-topic model;Blogs;Collaboration;Educational institutions;Noise;Resource management;Semantics;Tagging;tag recommendation;tag-topic model},
}

@InProceedings{chidlovskii_2012,
  author        = {B. Chidlovskii},
  title         = {Tag Ranking by Linear Relational Neighbourhood Propagation},
  booktitle     = {2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining},
  year          = {2012},
  pages         = {184-188},
  month         = {Aug},
  __markedentry = {[felipe:1]},
  abstract      = {We propose a tag recommendation method which can assist users in tagging process by suggesting relevant tags. The method is based on query-based ranking on relational multi-type graphs which capture the annotation relationship between objects and tags, as well as the object similarity and tag correlation. The additional advance consists in extending the linear neighbourhood propagation to the relational graphs with the Laplacian regularization framework. We report evaluation results on a large-scale Flickr data set.},
  doi           = {10.1109/ASONAM.2012.40},
  keywords      = {graph theory;query processing;recommender systems;social networking (online);Laplacian regularization framework;large-scale Flickr data set;linear relational neighbourhood propagation;object similarity;object-tag annotation relationship;query-based ranking;relational graphs;relational multitype graphs;tag correlation;tag ranking;tag recommendation method;tagging process;Correlation;Image edge detection;Image reconstruction;Laplace equations;Minimization;Tagging;Vectors;Laplacian regularization;Tag ranking;linear neighbourhood propagation;relational graphs},
}

@Article{vanleeuwen_puspitaningrum_2012,
  author        = {Van Leeuwen, M. and Puspitaningrum, D.},
  title         = {Improving tag recommendation using few associations},
  journal       = {Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)},
  year          = {2012},
  volume        = {7619 LNCS},
  pages         = {184-194},
  note          = {cited By},
  __markedentry = {[felipe:1]},
  abstract      = {Collaborative tagging services allow users to freely assign tags to resources. As the large majority of users enters only very few tags, good tag recommendation can vastly improve the usability of tags for techniques such as searching, indexing, and clustering. Previous research has shown that accurate recommendation can be achieved by using conditional probabilities computed from tag associations. The main problem, however, is that enormous amounts of associations are needed for optimal recommendation. We argue and demonstrate that pattern selection techniques can improve tag recommendation by giving a very favourable balance between accuracy and computational demand. That is, few associations are chosen to act as information source for recommendation, providing high-quality recommendation and good scalability at the same time. We provide a proof-of-concept using an off-the-shelf pattern selection method based on the Minimum Description Length principle. Experiments on data from Delicious, LastFM and YouTube show that our proposed methodology works well: applying pattern selection gives a very favourable trade-off between runtime and recommendation quality. © Springer-Verlag Berlin Heidelberg 2012.},
  affiliation   = {Dept. of Information and Computing Sciences, Universiteit Utrecht, Netherlands; Dept. of Computer Science, KU Leuven, Belgium},
  document_type = {Conference Paper},
  doi           = {10.1007/978-3-642-34156-4_18},
  source        = {Scopus},
  url           = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84868032350&doi=10.1007%2f978-3-642-34156-4_18&partnerID=40&md5=9207f41d00f57a8bec31e43559703813},
}

@Conference{trabelsi_etal_2012,
  author          = {Trabelsi, C. and Moulahi, B. and Yahia, S.B.},
  title           = {HMM-CARe: Hidden Markov models for context-aware tag recommendation in folksonomies},
  year            = {2012},
  pages           = {957-961},
  note            = {cited By},
  __markedentry   = {[felipe:1]},
  abstract        = {Collaborative tagging systems allow users to manually annotate web resources with freely chosen keywords aka tags without any restriction to a certain vocabulary. The resulting collection of all these users annotations constitute the so-called folksonomy. Such systems typically provide simple tag recommendations skills to increase the number of tags assigned to resources. In this this paper, we propose a novel Hidden Markov Model (HMM) based approach, called HMM-CARE, for tags recommendation. Specifically, we extend the HMM to include user's tagging intents, formally represented as triadic concepts. Carried out experiments emphasize the relevance of our proposal and open many thriving issues. © 2012 ACM.},
  affiliation     = {Faculty of Sciences of Tunis, University Tunis El-Manar, Tunis, Tunisia},
  author_keywords = {folksonomies; Hidden Markov Models; tag recommendation; triadic concepts},
  document_type   = {Conference Paper},
  doi             = {10.1145/2245276.2245461},
  journal         = {Proceedings of the ACM Symposium on Applied Computing},
  keywords        = {rank1},
  source          = {Scopus},
  url             = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84863575351&doi=10.1145%2f2245276.2245461&partnerID=40&md5=c1941e7f351c3dba08f4ff967543524b},
}

@InProceedings{kakade_kakade_2013,
  author        = {S. R. Kakade and N. R. Kakade},
  title         = {A novel approach to link semantic gap between images and tags via probabilistic ranking},
  booktitle     = {2013 IEEE International Conference on Computational Intelligence and Computing Research},
  year          = {2013},
  pages         = {1-6},
  month         = {Dec},
  __markedentry = {[felipe:1]},
  abstract      = {Nowadays there is tremendous fame of social networking sites that leads to a broad investigation in tag-based social image search. Both visual features and tags plays important role in the research field. Existing approach uses tags and visual features consecutively or independently to estimate the relevance of images. In this paper we tackle the problems of Automatic Image Annotation (AIA) for image retrieval that delivers the global image search engine that covers different applications such as Image to Image Retrieval, Image to Tag suggestions, Tag to Image Retrieval and Tag to Tag suggestion which is based on user contributed tags on the websites such as flicker. The proposed approach comprises of global feature extractions of images. Then we predicate relationship between image and tag by using asymmetric graph as a probabilistic view. Once the relationship is found ranking is employed with the help of t-step random walk model. Finally pseudo relevance feedback is applied on the ranked images. Additionally experimental analysis of above model is conducted on Corel image dataset to illustrate the effectiveness of this approach.},
  doi           = {10.1109/ICCIC.2013.6724166},
  keywords      = {feature extraction;image retrieval;random processes;search engines;AIA;Corel image dataset;Flicker;Web sites;asymmetric graph;automatic image annotation;global feature extractions;global image search engine;image retrieval;link semantic gap;probabilistic ranking;pseudo relevance feedback;random walk model;social networking sites;tag suggestion;tag-based social image search;user contributed tags;visual features;Bipartite graph;Feature extraction;Image color analysis;Image edge detection;Image retrieval;Vectors;Visualization;Automatic Image Annotation;Content-based image retrieval;Image annotation;Random Walk;Text-based image retrieval},
}

@InProceedings{liu_etal_2013,
  author        = {S. Liu and Y. Zhu and J. Guo and Y. Wang and X. Cheng and Y. Liu},
  title         = {A Blending Method for Automated Social Tagging},
  booktitle     = {2013 IEEE/WIC/ACM International Joint Conferences on Web Intelligence (WI) and Intelligent Agent Technologies (IAT)},
  year          = {2013},
  volume        = {1},
  pages         = {115-120},
  month         = {Nov},
  __markedentry = {[felipe:1]},
  abstract      = {Social tagging has grown in popularity on the web due to its effectiveness in organizing and accessing web pages. This short paper addresses the problem of automated social tagging, which aims to predict tags for web pages automatically and help with future navigation, filtering or search. We explore and find three foundations of the collaborative tags in social tagging services, that are consistency, sharability and stability. The complementary advantages are studied among three well-known methods, i.e. TF-weighted keyword extraction, collaborative filtering approach, and Corr-LDA (correspondence latent Dirichlet allocation) topic model. We then propose a blending model for automated social tagging to emphasize all the foundations, which linearly combines those tags generated by the three methods, and a permutation probability model is built to learn the linear blending. With the experiments on 50,000 training and 10,000 testing web pages from Delicious database, the results show that our blending method outperforms the four baselines. Furthermore, compared with both topic models, Corr-LDA and mixed membership LDA, our approach results in 14.2% and 25.6% of NDCG10 improvement separately.},
  doi           = {10.1109/WI-IAT.2013.17},
  keywords      = {Internet;collaborative filtering;data integrity;social networking (online);Corr-LDA topic model;Del.icio.us database;TF-weighted keyword extraction;automated social tagging services;blending model;collaborative filtering approach;collaborative tags;consistency;correspondence latent Dirichlet allocation topic model;linear blending;permutation probability model;sharability;stability;testing Web pages;training Web pages;Accuracy;Collaboration;Data models;Equations;Mathematical model;Stability analysis;Tagging;automatic annotation;collaborative filtering;social tagging;topic model, rank1},
}

@InProceedings{sattigeri_etal_2014,
  author        = {P. Sattigeri and J. J. Thiagarajan and M. Shah and K. N. Ramamurthy and A. Spanias},
  title         = {A scalable feature learning and tag prediction framework for natural environment sounds},
  booktitle     = {2014 48th Asilomar Conference on Signals, Systems and Computers},
  year          = {2014},
  pages         = {1779-1783},
  month         = {Nov},
  __markedentry = {[felipe:1]},
  abstract      = {Building feature extraction approaches that can effectively characterize natural environment sounds is challenging due to the dynamic nature. In this paper, we develop a framework for feature extraction and obtaining semantic inferences from such data. In particular, we propose a new pooling strategy for deep architectures, that can preserve the temporal dynamics in the resulting representation. By constructing an ensemble of semantic embeddings, we employ an l1-reconstruction based prediction algorithm for estimating the relevant tags. We evaluate our approach on challenging environmental sound recognition datasets, and show that the proposed features outperform traditional spectral features.},
  doi           = {10.1109/ACSSC.2014.7094773},
  keywords      = {acoustic signal processing;feature extraction;learning (artificial intelligence);Iι-reconstruction based prediction algorithm;environmental sound recognition;feature extraction approach;scalable feature learning;semantic inferences;tag prediction framework;Computational modeling;Computer architecture;Correlation;Dictionaries;Feature extraction;Predictive models;Semantics, rank1},
}

@Article{zhang_etal_2014,
  author        = {Yin Zhang and Deng Yi and Baogang Wei and Yueting Zhuang},
  title         = {A GPU-accelerated non-negative sparse latent semantic analysis algorithm for social tagging data},
  journal       = {Information Sciences},
  year          = {2014},
  volume        = {281},
  pages         = {687 - 702},
  issn          = {0020-0255},
  note          = {Multimedia Modeling},
  __markedentry = {[felipe:1]},
  doi           = {http://dx.doi.org/10.1016/j.ins.2014.04.047},
  keywords      = {Non-negative Sparse LSA, Social tagging, GPU computing, Tag recommendation, Image classification, rank1},
  url           = {http://www.sciencedirect.com/science/article/pii/S002002551400512X},
}

@InProceedings{charte_etal_2015,
  author        = {F. Charte and A. J. Rivera and M. J. del Jesus and F. Herrera},
  title         = {QUINTA: A question tagging assistant to improve the answering ratio in electronic forums},
  booktitle     = {IEEE EUROCON 2015 - International Conference on Computer as a Tool (EUROCON)},
  year          = {2015},
  pages         = {1-6},
  month         = {Sept},
  __markedentry = {[felipe:1]},
  abstract      = {The Web is broadly used nowadays to obtain information about almost any topic, from scientific procedures to cooking recipes. Electronic forums are very popular, with thousands of questions asked and answered every day. Correctly tagging the questions posted by users usually produces quicker and better answers by other users and experts. In this paper a prototype of a system aimed to assist the users while tagging their questions is proposed. To accomplish this task, firstly the text of each post is processed to produce a multilabel dataset, then a lazy nearest neighbor multilabel classification algorithm is used to predict the tags on new posts. The obtained results are promising, opening the door to the developing of a full automated system for this task.},
  doi           = {10.1109/EUROCON.2015.7313677},
  keywords      = {Web sites;data mining;pattern classification;question answering (information retrieval);text analysis;QUINTA;answering ratio improvement;electronic forums;lazy nearest neighbor multilabel classification algorithm;multilabel dataset;post-text processing;question tagging assistant;tag prediction;Computer science;Electronic mail;Measurement;Pipelines;Prediction algorithms;Tagging;Text mining, rank1},
}

@InProceedings{kataria_agarwal_2015,
  author        = {S. Kataria and A. Agarwal},
  title         = {Distributed Representations for Content-Based and Personalized Tag Recommendation},
  booktitle     = {2015 IEEE International Conference on Data Mining Workshop (ICDMW)},
  year          = {2015},
  pages         = {1388-1395},
  month         = {Nov},
  __markedentry = {[felipe:1]},
  abstract      = {We consider the problem of learning distributed representations for documents from their content and associated tags, and of distributed representations of users from documents and tags provided by users. The documents, words, and tags are represented as low-dimensional vectors and are jointly learned with a multi-layered neural language model. We propose a two stage method where in the first stage which consists of two layers, we exploit the corpus wide topic-level information contained in tags to model one layer of neural language model and use document level words sequence information to model other layer of the proposed architecture. In the second stage, we use thus obtained document and tags representations to learn user representations. We utilize these jointly trained vector representations for personalized tag recommendation tasks. Our experiments on two widely used bookmarking datasets show a significant improvements for quality of recommendations. These continuous vector representations has the added advantages of conceptually meaningful which we show by our qualitative analysis on tag suggestion tasks.},
  doi           = {10.1109/ICDMW.2015.240},
  keywords      = {indexing;learning (artificial intelligence);natural language processing;neural nets;recommender systems;search problems;book-marking datasets;content-based recommendation;corpus wide topic-level information;distributed representation learning problem;document level word sequence information;jointly trained vector representations;low-dimensional vectors;multilayered neural language model;personalized tag recommendation tasks;two stage method;Conferences;Context;Context modeling;Optimization;Semantics;Tagging;Training, rank4},
}

@Article{zhang_etal_2015,
  author        = {Jing Zhang and Xin Liu and Li Zhuo and Chao Wang},
  title         = {Social images tag ranking based on visual words in compressed domain},
  journal       = {Neurocomputing},
  year          = {2015},
  volume        = {153},
  pages         = {278 - 285},
  issn          = {0925-2312},
  __markedentry = {[felipe:1]},
  doi           = {http://dx.doi.org/10.1016/j.neucom.2014.11.027},
  keywords      = {Social images, Tag ranking, Visual words, Compressed domain, Neighbor voting},
  url           = {http://www.sciencedirect.com/science/article/pii/S0925231214015604},
}

@Article{zhao_etal_2015,
  author          = {Zhao, W. and Guan, Z. and Liu, Z.},
  title           = {Ranking on heterogeneous manifolds for tag recommendation in social tagging services},
  journal         = {Neurocomputing},
  year            = {2015},
  volume          = {148},
  pages           = {521-534},
  issn            = {0925-2312},
  note            = {cited By},
  __markedentry   = {[felipe:1]},
  abstract        = {Nowadays, most social Websites allow users to annotate resources (such as Web pages and images) with keywords, i.e. tags. Collaborative tagging data reflects the semantic perception of users, thus providing valuable information for the related recommendation problems, e.g. tag recommendation, resource recommendation. In this paper, we tackle the problem of personalized tag recommendation in social tagging services by generalizing the traditional manifold ranking idea. Specifically, we model the complex relationships in tagging data as a heterogeneous graph and propose a novel ranking algorithmic framework for heterogeneous manifolds, named GRoMO (Graph-based Ranking of Multi-type interrelated Objects). In our system both the resource to be tagged (accounting for relevance) and the user[U+05F3]s historical tags (accounting for personalization) are treated as query inputs. Then tags are ranked according to the output of GRoMO and the top tags are recommended to that user. We also explore adapting GRoMO for resource recommendation. For experiments we crawled a tagging dataset from the well-known tagging service, Del.icio.us. Experimental results indicate (1) the proposed method is effective and significantly outperforms baseline methods; (2) the iterative form solutions of GRoMO converge very fast and can be used when the dataset is large; (3) GRoMO can also be used for resource recommendation. © 2014 Elsevier B.V.},
  affiliation     = {College of Information and Technology, Northwest University of China, Xi'an, Shaanxi, China; National Laboratory of Radar Signal Processing, Xidian University, Xi'an, Shaanxi, China},
  author_keywords = {Heterogeneous graphs; Manifold ranking; Recommendation; Social tagging},
  document_type   = {Article},
  doi             = {10.1016/j.neucom.2014.07.011},
  keywords        = {Manifold ranking, Heterogeneous graphs, Social tagging, Recommendation},
  source          = {Scopus},
  url             = {http://www.sciencedirect.com/science/article/pii/S0925231214008893},
}

@Article{wang_etal_2015,
  author          = {Jun Wang and Jiaxu Peng and Ou Liu},
  title           = {A classification approach for less popular webpages based on latent semantic analysis and rough set model},
  journal         = {Expert Systems with Applications},
  year            = {2015},
  volume          = {42},
  number          = {1},
  pages           = {642 - 648},
  issn            = {0957-4174},
  note            = {cited By},
  __markedentry   = {[felipe:1]},
  abstract        = {Nowadays, with the explosive growth of web information, the webpage classification faces great challenge. Computers have difficulty in understanding the semantic meaning of textual or non-textual webpages. Fortunately, Web 2.0 based collaborative tagging system brings new opportunities to solve this problem. It abstracts structured tags from unstructured content in webpages. However, large numbers of webpages on the Internet are less popular. Their tagging information is sparse, which makes their topic unclear and leads to ambiguous classification. Illuminated by the "ambiguous classification", we name the less popular webpage "hesitant webpage". In this paper, we propose an advanced approach for hesitant webpages classification. Firstly, hesitant webpages are divided into bridges, hubs and attached webpages according to their roles on the Internet. Secondly, attached webpages are classified by mining and extending their information in two perspectives. One is the latent semantic analysis (LSA) which is applied to fully explore the semantic meaning of sparse tags. It promotes accurate cognition of webpages semantically close to attached webpages. Another is the proposed density-relation-based rough set model which measures the affiliation degree of attached webpages in different categories. Experiment on real data shows that our approach effectively classifies the hesitant webpages base on the semantic meaning. © 2014 Elsevier Ltd. All rights reserved.},
  affiliation     = {School of Economics and Management, BeiHang University, Beijing, China; School of Accounting and Finance, Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong},
  author_keywords = {Complex network analysis; Latent semantic analysis; Rough set; Webpage classification},
  document_type   = {Article},
  doi             = {http://dx.doi.org/10.1016/j.eswa.2014.08.013},
  keywords        = {Webpage classification, Complex network analysis, Rough set, Latent semantic analysis},
  source          = {Scopus},
  url             = {http://www.sciencedirect.com/science/article/pii/S0957417414004898},
}

@InProceedings{tao_yao_2016,
  author        = {Ruilin Tao and Tianfang Yao},
  title         = {Tag recommendation based on Paragraph Vector},
  booktitle     = {2016 2nd IEEE International Conference on Computer and Communications (ICCC)},
  year          = {2016},
  pages         = {2786-2789},
  month         = {Oct},
  __markedentry = {[felipe:1]},
  abstract      = {With the rapid development of Q&A communities, the amount of questions present on Internet as “explosive” growth. Using tags to organize these questions is a reasonable and effective way. Most of traditional tag recommendations are based on bag-of-words as features. Despite their popularity, bag-of-words features ignore semantics and lose the ordering of the words. Paragraph Vector overcomes the weakness of bag-of-words model. Empirical results show that Paragraph Vector outperforms bag-of-words for tag recommendation.},
  doi           = {10.1109/CompComm.2016.7925205},
  keywords      = {question answering (information retrieval);recommender systems;vectors;Internet;Q and A communities;bag-of-words model;paragraph vector;tag recommendation;Explosives;Films;multilabel;paragraph vector;tag recommendation, rank2},
}

@Article{gong_etal_2017,
  author        = {Yeyun Gong and Qi Zhang and Xuanjing Huang},
  title         = {Hashtag recommendation for multimodal microblog posts},
  journal       = {Neurocomputing},
  year          = {2017},
  issn          = {0925-2312},
  __markedentry = {[felipe:1]},
  doi           = {http://dx.doi.org/10.1016/j.neucom.2017.06.056},
  keywords      = {Hashtag recommendation, Topical model, Social media},
  url           = {http://www.sciencedirect.com/science/article/pii/S0925231217311840},
}

@InBook{jaeschke_etal_2007,
  pages         = {506--514},
  title         = {Tag Recommendations in Folksonomies},
  publisher     = {Springer Berlin Heidelberg},
  year          = {2007},
  author        = {J{\"a}schke, Robert and Marinho, Leandro and Hotho, Andreas and Schmidt-Thieme, Lars and Stumme, Gerd},
  editor        = {Kok, Joost N. and Koronacki, Jacek and Lopez de Mantaras, Ramon and Matwin, Stan and Mladeni{\v{c}}, Dunja and Skowron, Andrzej},
  address       = {Berlin, Heidelberg},
  isbn          = {978-3-540-74976-9},
  __markedentry = {[felipe:1]},
  abstract      = {Collaborative tagging systems allow users to assign keywords---so called ``tags''---to resources. Tags are used for navigation, finding resources and serendipitous browsing and thus provide an immediate benefit for users. These systems usually include tag recommendation mechanisms easing the process of finding good tags for a resource, but also consolidating the tag vocabulary across users. In practice, however, only very basic recommendation strategies are applied.},
  booktitle     = {Knowledge Discovery in Databases: PKDD 2007: 11th European Conference on Principles and Practice of Knowledge Discovery in Databases, Warsaw, Poland, September 17-21, 2007. Proceedings},
  doi           = {10.1007/978-3-540-74976-9_52},
  keywords      = {rank1},
  url           = {https://doi.org/10.1007/978-3-540-74976-9_52},
}

@InProceedings{katakis_etal_2008,
  author               = {Katakis, Ioannis and Tsoumakas, Grigorios and Vlahavas, Ioannis},
  title                = {Multilabel Text Classification for Automated Tag Suggestion},
  booktitle            = {Proceedings of the ECML/PKDD 2008 Discovery Challenge},
  year                 = {2008},
  __markedentry        = {[felipe:1]},
  abstract             = {The increased popularity of tagging during the last few years can be mainly attributed to its embracing by most of the recently thriving user-centric content publishing and management Web 2.0 applications. However, tagging systems have some limitations that have led researchers to develop methods that assist users in the tagging process, by automatically suggesting an appropriate set of tags. We have tried to model the automated tag suggestion problem as a multilabel text classification task in order to participate in the ECML/PKDD 2008 Discovery Challenge.},
  added-at             = {2009-05-26T15:00:29.000+0200},
  biburl               = {http://www.bibsonomy.org/bibtex/23900b844ce31bd81a537081116861a73/lysander07},
  citeulike-article-id = {3790266},
  interhash            = {b9a5f40935e1d4799be3e2fb7db9747d},
  intrahash            = {3900b844ce31bd81a537081116861a73},
  keywords             = {Semantic_Web folksonomy tagging, rank2},
  location             = {Antwerp, Belgium},
  posted-at            = {2008-12-15 21:05:09},
  timestamp            = {2009-06-02T11:03:17.000+0200},
  url                  = {http://lpis.csd.auth.gr/publications/katakis_ecmlpkdd08_challenge.pdf},
}

@InProceedings{heymann_etal_2008,
  author        = {Heymann, Paul and Ramage, Daniel and Garcia-Molina, Hector},
  title         = {Social Tag Prediction},
  booktitle     = {Proceedings of the 31st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval},
  year          = {2008},
  series        = {SIGIR '08},
  pages         = {531--538},
  address       = {New York, NY, USA},
  publisher     = {ACM},
  __markedentry = {[felipe:1]},
  acmid         = {1390425},
  doi           = {10.1145/1390334.1390425},
  isbn          = {978-1-60558-164-4},
  keywords      = {association rules, collaborative tagging, social bookmarking, text classification, rank2},
  location      = {Singapore, Singapore},
  numpages      = {8},
  url           = {http://doi.acm.org/10.1145/1390334.1390425},
}

@InProceedings{song_etal_2008,
  author        = {Song, Yang and Zhuang, Ziming and Li, Huajing and Zhao, Qiankun and Li, Jia and Lee, Wang-Chien and Giles, C. Lee},
  title         = {Real-time Automatic Tag Recommendation},
  booktitle     = {Proceedings of the 31st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval},
  year          = {2008},
  series        = {SIGIR '08},
  pages         = {515--522},
  address       = {New York, NY, USA},
  publisher     = {ACM},
  __markedentry = {[felipe:1]},
  acmid         = {1390423},
  doi           = {10.1145/1390334.1390423},
  isbn          = {978-1-60558-164-4},
  keywords      = {graph partitioning, mixture model, tagging system, rank2},
  location      = {Singapore, Singapore},
  numpages      = {8},
  url           = {http://doi.acm.org/10.1145/1390334.1390423},
}

@INPROCEEDINGS{moxley_etal_2008, 
author={E. Moxley and T. Mei and X. S. Hua and W. Y. Ma and B. S. Manjunath}, 
booktitle={2008 IEEE International Conference on Multimedia and Expo}, 
title={Automatic video annotation through search and mining}, 
year={2008}, 
volume={}, 
number={}, 
pages={685-688}, 
keywords={data mining;video retrieval;video signal processing;multimodal search;query keywords;speech-recognized transcripts;supervised identification;unsupervised annotation;unsupervised automatic video annotation;video mining;video query;video search;visual content transcripts;Asia;Automatic speech recognition;Data mining;Image databases;Machine learning;Noise figure;Support vector machines;Tagging;Video sharing;Vocabulary;data mining;video annotation;video search}, 
doi={10.1109/ICME.2008.4607527}, 
ISSN={1945-7871}, 
month={June},}

@Article{bertin-mahieux_etal_2008,
  author        = {Thierry Bertin-Mahieux and Douglas Eck and François Maillet and Paul Lamere},
  title         = {Autotagger: A Model for Predicting Social Tags from Acoustic Features on Large Music Databases},
  journal       = {Journal of New Music Research},
  year          = {2008},
  volume        = {37},
  number        = {2},
  pages         = {115-135},
  __markedentry = {[felipe:1]},
  abstract      = { Abstract Social tags are user-generated keywords associated with some resource on the Web. In the case of music, social tags have become an important component of “Web 2.0” recommender systems, allowing users to generate playlists based on use-dependent terms such as chill or jogging that have been applied to particular songs. In this paper, we propose a method for predicting these social tags directly from MP3 files. Using a set of 360 classifiers trained using the online ensemble learning algorithm FilterBoost, we map audio features onto social tags collected from the Web. The resulting automatic tags (or autotags) furnish information about music that is otherwise untagged or poorly tagged, allowing for insertion of previously unheard music into a social recommender. This avoids the “cold-start problem” common in such systems. Autotags can also be used to smooth the tag space from which similarities and recommendations are made by providing a set of comparable baseline tags for all tracks in a recommender system. Because the words we learn are the same as those used by people who label their music collections, it is easy to integrate our predictions into existing similarity and prediction methods based on web data. },
  doi           = {10.1080/09298210802479250},
  eprint        = {http://dx.doi.org/10.1080/09298210802479250},
  keywords      = {rank1},
  url           = {http://www.iro.umontreal.ca/~eckdoug/papers/2008_jnmr.pdf
    
},
}

@Article{mrosek_etal_2009,
  author        = {Johannes Mrosek and Stefan Bussmann and Hendrik Albers and Kai Posdziech and Benedikt Hengefeld},
  title         = {Content- and Graph-based Tag Recommendation: Two Variations},
  year          = {2009},
  __markedentry = {[felipe:1]},
  url           = {https://www.kde.cs.uni-kassel.de/ws/dc09/papers/paper_18.pdf},
}

@InProceedings{rendle_schmidt-thieme_2009,
  author        = {Rendle, Steffen and Schmidt-Thieme, Lars},
  title         = {Factor Models for Tag Recommendation in Bibsonomy},
  booktitle     = {Proceedings of the 2009th International Conference on ECML PKDD Discovery Challenge - Volume 497},
  year          = {2009},
  series        = {ECMLPKDDDC'09},
  pages         = {235--242},
  address       = {Aachen, Germany, Germany},
  publisher     = {CEUR-WS.org},
  __markedentry = {[felipe:1]},
  acmid         = {3056166},
  keywords      = {rank1},
  location      = {Bled, Slovenia},
  numpages      = {8},
  url           = {https://www.kde.cs.uni-kassel.de/ws/dc09/papers/paper_13.pdf},
}

@InProceedings{rendle_etal_2009,
  author        = {Rendle, Steffen and Balby Marinho, Leandro and Nanopoulos, Alexandros and Schmidt-Thieme, Lars},
  title         = {Learning Optimal Ranking with Tensor Factorization for Tag Recommendation},
  booktitle     = {Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining},
  year          = {2009},
  series        = {KDD '09},
  pages         = {727--736},
  address       = {New York, NY, USA},
  publisher     = {ACM},
  __markedentry = {[felipe:1]},
  acmid         = {1557100},
  doi           = {10.1145/1557019.1557100},
  isbn          = {978-1-60558-495-9},
  keywords      = {ranking, tag recommendation, tensor factorization, rank2},
  location      = {Paris, France},
  numpages      = {10},
  url           = {http://doi.acm.org/10.1145/1557019.1557100},
}

@Article{song_etal_2011,
  author        = {Song, Yang and Zhang, Lu and Giles, C. Lee},
  title         = {Automatic Tag Recommendation Algorithms for Social Recommender Systems},
  journal       = {ACM Trans. Web},
  year          = {2011},
  volume        = {5},
  number        = {1},
  pages         = {4:1--4:31},
  month         = feb,
  issn          = {1559-1131},
  __markedentry = {[felipe:1]},
  abstract      = {The emergence of Web 2.0 and the consequent success of social network Web sites such as Del.icio.us and Flickr introduce us to a new concept called social bookmarking, or tagging. Tagging is the action of connecting a relevant user-defined keyword to a document, image, or video, which helps the user to better organize and share their collections of interesting stuff. With the rapid growth of Web 2.0, tagged data is becoming more and more abundant on the social network Web sites. An interesting problem is how to automate the process of making tag recommendations to users when a new resource becomes available.

In this article, we address the issue of tag recommendation from a machine learning perspective. From our empirical observation of two large-scale datasets, we first argue that the user-centered approach for tag recommendation is not very effective in practice. Consequently, we propose two novel document-centered approaches that are capable of making effective and efficient tag recommendations in real scenarios. The first, graph-based, method represents the tagged data in two bipartite graphs, (document, tag) and (document, word), then finds document topics by leveraging graph partitioning algorithms. The second, prototype-based, method aims at finding the most representative documents within the data collections and advocates a sparse multiclass Gaussian process classifier for efficient document classification. For both methods, tags are ranked within each topic cluster/class by a novel ranking method. Recommendations are performed by first classifying a new document into one or more topic clusters/classes, and then selecting the most relevant tags from those clusters/classes as machine-recommended tags.

Experiments on real-world data from Del.icio.us, CiteULike, and BibSonomy examine the quality of tag recommendation as well as the efficiency of our recommendation algorithms. The results suggest that our document-centered models can substantially improve the performance of tag recommendations when compared to the user-centered methods, as well as topic models LDA and SVM classifiers.
},
  acmid         = {1921595},
  address       = {New York, NY, USA},
  articleno     = {4},
  doi           = {10.1145/1921591.1921595},
  issue_date    = {February 2011},
  keywords      = {Gaussian processes, Tagging system, graph partitioning, mixture model, multi-label classification, prototype selection, rank2},
  numpages      = {31},
  publisher     = {ACM},
  url           = {http://doi.acm.org/10.1145/1921591.1921595},
}

@InBook{leginus_etal_2012,
  pages         = {151--163},
  title         = {Improving Tensor Based Recommenders with Clustering},
  publisher     = {Springer Berlin Heidelberg},
  year          = {2012},
  author        = {Leginus, Martin and Dolog, Peter and {\v{Z}}emaitis, Valdas},
  editor        = {Masthoff, Judith and Mobasher, Bamshad and Desmarais, Michel C. and Nkambou, Roger},
  address       = {Berlin, Heidelberg},
  isbn          = {978-3-642-31454-4},
  __markedentry = {[felipe:1]},
  abstract      = {Social tagging systems (STS) model three types of entities (i.e. tag-user-item) and relationships between them are encoded into a 3-order tensor. Latent relationships and patterns can be discovered by applying tensor factorization techniques like Higher Order Singular Value Decomposition (HOSVD), Canonical Decomposition etc. STS accumulate large amount of sparse data that restricts factorization techniques to detect latent relations and also significantly slows down the process of a factorization. We propose to reduce tag space by exploiting clustering techniques so that the quality of the recommendations and execution time are improved and memory requirements are decreased. The clustering is motivated by the fact that many tags in a tag space are semantically similar thus the tags can be grouped. Finally, promising experimental results are presented.},
  booktitle     = {User Modeling, Adaptation, and Personalization: 20th International Conference, UMAP 2012, Montreal, Canada, July 16-20, 2012. Proceedings},
  doi           = {10.1007/978-3-642-31454-4_13},
  url           = {https://doi.org/10.1007/978-3-642-31454-4_13},
}

@Article{hu_etal_2012_1,
  author        = {Hu, Jun and Wang, Bing and Liu, Yu and Li, De-Yi},
  title         = {Personalized Tag Recommendation Using Social Influence},
  journal       = {Journal of Computer Science and Technology},
  year          = {2012},
  volume        = {27},
  number        = {3},
  pages         = {527--540},
  month         = {Jan},
  issn          = {1860-4749},
  __markedentry = {[felipe:1]},
  abstract      = {Tag recommendation encourages users to add more tags in bridging the semantic gap between human concept and the features of media object, which provides a feasible solution for content-based multimedia information retrieval. In this paper, we study personalized tag recommendation in a popular online photo sharing site --- Flickr. Social relationship information of users is collected to generate an online social network. From the perspective of network topology, we propose node topological potential to characterize user's social influence. With this metric, we distinguish different social relations between users and find out those who really have influence on the target users. Tag recommendations are based on tagging history and the latent personalized preference learned from those who have most influence in user's social network. We evaluate our method on large scale real-world data. The experimental results demonstrate that our method can outperform the non-personalized global co-occurrence method and other two state-of-the-art personalized approaches using social networks. We also analyze the further usage of our approach for the cold-start problem of tag recommendation.},
  day           = {01},
  doi           = {10.1007/s11390-012-1241-0},
  url           = {https://doi.org/10.1007/s11390-012-1241-0},
}

@InProceedings{wu_etal_2016,
  author        = {Wu, Yong and Yao, Yuan and Xu, Feng and Tong, Hanghang and Lu, Jian},
  title         = {Tag2Word: Using Tags to Generate Words for Content Based Tag Recommendation},
  booktitle     = {Proceedings of the 25th ACM International on Conference on Information and Knowledge Management},
  year          = {2016},
  series        = {CIKM '16},
  pages         = {2287--2292},
  address       = {New York, NY, USA},
  publisher     = {ACM},
  __markedentry = {[felipe:1]},
  abstract      = {Tag recommendation is helpful for the categorization and searching of online content. Existing tag recommendation methods can be divided into collaborative filtering methods and content based methods. In this paper, we put our focus on the content based tag recommendation due to its wider applicability. Our key observation is the tag-content co-occurrence, i.e., many tags have appeared multiple times in the corresponding content. Based on this observation, we propose a generative model (Tag2Word), where we generate the words based on the tag-word distribution as well as the tag itself. Experimental evaluations on real data sets demonstrate that the proposed method outperforms several existing methods in terms of recommendation accuracy, while enjoying linear scalability.},
  acmid         = {2983682},
  doi           = {10.1145/2983323.2983682},
  isbn          = {978-1-4503-4073-1},
  keywords      = {generative model, tag recommendation, tag-content co-occurrence, rank1},
  location      = {Indianapolis, Indiana, USA},
  numpages      = {6},
  url           = {http://doi.acm.org/10.1145/2983323.2983682},
}

@Article{si_sun_2008,
  author        = {Si, Xiance and Sun, Maosong},
  title         = {Tag-LDA for Scalable Real-time Tag Recommendation},
  year          = {2008},
  volume        = {6},
  month         = {11},
  __markedentry = {[felipe:1]},
  booktitle     = {Journal of Computational Information Systems},
  url           = {https://www.yumpu.com/en/document/view/40719286/tag-lda-for-scalable-real-time-tag-recommendation},
}

@InProceedings{si_etal_2009,
  author        = {Si, Xiance and Liu, Zhiyuan and Li, Peng and Jiang, Qixia and Sun, Maosong},
  title         = {Content-based and Graph-based Tag Suggestion},
  booktitle     = {Proceedings of the 2009th International Conference on ECML PKDD Discovery Challenge - Volume 497},
  year          = {2009},
  series        = {ECMLPKDDDC'09},
  pages         = {243--260},
  address       = {Aachen, Germany, Germany},
  publisher     = {CEUR-WS.org},
  __markedentry = {[felipe:1]},
  acmid         = {3056167},
  keywords      = {rank2},
  location      = {Bled, Slovenia},
  numpages      = {18},
  url           = {https://www.kde.cs.uni-kassel.de/ws/dc09/results/papers/paper_14.pdf},
}

@Article{kataria_2016,
  author        = {Saurabh Kataria},
  title         = {Recursive Neural Language Architecture for Tag Prediction},
  journal       = {CoRR},
  year          = {2016},
  volume        = {abs/1603.07646},
  __markedentry = {[felipe:1]},
  bibsource     = {dblp computer science bibliography, http://dblp.org},
  biburl        = {http://dblp.uni-trier.de/rec/bib/journals/corr/Kataria16},
  keywords      = {rank3},
  timestamp     = {Wed, 07 Jun 2017 14:43:01 +0200},
  url           = {http://arxiv.org/abs/1603.07646},
}

@PhdThesis{choubey_2011,
  author        = {Choubey, Rahul},
  title         = {Tag recommendation using Latent Dirichlet Allocation.},
  school        = {Kansas State University},
  year          = {2011},
  __markedentry = {[felipe:1]},
  added-at      = {2014-07-25T15:53:46.000+0200},
  biburl        = {https://www.bibsonomy.org/bibtex/28dfcc809b6e8ca5cf914b134e1d3592b/schwemmlein},
  interhash     = {6522e8899df29e11c1f8062df9c4f20b},
  intrahash     = {8dfcc809b6e8ca5cf914b134e1d3592b},
  keywords      = {allocation dirichlet latent lda recommendation tag, rank3},
  timestamp     = {2014-07-25T15:53:46.000+0200},
  url           = {http://krex.k-state.edu/dspace/handle/2097/9785},
}

@Conference{halpin_etal_2006,
  author        = {Halpin, H. and Robu, V. and Shepherd, H.},
  title         = {The dynamics and semantics of collaborative tagging},
  year          = {2006},
  volume        = {209},
  note          = {cited By},
  abstract      = {The debate within the Web community over the optimal means by which to organize information often pits formalized classifications against distributed collaborative tagging systems. A number of questions remain unanswered, however, regarding the nature of collaborative tagging systems including the dynamics of such systems and whether coherent classification schemes can emerge from undirected tagging by users. Currently millions of users are using collaborative tagging without centrally organizing principles, and many suspect this exhibits features considered to be indicative of a complex system. If this is the case, it remains to be seem whether collaborative tagging by users over time leads to emergent classification schemes that could be formalized into an ontology usable by the Semantic Web. This paper uses data from "popular" tagged sites on the social bookmarking site del.icio.us to examine the dynamics of such collaborative tagging systems. In particular, we are trying to determine whether the distribution of tag frequencies stabilizes, which indicates a degree of cohesion or consensus among users about the optimal tags to describe particular sites. We use tag co-occurrence networks for a sample domain of tags to analyze the meaning of particular tags given their relationship to other tags and automatically create an ontology. We also produce a generative model of collaborative tagging in order to model and understand some of the basic dynamics behind the process.},
  affiliation   = {University of Edinburgh, 2 Buccleuch Place, Edinburgh, United Kingdom; Dutch Center for Mathematics and Computer Science, Kruislaan 413, Amsterdam, Netherlands; Princeton University, Wallace Hall, Princeton, NJ, United States},
  document_type = {Conference Paper},
  journal       = {CEUR Workshop Proceedings},
  keywords      = {rank2},
  page_count    = {10},
  source        = {Scopus},
  url           = {http://ceur-ws.org/Vol-209/saaw06-full01-halpin.pdf},
}

@Article{floeck_etal_2010,
  author          = {Floeck, F. and Putzke, J. and Steinfels, S. and Fischbach, K. and Schoder, D.},
  title           = {Imitation and quality of tags in social bookmarking systems - Collective intelligence leading to folksonomies},
  journal         = {Advances in Intelligent and Soft Computing},
  year            = {2010},
  volume          = {76},
  pages           = {75-91},
  note            = {cited By},
  abstract        = {Social bookmarking platforms often allow users to see a list of tags that have been used previously for the webpage they are currently bookmarking, and from which they can select. In this paper, the authors analyze the influences of this feature on the tag categorizations resulting from the collaborative tagging effort. The main research goal is to show how the interface design of social bookmarking systems can influence the quality of the collective output of their users. Findings from a joint research project with the largest Russian social bookmarking site BobrDobr.ru suggest that if social bookmarking systems allow users to view the most popular tags, the overall variation of keywords used that are assigned to websites by all users decreases. © 2010 Springer-Verlag Berlin Heidelberg.},
  affiliation     = {University of Cologne, Department of Information Systems and Information Management, Pohligstr. 1, 50969 Cologne (Köln), Germany},
  author_keywords = {Collaborative Tagging; Collective Intelligence; Folksonomies; Shared Knowledge; Social-Bookmarking-Systems},
  document_type   = {Conference Paper},
  doi             = {10.1007/978-3-642-14481-3_7},
  source          = {Scopus},
  url             = {https://link.springer.com/chapter/10.1007%2F978-3-642-14481-3_7},
}

@InProceedings{marlow_etal_2006,
  author    = {Marlow, Cameron and Naaman, Mor and Boyd, Danah and Davis, Marc},
  title     = {HT06, Tagging Paper, Taxonomy, Flickr, Academic Article, to Read},
  booktitle = {Proceedings of the Seventeenth Conference on Hypertext and Hypermedia},
  year      = {2006},
  series    = {HYPERTEXT '06},
  pages     = {31--40},
  address   = {New York, NY, USA},
  publisher = {ACM},
  acmid     = {1149949},
  doi       = {10.1145/1149941.1149949},
  isbn      = {1-59593-417-0},
  keywords  = {Flickr, categorization, classification, folksonomy, incentives, models, research, social networks, social software, tagging systems, tagsonomy, taxonomy},
  location  = {Odense, Denmark},
  numpages  = {10},
  url       = {http://doi.acm.org/10.1145/1149941.1149949},
}

@Article{golder_huberman_2006,
  author          = {Golder, S.A. and Huberman, B.A.},
  title           = {Usage patterns of collaborative tagging systems},
  journal         = {Journal of Information Science},
  year            = {2006},
  volume          = {32},
  number          = {2},
  pages           = {198-208},
  note            = {cited By},
  abstract        = {Collaborative tagging describes the process by which many users add metadata in the form of keywords to shared content. Recently, collaborative tagging has grown in popularity on the web, on sites that allow users to tag bookmarks, photographs and other content. In this paper we analyze the structure of collaborative tagging systems as well as their dynamic aspects. Specifically, we discovered regularities in user activity, tag frequencies, kinds of tags used, bursts of popularity in bookmarking and a remarkable stability in the relative proportions of tags within a given URL. We also present a dynamic model of collaborative tagging that predicts these stable patterns and relates them to imitation and shared knowledge. © CILIP.},
  affiliation     = {Information Dynamics Lab., HP Labs., Palo Alto, CA, United States; HP Labs., MS 1139, 1501 Page Mill Rd, Palo Alto, CA 94304, United States},
  author_keywords = {Bookmarks; Collaborative tagging; Del.icio.us; Folksonomy; Sharing; Web},
  document_type   = {Article},
  doi             = {10.1177/0165551506062337},
  keywords        = {rank3},
  source          = {Scopus},
  url             = {http://journals.sagepub.com/doi/pdf/10.1177/0165551506062337},
}

@Misc{wal_2005_broad_and_narrow,
  author = {Wal, V. d.},
  title  = {Explaining and showing broad and narrow folksonomies},
  year   = {2005},
  url    = {http://www.vanderwal.net/random/entrysel.php?blog=1635},
}

@Misc{wal_2005_folksonomy,
  author = {Wal, V. d.},
  title  = {Folksonomy Explanations},
  year={2005},
  url = {http://www.vanderwal.net/random/entrysel.php?blog=1622}
}

@Article{mathes_2004,
  author        = {Adam Mathes},
  title         = {Folksonomies - cooperative classification and communication through shared metadata},
  journal       = {Computer Mediated Communication},
  year          = {2004},
  month         = {December},
  __markedentry = {[felipe:1]},
  abstract      = {This paper examines user-‍generated metadata as implemented and applied in two web services designed to share and organize digital media to better understand grassroots classification. Metadata - data about data - allows systems to collocate related information, and helps users find relevant information. The creation of metadata has generally been approached in two ways: professional creation and author creation. In libraries and other organizations, creating metadata, primarily in the form of catalog records, has traditionally been the domain of dedicated professionals working with complex, detailed rule sets and vocabularies. The primary problem with this approach is scalability and its impracticality for the vast amounts of content being produced and used, especially on the World Wide Web. The apparatus and tools built around professional cataloging systems are generally too complicated for anyone without specialized training and knowledge. A second approach is for metadata to be created by authors. The movement towards creator described documents was heralded by SGML, the WWW, and the Dublin Core Metadata Initiative. There are problems with this approach as well - often due to inadequate or inaccurate description, or outright deception. This paper examines a third approach: user-‍created metadata, where users of the documents and media create metadata for their own individual use that is also shared throughout a community.},
  url           = {http://adammathes.com/academic/computer-mediated-communication/folksonomies.html},
}

@Article{golder_huberman_2005,
  author    = {Scott A. Golder and Bernardo A. Huberman},
  title     = {The Structure of Collaborative Tagging Systems},
  journal   = {CoRR},
  year      = {2005},
  volume    = {abs/cs/0508082},
  bibsource = {dblp computer science bibliography, http://dblp.org},
  biburl    = {http://dblp.uni-trier.de/rec/bib/journals/corr/abs-cs-0508082},
  keywords  = {rank3},
  timestamp = {Wed, 07 Jun 2017 14:40:39 +0200},
  url       = {http://arxiv.org/abs/cs/0508082},
}

@Article{mika_2007,
  author   = {Peter Mika},
  title    = {Ontologies are us: A unified model of social networks and semantics},
  journal  = {Web Semantics: Science, Services and Agents on the World Wide Web},
  year     = {2007},
  volume   = {5},
  number   = {1},
  pages    = {5 - 15},
  issn     = {1570-8268},
  note     = {Selected Papers from the International Semantic Web Conference},
  doi      = {http://dx.doi.org/10.1016/j.websem.2006.11.002},
  keywords = {Knowledge representation, Folksonomies, Tagging},
  url      = {http://www.sciencedirect.com/science/article/pii/S1570826806000552},
}

@InProceedings{dattolo_etal_2010,
  author        = {A. Dattolo and F. Ferrara and C. Tasso},
  title         = {The role of tags for recommendation: A survey},
  booktitle     = {3rd International Conference on Human System Interaction},
  year          = {2010},
  pages         = {548-555},
  month         = {May},
  __markedentry = {[felipe:1]},
  abstract      = {Social tagging is an innovative and powerful mechanism introduced with Web 2.0: it shifts the task of classifying resources from a reduced set of knowledge engineers to the wide set of Web users. Users of social tagging systems define personal classifications which can be used by other peers for browsing available resources. However, due to the absence of rules for managing the tagging process, and to the lack of predefined schemas or structures for inserting metadata and relationships among tags, current user generated classifications dop not produce sound taxonomies. This is a strong limitation which prevents an effective and informed resource sharing. For this reason researchers are modeling innovative recommender systems capable to better support tagging, browsing, and searching for new resources. This paper is a survey which discusses the role of tags in recommender systems: starting from social tagging systems, we analyze various techniques for suggesting content and we introduce the approaches exploited for proposing tags for classifying resources, considering both personalized and not-personalized recommendation.},
  doi           = {10.1109/HSI.2010.5514515},
  issn          = {2158-2246},
  keywords      = {Internet;meta data;pattern classification;personal computing;recommender systems;user interfaces;Web 2.0;informed resource sharing;metadata;personal classification;recommender system;social tagging;Decision support systems;Helium;Knowledge engineering;Mercury (metals);Power engineering and energy;Recommender systems;Resource management;Tagging;Taxonomy;Tag;personalization;recommender system;social tagging, rank1},
}

@Article{schifanella_etal_2010,
  author    = {Rossano Schifanella and Alain Barrat and Ciro Cattuto and Benjamin Markines and Filippo Menczer},
  title     = {Folks in Folksonomies: Social Link Prediction from Shared Metadata},
  journal   = {CoRR},
  year      = {2010},
  volume    = {abs/1003.2281},
  bibsource = {dblp computer science bibliography, http://dblp.org},
  biburl    = {http://dblp.uni-trier.de/rec/bib/journals/corr/abs-1003-2281},
  timestamp = {Wed, 07 Jun 2017 14:40:04 +0200},
  url       = {http://arxiv.org/abs/1003.2281},
}

@InCollection{bordes_etal_2013,
  author    = {Bordes, Antoine and Usunier, Nicolas and Garcia-Duran, Alberto and Weston, Jason and Yakhnenko, Oksana},
  title     = {Translating Embeddings for Modeling Multi-relational Data},
  booktitle = {Advances in Neural Information Processing Systems 26},
  publisher = {Curran Associates, Inc.},
  year      = {2013},
  editor    = {C. J. C. Burges and L. Bottou and M. Welling and Z. Ghahramani and K. Q. Weinberger},
  pages     = {2787--2795},
  url       = {http://papers.nips.cc/paper/5071-translating-embeddings-for-modeling-multi-relational-data.pdf},
}

@InProceedings{illig_etal_2011,
  author        = {Illig, Jens and Hotho, Andreas and J\"{a}schke, Robert and Stumme, Gerd},
  title         = {A Comparison of Content-based Tag Recommendations in Folksonomy Systems},
  booktitle     = {Proceedings of the First International Conference on Knowledge Processing and Data Analysis},
  year          = {2011},
  series        = {KONT'07/KPP'07},
  pages         = {136--149},
  address       = {Berlin, Heidelberg},
  publisher     = {Springer-Verlag},
  __markedentry = {[felipe:1]},
  acmid         = {2022778},
  isbn          = {978-3-642-22139-2},
  location      = {Novosibirsk, Russia},
  numpages      = {14},
  url           = {http://dl.acm.org/citation.cfm?id=2022767.2022778},
}

@InProceedings{zubiaga_etal_2009,
  author        = {A. Zubiaga and A. P. García-Plaza and V. Fresno and R. Martínez},
  title         = {Content-Based Clustering for Tag Cloud Visualization},
  booktitle     = {2009 International Conference on Advances in Social Network Analysis and Mining},
  year          = {2009},
  pages         = {316-319},
  month         = {July},
  __markedentry = {[felipe:1]},
  abstract      = {Social tagging systems are becoming an interesting way to retrieve web information from previously annotated data. These sites present a tag cloud made up by the most popular tags, where neither tag grouping nor their corresponding content is considered. We present a methodology to obtain and visualize a cloud of related tags based on the use of self-organizing maps, and where the relations among tags are established taking into account the textual content of tagged documents. Each map unit can be represented by the most relevant terms of the tags it contains, so that it is possible to study and analyze the groups as well as to visualize and navigate through the relevant terms and tags.},
  doi           = {10.1109/ASONAM.2009.19},
  keywords      = {Internet;data visualisation;information retrieval;pattern clustering;self-organising feature maps;social networking (online);content-based clustering;self-organizing maps;social tagging;tag cloud visualisation;tagged documents;textual content;web information retrieval;Data visualization;Feeds;Navigation;Ontologies;Self organizing feature maps;Semantic Web;Social network services;Subscriptions;Tag clouds;Tagging;clustering;information access;social-tagging;visualization},
}

@InProceedings{kataria_agarwal_2015:c2,
  author    = {S. Kataria and A. Agarwal},
  title     = {Distributed Representations for Content-Based and Personalized Tag Recommendation},
  booktitle = {2015 IEEE International Conference on Data Mining Workshop (ICDMW)},
  year      = {2015},
  pages     = {1388-1395},
  month     = {Nov},
  abstract  = {Summary form only given. Strong light-matter coupling has been recently successfully explored in the GHz and THz [1] range with on-chip platforms. New and intriguing quantum optical phenomena have been predicted in the ultrastrong coupling regime [2], when the coupling strength Ω becomes comparable to the unperturbed frequency of the system ω. We recently proposed a new experimental platform where we couple the inter-Landau level transition of an high-mobility 2DEG to the highly subwavelength photonic mode of an LC meta-atom [3] showing very large Ω/ωc = 0.87. Our system benefits from the collective enhancement of the light-matter coupling which comes from the scaling of the coupling Ω ∝ √n, were n is the number of optically active electrons. In our previous experiments [3] and in literature [4] this number varies from 104-103 electrons per meta-atom. We now engineer a new cavity, resonant at 290 GHz, with an extremely reduced effective mode surface Seff = 4 × 10-14 m2 (FE simulations, CST), yielding large field enhancements above 1500 and allowing to enter the few (<;100) electron regime. It consist of a complementary metasurface with two very sharp metallic tips separated by a 60 nm gap (Fig.1(a, b)) on top of a single triangular quantum well. THz-TDS transmission experiments as a function of the applied magnetic field reveal strong anticrossing of the cavity mode with linear cyclotron dispersion. Measurements for arrays of only 12 cavities are reported in Fig.1(c). On the top horizontal axis we report the number of electrons occupying the topmost Landau level as a function of the magnetic field. At the anticrossing field of B=0.73 T we measure approximately 60 electrons ultra strongly coupled (Ω/ω- ||},
  doi       = {10.1109/ICDMW.2015.240},
  keywords  = {indexing;learning (artificial intelligence);natural language processing;neural nets;recommender systems;search problems;book-marking datasets;content-based recommendation;corpus wide topic-level information;distributed representation learning problem;document level word sequence information;jointly trained vector representations;low-dimensional vectors;multilayered neural language model;personalized tag recommendation tasks;two stage method;Conferences;Context;Context modeling;Optimization;Semantics;Tagging;Training},
}

@article{peralta_2007,
author = {Peralta, Veronika},
year = {2007},
month = {08},
pages = {},
title = {Extraction and Integration of MovieLens and IMDb Data}
}

@article{obar_wildman_2015,
title = "Social media definition and the governance challenge: An introduction to the special issue",
journal = "Telecommunications Policy",
volume = "39",
number = "9",
pages = "745 - 750",
year = "2015",
note = "SPECIAL ISSUE ON THE GOVERNANCE OF SOCIAL MEDIA",
issn = "0308-5961",
doi = "https://doi.org/10.1016/j.telpol.2015.07.014",
url = "http://www.sciencedirect.com/science/article/pii/S0308596115001172",
author = "Jonathan A. Obar and Steve Wildman"
}

@inbook{hamburger_etal_2017,
title = {Social Networking},
author = {Amichai-Hamburger, Yair and Hayat, Tsahi},
author = {},
publisher = {John Wiley \& Sons, Inc.},
isbn = {9781118783764},
url = {http://dx.doi.org/10.1002/9781118783764.wbieme0170},
doi = {10.1002/9781118783764.wbieme0170},
keywords = {computer-mediated communication, digital culture, social networks},
booktitle = {The International Encyclopedia of Media Effects},
year = {2017},
}

@article{aiello_2012,
 author = {Aiello, Luca Maria and Barrat, Alain and Schifanella, Rossano and Cattuto, Ciro and Markines, Benjamin and Menczer, Filippo},
 title = {Friendship Prediction and Homophily in Social Media},
 journal = {ACM Trans. Web},
 issue_date = {May 2012},
 volume = {6},
 number = {2},
 month = jun,
 year = {2012},
 issn = {1559-1131},
 pages = {9:1--9:33},
 articleno = {9},
 numpages = {33},
 url = {http://doi.acm.org/10.1145/2180861.2180866},
 doi = {10.1145/2180861.2180866},
 acmid = {2180866},
 publisher = {ACM},
 address = {New York, NY, USA},
 keywords = {Social media, collaborative tagging, folksonomies, homophily, link prediction, maximum Information path, social network, topical similarity},
}

@inproceedings{helic_etal_2012,
 author = {Helic, Denis and K\"{o}rner, Christian and Granitzer, Michael and Strohmaier, Markus and Trattner, Christoph},
 title = {Navigational Efficiency of Broad vs. Narrow Folksonomies},
 booktitle = {Proceedings of the 23rd ACM Conference on Hypertext and Social Media},
 series = {HT '12},
 year = {2012},
 isbn = {978-1-4503-1335-3},
 location = {Milwaukee, Wisconsin, USA},
 pages = {63--72},
 numpages = {10},
 url = {http://doi.acm.org/10.1145/2309996.2310008},
 doi = {10.1145/2309996.2310008},
 acmid = {2310008},
 publisher = {ACM},
 address = {New York, NY, USA},
 keywords = {folksonomy, keywords, navigation, tags},
}

@book{peters_2009,
 author = {Peters, Isabella},
 title = {Folksonomies. Indexing and Retrieval in Web 2.0},
 year = {2009},
 isbn = {3598251793, 9783598251795},
 edition = {1st},
 publisher = {Walter de Gruyter \& Co.},
 address = {Hawthorne, NJ, USA},
}

@InCollection{baumeister_leary_1997,
  author    = {Baumeister, Roy F. and Leary, Mark R.},
  title     = {Writing narrative literature reviews},
  booktitle = {Review of General Psychology},
  year      = {1997},
  url       = {http://psycnet.apa.org/doiLanding?doi=10.1037\%2F1089-2680.1.3.311},
}

@ARTICLE{bem_1995,
    author = {Daryl J. Bem},
    title = {Writing a Review Article for Psychological Bulletin},
    journal = {Psychological Bulletin},
    year = {1995},
    pages = {172--177}
}

@article{blei_etal_2003,
 author = {Blei, David M. and Ng, Andrew Y. and Jordan, Michael I.},
 title = {Latent Dirichlet Allocation},
 journal = {J. Mach. Learn. Res.},
 issue_date = {3/1/2003},
 volume = {3},
 month = mar,
 year = {2003},
 issn = {1532-4435},
 pages = {993--1022},
 numpages = {30},
 url = {http://dl.acm.org/citation.cfm?id=944919.944937},
 acmid = {944937},
 publisher = {JMLR.org},
}

@article{kullback_leibler_1951,
author = "Kullback, S. and Leibler, R. A.",
doi = "10.1214/aoms/1177729694",
fjournal = "The Annals of Mathematical Statistics",
journal = "Ann. Math. Statist.",
month = "03",
number = "1",
pages = "79--86",
publisher = "The Institute of Mathematical Statistics",
title = "On Information and Sufficiency",
url = "https://doi.org/10.1214/aoms/1177729694",
volume = "22",
year = "1951"
}

@article{dietterich_etal_1997,
 author = {Dietterich, Thomas G. and Lathrop, Richard H. and Lozano-P{\'e}rez, Tom\'{a}s},
 title = {Solving the Multiple Instance Problem with Axis-parallel Rectangles},
 journal = {Artif. Intell.},
 issue_date = {Jan. 1997},
 volume = {89},
 number = {1-2},
 month = jan,
 year = {1997},
 issn = {0004-3702},
 pages = {31--71},
 numpages = {41},
 url = {http://dx.doi.org/10.1016/S0004-3702(96)00034-3},
 doi = {10.1016/S0004-3702(96)00034-3},
 acmid = {249682},
 publisher = {Elsevier Science Publishers Ltd.},
 address = {Essex, UK},
 keywords = {drug design, machine learning, structure-activity relationships},
}

@inproceedings{hearst_1994,
 author = {Hearst, Marti A.},
 title = {Multi-paragraph Segmentation of Expository Text},
 booktitle = {Proceedings of the 32Nd Annual Meeting on Association for Computational Linguistics},
 series = {ACL '94},
 year = {1994},
 location = {Las Cruces, New Mexico},
 pages = {9--16},
 numpages = {8},
 url = {https://doi.org/10.3115/981732.981734},
 doi = {10.3115/981732.981734},
 acmid = {981734},
 publisher = {Association for Computational Linguistics},
 address = {Stroudsburg, PA, USA},
}

@Book{edgar_2008,
author={Edgar,Gerald},
title = {Measure, Topology, and Fractal Geometry},
booktitle = {Measure, Topology, and Fractal Geometry},
year={2008},
publisher={Springer-Verlag New York}

}

@article{sokolova_and_napalme_2009,
title = "A systematic analysis of performance measures for classification tasks",
journal = "Information Processing \& Management",
volume = "45",
number = "4",
pages = "427 - 437",
year = "2009",
issn = "0306-4573",
doi = "https://doi.org/10.1016/j.ipm.2009.03.002",
url = "http://www.sciencedirect.com/science/article/pii/S0306457309000259",
author = "Marina Sokolova and Guy Lapalme",
keywords = "Performance evaluation, Machine Learning, Text classification"
}

@inproceedings{le_mikolov_2014,
 author = {Le, Quoc and Mikolov, Tomas},
 title = {Distributed Representations of Sentences and Documents},
 booktitle = {Proceedings of the 31st International Conference on International Conference on Machine Learning - Volume 32},
 series = {ICML'14},
 year = {2014},
 location = {Beijing, China},
 pages = {II-1188--II-1196},
 url = {http://dl.acm.org/citation.cfm?id=3044805.3045025},
 acmid = {3045025},
 publisher = {JMLR.org},
}

@article{correa_etal_2017,
  author    = {Edilson Anselmo Corr{\^{e}}a J{\'{u}}nior and
               Vanessa Queiroz Marinho and
               Leandro Borges dos Santos},
  title     = {{NILC-USP} at SemEval-2017 Task 4: {A} Multi-view Ensemble for Twitter
               Sentiment Analysis},
  journal   = {CoRR},
  volume    = {abs/1704.02263},
  year      = {2017},
  url       = {http://arxiv.org/abs/1704.02263},
  archivePrefix = {arXiv},
  eprint    = {1704.02263},
  timestamp = {Wed, 25 Oct 2017 16:39:35 +0200},
  biburl    = {https://dblp.org/rec/bib/journals/corr/CorreaMS17},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}

@ARTICLE{tsoumakas_katakis_2007,
    author = {Grigorios Tsoumakas and Ioannis Katakis},
    title = {Multi-label classification: An overview},
    journal = {Int J Data Warehousing and Mining},
    year = {2007},
    volume = {2007},
    pages = {1--13}
}

@article{rifkin_klautau_2004,
 author = {Rifkin, Ryan and Klautau, Aldebaro},
 title = {In Defense of One-Vs-All Classification},
 journal = {J. Mach. Learn. Res.},
 issue_date = {12/1/2004},
 volume = {5},
 month = dec,
 year = {2004},
 issn = {1532-4435},
 pages = {101--141},
 numpages = {41},
 url = {http://dl.acm.org/citation.cfm?id=1005332.1005336},
 acmid = {1005336},
 publisher = {JMLR.org},
}

@incollection{zhou_zhang_2006,
title = {Multi-Instance Multi-Label Learning with Application to Scene Classification},
author = {Zhang, Min-Ling and Zhou, Zhi-Hua},
booktitle = {Advances in Neural Information Processing Systems 19},
editor = {B. Sch\"{o}lkopf and J. C. Platt and T. Hoffman},
pages = {1609--1616},
year = {2006},
publisher = {MIT Press},
url = {http://papers.nips.cc/paper/3047-multi-instance-multi-label-learning-with-application-to-scene-classification.pdf}
}

@misc{kaufmanl_rousseeuw_1987,
  added-at = {2017-07-03T12:14:53.000+0200},
  author = {Kaufman, Leonard and Rousseeuw, Peter J.},
  biburl = {https://www.bibsonomy.org/bibtex/22b132bef56453004f6326777af4ee791/nosebrain},
  editor = {Y, In: Dodge and editor},
  interhash = {ef930c034e33409e95869d0016cc74a0},
  intrahash = {2b132bef56453004f6326777af4ee791},
  keywords = {clustering kmediods lecture:data-mining pam},
  location = {Amsterdam:},
  pages = {405–416},
  publisher = {North Holland / Elsevier},
  timestamp = {2017-08-03T17:03:26.000+0200},
  title = {Clustering by means of medoids},
  year = 1987
}

@article{huttenlocher_etal_1993,
 author = {Huttenlocher, D. P. and Klanderman, G. A. and Rucklidge, W. A.},
 title = {Comparing Images Using the Hausdorff Distance},
 journal = {IEEE Trans. Pattern Anal. Mach. Intell.},
 issue_date = {September 1993},
 volume = {15},
 number = {9},
 month = sep,
 year = {1993},
 issn = {0162-8828},
 pages = {850--863},
 numpages = {14},
 url = {https://doi.org/10.1109/34.232073},
 doi = {10.1109/34.232073},
 acmid = {628513},
 publisher = {IEEE Computer Society},
 address = {Washington, DC, USA},
 keywords = {Hausdorff distance, binary image, image comparison, image processing, position error tolerance, rigid motion, shape comparison methods, translation},
}

@article{zhang_zhou_2007,
 author = {Zhang, Min-Ling and Zhou, Zhi-Hua},
 title = {ML-KNN: A Lazy Learning Approach to Multi-label Learning},
 journal = {Pattern Recogn.},
 issue_date = {July, 2007},
 volume = {40},
 number = {7},
 month = jul,
 year = {2007},
 issn = {0031-3203},
 pages = {2038--2048},
 numpages = {11},
 url = {http://dx.doi.org/10.1016/j.patcog.2006.12.019},
 doi = {10.1016/j.patcog.2006.12.019},
 acmid = {1234635},
 publisher = {Elsevier Science Inc.},
 address = {New York, NY, USA},
 keywords = {Functional genomics, K-nearest neighbor, KNN, K-nearest neighbor, Lazy learning, MAP, maximum a posteriori, ML-KNN, multi-label K-nearest neighbor, Machine learning, Multi-label learning, Natural scene classification, PMM, parametric mixture model, Text categorization},
}

@inproceedings{mikolov_etal_2013,
  author    = {Tomas Mikolov and
               Ilya Sutskever and
               Kai Chen and
               Gregory S. Corrado and
               Jeffrey Dean},
  title     = {Distributed Representations of Words and Phrases and their Compositionality},
  booktitle = {Advances in Neural Information Processing Systems 26: 27th Annual
               Conference on Neural Information Processing Systems 2013. Proceedings
               of a meeting held December 5-8, 2013, Lake Tahoe, Nevada, United States.},
  pages     = {3111--3119},
  year      = {2013},
  url       = {http://papers.nips.cc/paper/5021-distributed-representations-of-words-and-phrases-and-their-compositionality},
  timestamp = {Thu, 07 May 2015 20:02:01 +0200},
  biburl    = {http://dblp.uni-trier.de/rec/bib/conf/nips/MikolovSCCD13},
  bibsource = {dblp computer science bibliography, http://dblp.org}
}

@INPROCEEDINGS{arthur_vassilvitskii_2007,
    author = {David Arthur and Sergei Vassilvitskii},
    title = {K-means++: the advantages of careful seeding},
    booktitle = {In Proceedings of the 18th Annual ACM-SIAM Symposium on Discrete Algorithms},
    year = {2007}
}

@INPROCEEDINGS{macqueen_1967,
    author = {J. Macqueen},
    title = {Some methods for classification and analysis of multivariate observations},
    booktitle = {In 5-th Berkeley Symposium on Mathematical Statistics and Probability},
    year = {1967},
    pages = {281--297}
}

@book{jain_dubes_1988,
 author = {Jain, Anil K. and Dubes, Richard C.},
 title = {Algorithms for Clustering Data},
 year = {1988},
 isbn = {0-13-022278-X},
 publisher = {Prentice-Hall, Inc.},
 address = {Upper Saddle River, NJ, USA},
}

@article{wieting_etal_2016,
  author    = {John Wieting and
               Mohit Bansal and
               Kevin Gimpel and
               Karen Livescu},
  title     = {Towards Universal Paraphrastic Sentence Embeddings},
  journal   = {CoRR},
  volume    = {abs/1511.08198},
  year      = {2015},
  url       = {http://arxiv.org/abs/1511.08198},
  archivePrefix = {arXiv},
  eprint    = {1511.08198},
  timestamp = {Wed, 07 Jun 2017 14:40:28 +0200},
  biburl    = {https://dblp.org/rec/bib/journals/corr/WietingBGL15a},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}

@article{arora_etal_2017, 
	author = {Sanjeev Arora and Yingyu Liang and Tengyu Ma}, 
	title = {A Simple but Tough-to-Beat Baseline for Sentence Embeddings}, 
	booktitle = {International Conference on Learning Representations},
	year = {2017}
}

@INPROCEEDINGS{maron_ratan_98,
    author = {Oded Maron and Aparna Lakshmi Ratan},
    title = {Multiple-Instance Learning for Natural Scene Classification},
    booktitle = {In The Fifteenth International Conference on Machine Learning},
    year = {1998},
    pages = {341--349},
    publisher = {Morgan Kaufmann}
}

@INPROCEEDINGS{andrews_etal_2003,
    author = {Stuart Andrews and Ioannis Tsochantaridis and Thomas Hofmann},
    title = {Support vector machines for multiple-instance learning},
    booktitle = {Advances in Neural Information Processing Systems 15},
    year = {2003},
    pages = {561--568},
    publisher = {MIT Press}
}

@book{marinho_etal_2012,
 author = {Marinho, Leandro Balby and Hotho, Andreas and Jschke, Robert and Nanopoulos, Alexandros and Rendle, Steffen and Schmidt-Thieme, Lars and Stumme, Gerd and Symeonidis, Panagiotis},
 title = {Recommender Systems for Social Tagging Systems},
 year = {2012},
 isbn = {1461418933, 9781461418931},
 publisher = {Springer Publishing Company, Incorporated},
}

@ARTICLE{deerwester_etal_1990,
    author = {Scott Deerwester and Susan T. Dumais and George W. Furnas and Thomas K. Landauer and Richard Harshman},
    title = {Indexing by latent semantic analysis},
    journal = {JOURNAL OF THE AMERICAN SOCIETY FOR INFORMATION SCIENCE},
    year = {1990},
    volume = {41},
    number = {6},
    pages = {391--407}
}

@ARTICLE{gillis_2014,
   author = {{Gillis}, N.},
    title = "{The Why and How of Nonnegative Matrix Factorization}",
  journal = {ArXiv e-prints},
archivePrefix = "arXiv",
   eprint = {1401.5226},
 primaryClass = "stat.ML",
 keywords = {Statistics - Machine Learning, Computer Science - Information Retrieval, Computer Science - Learning, Mathematics - Optimization and Control},
     year = 2014,
    month = jan,
   adsurl = {http://adsabs.harvard.edu/abs/2014arXiv1401.5226G},
  adsnote = {Provided by the SAO/NASA Astrophysics Data System}
}

@article{huang_etal_2011,
author = {Huang, Jianbin and Sun, Heli and Han, Jiawei and Feng, Boqin},
year = {2011},
month = {06},
pages = {2160-2171},
title = {Density-based shrinkage for revealing hierarchical and overlapping community structure in networks},
volume = {390},
booktitle = {Physica A-statistical Mechanics and Its Applications - PHYSICA A}
}

@incollection{rabiner_1989,
 author = {Rabiner, Lawrence R.},
 chapter = {A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition},
 title = {Readings in Speech Recognition},
 editor = {Waibel, Alex and Lee, Kai-Fu},
 year = {1990},
 isbn = {1-55860-124-4},
 pages = {267--296},
 numpages = {30},
 url = {http://dl.acm.org/citation.cfm?id=108235.108253},
 acmid = {108253},
 publisher = {Morgan Kaufmann Publishers Inc.},
 address = {San Francisco, CA, USA},
}

@Article{rawashdeh_etal_2012,
author="Rawashdeh, Majdi
and Kim, Heung-Nam
and Alja'am, Jihad Mohamad
and El Saddik, Abdulmotaleb",
title="Folksonomy link prediction based on a tripartite graph for tag recommendation",
journal="Journal of Intelligent Information Systems",
year="2013",
month="Apr",
day="01",
volume="40",
number="2",
pages="307--325",
abstract="Nowadays social tagging has become a popular way to annotate, search, navigate and discover online resources, in turn leading to the sheer amount of user-generated metadata. This paper addresses the problem of recommending suitable tags during folksonomy development from a graph-based perspective. The proposed approach adapts the Katz measure, a path-ensemble based proximity measure, for the use in social tagging systems. We model a folksonomy as a weighted, undirected tripartite graph. We then apply the Katz measure to this graph, and exploit it to provide tag recommendations for individual users. We evaluate our method on two real-world folksonomies collected from CiteULike and Last.fm. The experimental results demonstrate that the proposed method improves the recommendation performance and is effective for both active taggers and cold-start taggers compared to existing algorithms.",
issn="1573-7675",
doi="10.1007/s10844-012-0227-2",
url="https://doi.org/10.1007/s10844-012-0227-2"
}

@book{berge_1985,
 author = {Berge, C.},
 title = {Graphs and Hypergraphs},
 year = {1985},
 isbn = {0720404797},
 publisher = {Elsevier Science Ltd.},
 address = {Oxford, UK, UK},
}

@MISC{maaten_etal_2008,
    author = {L.J.P. van der Maaten and E. O. Postma and H. J. van den Herik},
    title = { Dimensionality Reduction: A Comparative Review},
    year = {2008}
}

@article{wold_etal_1987,
title = "Principal component analysis",
journal = "Chemometrics and Intelligent Laboratory Systems",
volume = "2",
number = "1",
pages = "37 - 52",
year = "1987",
note = "Proceedings of the Multivariate Statistical Workshop for Geologists and Geochemists",
issn = "0169-7439",
doi = "https://doi.org/10.1016/0169-7439(87)80084-9",
url = "http://www.sciencedirect.com/science/article/pii/0169743987800849",
author = "Svante Wold and Kim Esbensen and Paul Geladi"
}

@incollection{bengio_etal_2007,
title = {Greedy Layer-Wise Training of Deep Networks},
author = {Bengio, Yoshua and Pascal Lamblin and Dan Popovici and Larochelle, Hugo},
booktitle = {Advances in Neural Information Processing Systems 19},
editor = {B. Sch\"{o}lkopf and J. C. Platt and T. Hoffman},
pages = {153--160},
year = {2007},
publisher = {MIT Press},
url = {http://papers.nips.cc/paper/3048-greedy-layer-wise-training-of-deep-networks.pdf}
}

@article{jain_2010,
title = "Data clustering: 50 years beyond K-means",
journal = "Pattern Recognition Letters",
volume = "31",
number = "8",
pages = "651 - 666",
year = "2010",
note = "Award winning papers from the 19th International Conference on Pattern Recognition (ICPR)",
issn = "0167-8655",
doi = "https://doi.org/10.1016/j.patrec.2009.09.011",
url = "http://www.sciencedirect.com/science/article/pii/S0167865509002323",
author = "Anil K. Jain",
keywords = "Data clustering, User’s dilemma, Historical developments, Perspectives on clustering, King-Sun Fu prize"
}

@article{pan_yang_2010,
 author = {Pan, Sinno Jialin and Yang, Qiang},
 title = {A Survey on Transfer Learning},
 journal = {IEEE Trans. on Knowl. and Data Eng.},
 issue_date = {October 2010},
 volume = {22},
 number = {10},
 month = oct,
 year = {2010},
 issn = {1041-4347},
 pages = {1345--1359},
 numpages = {15},
 url = {http://dx.doi.org/10.1109/TKDE.2009.191},
 doi = {10.1109/TKDE.2009.191},
 acmid = {1850545},
 publisher = {IEEE Educational Activities Department},
 address = {Piscataway, NJ, USA},
 keywords = {Transfer learning, Transfer learning, survey, machine learning, data mining., data mining., machine learning, survey},
}

@incollection{bordes_etal_2013:c2,
title = {Translating Embeddings for Modeling Multi-relational Data},
author = {Bordes, Antoine and Usunier, Nicolas and Garcia-Duran, Alberto and Weston, Jason and Yakhnenko, Oksana},
booktitle = {Advances in Neural Information Processing Systems 26},
editor = {C. J. C. Burges and L. Bottou and M. Welling and Z. Ghahramani and K. Q. Weinberger},
pages = {2787--2795},
year = {2013},
publisher = {Curran Associates, Inc.},
url = {http://papers.nips.cc/paper/5071-translating-embeddings-for-modeling-multi-relational-data.pdf}
}

@article{haralick_shapiro_1985,
title = "Image segmentation techniques",
journal = "Computer Vision, Graphics, and Image Processing",
volume = "29",
number = "1",
pages = "100 - 132",
year = "1985",
issn = "0734-189X",
doi = "https://doi.org/10.1016/S0734-189X(85)90153-7",
url = "http://www.sciencedirect.com/science/article/pii/S0734189X85901537",
author = "Robert M. Haralick and Linda G. Shapiro"
}

@techreport{page_etal_1999,
          number = {1999-66},
           month = {November},
          author = {Lawrence Page and Sergey Brin and Rajeev Motwani and Terry Winograd},
            note = {Previous number = SIDL-WP-1999-0120},
           title = {The PageRank Citation Ranking: Bringing Order to the Web.},
            type = {Technical Report},
       publisher = {Stanford InfoLab},
            year = {1999},
     institution = {Stanford InfoLab},
             url = {http://ilpubs.stanford.edu:8090/422/},
        abstract = {The importance of a Web page is an inherently subjective matter, which depends on the readers interests, knowledge and attitudes. But there is still much that can be said objectively about the relative importance of Web pages. This paper describes PageRank, a mathod for rating Web pages objectively and mechanically, effectively measuring the human interest and attention devoted to them. We compare PageRank to an idealized random Web surfer. We show how to efficiently compute PageRank for large numbers of pages. And, we show how to apply PageRank to search and to user navigation.}
}

@InProceedings{hassan_etal_2009,
  author        = {Hassan, Malik Tahir and Karim, Asim and Manandhar, Suresh and Cussens, James},
  title         = {Discriminative Clustering for Content-based Tag Recommendation in Social Bookmarking Systems},
  booktitle     = {Proceedings of the 2009th International Conference on ECML PKDD Discovery Challenge - Volume 497},
  year          = {2009},
  series        = {ECMLPKDDDC'09},
  pages         = {85--97},
  address       = {Aachen, Germany, Germany},
  publisher     = {CEUR-WS.org},
  __markedentry = {[felipe:1]},
  acmid         = {3056154},
  location      = {Bled, Slovenia},
  numpages      = {13},
  url           = {http://ceur-ws.org/Vol-497/paper_24.pdf},
}

@Conference{sordo_etal_2007,
  author        = {Sordo, M. and Laurier, C. and Celma, {\`O}.},
  title         = {Annotating Music Collections How content-based similarity helps to propagate labels},
  booktitle     = {8th International Conference on Music Information Retrieval},
  year          = {2007},
  address       = {Vienna, Austria},
  __markedentry = {[felipe:1]},
  abstract      = {In this paper we present a way to annotate music collections by exploiting audio similarity. In this sense, similarity is used to propose labels (tags) to yet unlabeled songs, based on the content-based distance between them. The main goal of our work is to ease the process of annotating huge music collections, by using content-based similarity distances as a way to propagate labels among songs. <br />
<br />
We present two different experiments. The first one propagates labels that are related with the style of the piece, whereas the second experiment deals with mood labels. On the one hand, our approach shows that using a music collection annotated at 40\% with styles, and using content-based, the collection can be automatically annotated up to 78\% (that is, 40\% already annotated and the rest, 38\%, only using propagation), with a recall greater than 0.4. On the other hand, for a smaller music collection annotated at 30\% with moods, the collection can be automatically annotated up to 65\% (e.g. 30\% plus 35\% using propagation).
},
  keywords      = {rank1},
  url           = {http://mtg.upf.edu/files/publications/7c086c-ISMIR-2007-msordo-claurier.pdf},
}

@MISC{hsu_etal_2010,
    author = {Chih-wei Hsu and Chih-chung Chang and Chih-jen Lin},
    title = {A practical guide to support vector classification },
    year = {2010}
}

@inproceedings{li_etal_2015,
  title={Data Sparseness in Linear SVM},
  author={Xiang Li and Huaimin Wang and Bin Gu and Charles X. Ling},
  booktitle={IJCAI},
  year={2015}
}

@Inbook{tsoumakas_2010,
author="Tsoumakas, Grigorios
and Katakis, Ioannis
and Vlahavas, Ioannis",
editor="Maimon, Oded
and Rokach, Lior",
title="Mining Multi-label Data",
bookTitle="Data Mining and Knowledge Discovery Handbook",
year="2010",
publisher="Springer US",
address="Boston, MA",
pages="667--685",
abstract="A large body of research in supervised learning deals with the analysis of single-label data, where training examples are associated with a single label $\lambda$ from a set of disjoint labels L. However, training examples in several application domains are often associated with a set of labels Y ⊆ L. Such data are called multi-label.",
isbn="978-0-387-09823-4",
doi="10.1007/978-0-387-09823-4_34",
url="https://doi.org/10.1007/978-0-387-09823-4_34"
}

@article{kishida_2005,
title = "Property of average precision and its generalization: An examination of evaluation indicator for information retrieval experiments",
abstract = "In information retrieval experiments, indicators for measuring effectiveness of the systems or methods are important. Average precision is often used as an indicator for evaluating ranked output of documents in standard retrieval experiments. This report examines some properties of this indicator. First, we clarify mathematically that relevant documents at a higher position in the ranked list contribute much more to increasing the score of average precision. Second, influence of detecting unknown relevant documents on the score is discussed. Third, we examine statistical variation of average precision scores caused by fluctuations of results from relevance judgments. Another issue of this report is to explore evaluation indicators using data of multi-grade relevance. After reviewing indicators proposed by other researchers, i.e., modified sliding ratio, normalized discounted cumulative gain (nDCG), Q-measure and so on, a new indicator, generalized average precision is developed. We compare these indicators empirically using a simple and artificial example.",
author = "Kazuaki Kishida",
year = "2005",
month = "9",
day = "29",
language = "English",
volume = "2005",
pages = "1--19",
journal = "NII Technical Reports",
issn = "1346-5597",
publisher = "National Institute of Informatics",
number = "14",
}

@Article{cattuto_etal_2007,
  author    = {Ciro Cattuto and Andrea Baldassarri and Vito Domenico Pietro Servedio and Vittorio Loreto},
  title     = {Vocabulary growth in collaborative tagging systems},
  journal   = {CoRR},
  year      = {2007},
  volume    = {abs/0704.3316},
  bibsource = {dblp computer science bibliography, http://dblp.org},
  biburl    = {http://dblp.uni-trier.de/rec/bib/journals/corr/abs-0704-3316},
  timestamp = {Wed, 07 Jun 2017 14:40:28 +0200},
  url       = {http://arxiv.org/abs/0704.3316},
}

@inproceedings{koerner_etal_2010,
author = {Körner, Christian and Benz, Dominik and Hotho, Andreas and Strohmaier, Markus and Stumme, Gerd},
year = {2010},
month = {01},
pages = {521-530},
title = {Stop thinking, start tagging: Tag semantics emerge from collaborative verbosity},
booktitle = {Proceedings of the 19th International Conference on World Wide Web, WWW '10}
}

@article{bengio_etal_2003,
 author = {Bengio, Yoshua and Ducharme, Jean and Vincent, Pascal and Janvin, Christian},
 title = {A Neural Probabilistic Language Model},
 journal = {J. Mach. Learn. Res.},
 issue_date = {3/1/2003},
 volume = {3},
 month = mar,
 year = {2003},
 issn = {1532-4435},
 pages = {1137--1155},
 numpages = {19},
 url = {http://dl.acm.org/citation.cfm?id=944919.944966},
 acmid = {944966},
 publisher = {JMLR.org}
}

@inproceedings{buckley_vorhees_2000,
 author = {Buckley, Chris and Voorhees, Ellen M.},
 title = {Evaluating Evaluation Measure Stability},
 booktitle = {Proceedings of the 23rd Annual International ACM SIGIR Conference on Research and Development in Information Retrieval},
 series = {SIGIR '00},
 year = {2000},
 isbn = {1-58113-226-3},
 location = {Athens, Greece},
 pages = {33--40},
 numpages = {8},
 url = {http://doi.acm.org/10.1145/345508.345543},
 doi = {10.1145/345508.345543},
 acmid = {345543},
 publisher = {ACM},
 address = {New York, NY, USA},
} 

@inproceedings{agrawal_etal_93,
 author = {Agrawal, Rakesh and Imieli\'{n}ski, Tomasz and Swami, Arun},
 title = {Mining Association Rules Between Sets of Items in Large Databases},
 booktitle = {Proceedings of the 1993 ACM SIGMOD International Conference on Management of Data},
 series = {SIGMOD '93},
 year = {1993},
 isbn = {0-89791-592-5},
 location = {Washington, D.C., USA},
 pages = {207--216},
 numpages = {10},
 url = {http://doi.acm.org/10.1145/170035.170072},
 doi = {10.1145/170035.170072},
 acmid = {170072},
 publisher = {ACM},
 address = {New York, NY, USA},
} 

@InProceedings{hu_etal_2010,
  author        = {M. Hu and E. P. Lim and J. Jiang},
  title         = {A Probabilistic Approach to Personalized Tag Recommendation},
  booktitle     = {2010 IEEE Second International Conference on Social Computing},
  year          = {2010},
  pages         = {33-40},
  month         = {Aug},
  __markedentry = {[felipe:1]},
  abstract      = {In this work, we study the task of personalized tag recommendation in social tagging systems. To include candidate tags beyond the existing vocabularies of the query resource and of the query user, we examine recommendation methods that are based on personomy translation, and propose a probabilistic framework for adopting translations from similar users (neighbors). We propose to use distributional divergence to measure the similarity between users in the context of personomy translation, and examine two variations of such divergence (similarity) measures. We evaluate the proposed framework on a benchmark dataset collected from BibSonomy, and compare with two groups of baseline methods: (i) personomy translation methods based solely on the query user; and (ii) collaborative filtering. The experimental results show that our neighbor based translation methods outperform these baseline methods significantly. Moreover, we show that adopting translations from neighbors indeed helps including more relevant tags than that based solely on the query user.},
  doi           = {10.1109/SocialCom.2010.15},
  keywords      = {identification technology;recommender systems;benchmark dataset;collaborative filtering;distributional divergence;divergence measure;personalized tag recommendation;personomy translation;personomy translation method;probabilistic approach;query resource vocabulary;query user;recommendation method;social tagging system;Context;Equations;Measurement;Probabilistic logic;Tagging;Training;Vocabulary;personalization;tag recommendation, rank2},
  url           = {http://ieeexplore.ieee.org/document/5590886/},
}

@article{efficientestimation,
  author    = {Tomas Mikolov and
               Kai Chen and
               Greg Corrado and
               Jeffrey Dean},
  title     = {Efficient Estimation of Word Representations in Vector Space},
  journal   = {CoRR},
  volume    = {abs/1301.3781},
  year      = {2013b},
  url       = {http://arxiv.org/abs/1301.3781},
  timestamp = {Thu, 07 May 2015 20:02:01 +0200},
  biburl    = {http://dblp.uni-trier.de/rec/bib/journals/corr/abs-1301-3781},
  bibsource = {dblp computer science bibliography, http://dblp.org}
}

@InProceedings{chen_etal_2008,
  author        = {Chen, Hong-Ming and Chang, Ming-Hsiu and Chang, Ping-Chieh and Tien, Ming-Chun and Hsu, Winston H. and Wu, Ja-Ling},
  title         = {SheepDog: Group and Tag Recommendation for Flickr Photos by Automatic Search-based Learning},
  booktitle     = {Proceedings of the 16th ACM International Conference on Multimedia},
  year          = {2008},
  series        = {MM '08},
  pages         = {737--740},
  address       = {New York, NY, USA},
  publisher     = {ACM},
  __markedentry = {[felipe:1]},
  acmid         = {1459473},
  doi           = {10.1145/1459359.1459473},
  isbn          = {978-1-60558-303-7},
  keywords      = {concept detection, flickr, recommendation system},
  location      = {Vancouver, British Columbia, Canada},
  numpages      = {4},
  url           = {http://doi.acm.org/10.1145/1459359.1459473},
}

@InProceedings{goh_etal_2008,
  author        = {D. H. L. Goh and C. S. Lee and A. Y. K. Chua and K. Razikin},
  title         = {An Examination of the Effectiveness of Social Tagging for Resource Discovery},
  booktitle     = {2008 International Workshop on Information-Explosion and Next Generation Search},
  year          = {2008},
  pages         = {23-30},
  month         = {April},
  __markedentry = {[felipe:1]},
  abstract      = {Social tagging allows users to assign keywords (tags) to resources facilitating their future access by the tag creator, and possibly by other users. In terms of its support for resource discovery, social tagging has both proponents and critics. The goal of this paper investigates if tags are an effective means for helping users locate useful resources. Adopting techniques from text categorization, we downloaded Web pages and their associated tags from del.icio.us, and trained Support Vector Machine classifiers to determine if the documents could be assigned to their associated tags. Results from the classifiers in terms of precision, recall and F1 score were mixed, suggesting that that not all tags could be used by public users for resource discovery. Detailed analyses of our results revealed characteristics of effective and ineffective tags for resource discovery. From these, implications for social tagging systems are discussed.},
  doi           = {10.1109/INGS.2008.11},
  keywords      = {Internet;classification;information retrieval;learning (artificial intelligence);support vector machines;text analysis;Web document;Web resource access;Web resource discovery;Web searching;Web site;keyword assignment;social tagging systems;tag creator;text categorization;trained support vector machine classifier;Cultural differences;Navigation;Organizing;Support vector machine classification;Support vector machines;Tagging;Taxonomy;Text categorization;Vocabulary;Web pages;Resource discovery;Social tagging;Support Vector Machines;Text categorization, rank1},
}