Code for the CONLL-SIGMORPHON-2018 shared task

Natural Language Processing Machine learning with python and keras (text generator)

A keras implementation of Bidirectional-LSTM for Named Entity Recognition.

Applying NLP models to detect sarcasm for the Twitter datasets using Bidirectional LSTMs

A news article's title and description should be classified into the following groups in order to solve this classification problem: 1-World, 2-Sports, 3-Business and 4-Science/Tech .Here is a sequence of data. This is a sequential problem, thus we may use bidirectional LSTM for classification since we have access to the data.

predictive analysis for currency exchange rates, still under active development

A simple many-to-one string generating Bidirectional LSTM using pytorch

This repository aims is dedicated to exploring NLP problems using deep learning

AmazonReviewNLP is a deep learning project that utilizes LSTMs for sentiment analysis on Amazon customer reviews.

I crafted a robust sentiment analysis model featuring Bidirectional LSTM layers and embeddings. Leveraging a Sequential model structure and fine-tuning with 'sparse_categorical_crossentropy' loss and 'adam' optimizer, the implementation excels in capturing contextual nuances for precise emotion classification in natural language data.

Bidirectional RNNs are used to analyze the sentiment (positive, negative, neutral) of movie reviews. .

Master Thesis Code

Sentiment classification of IMDB movie reviews using Natural Language Processing models

This is a Pytorch implementation of the paper "Correlation Networks for Extreme Multi-label Text Classification" by Guangxu Xun, Kishlay Jha, Jianhui Sun, Aidong Zhang.

The repository focuses on developing a comprehensive business opportunity analysis system that uses geospatial data, sentiment analysis, and topic modeling. The objective is to leverage these techniques to identify and evaluate potential business opportunities in area of interest.

This repository is created for our final year research project

The model is character-based, for each character the model looks up the embedding, runs the GRU one timestep with the embedding as input, and applies the dense layer to generate logits predicting the log-likelihood of the next character.

A Deep Learning approach of classifying the news headlines and content as Fake or Real.