Deep-Convolutional-Neural-Networks-for-Sentiment-Analysis-of-Short-Texts

Libraries used: Tensorflow, PyTorch, Keras, Numpy, SciPy, Scikit-learn, and Theano.

Implemented the Character to Sentence Convolutional Neural Network (CharSCNN) architecture as described in the research paper (https://www.aclweb.org/anthology/C14-1008.pdf). For training the neural network: used the Stanford Twitter Sentiment corpus (STS), which contains Twitter messages.

To duplicate the project results, clone this repository and run the Torun.py file from the src folder.

Presentation of the project work: