In the file NLP-quora classification, explored text analysis and machine learning methods (merged NaiveBayes+Logistic regression) to get a comparable results with the state of art deep leanring technique;
In the file DL-quora, used word-embedding and LSTM to train the model.