The objective of this project is to perform sentiment analysis (only positive and negative) on an imbalanced hotel review dataset.

This project covers:

TF-IDF count features logistic regression naive bayes svm xgboost grid search word vectors (Universal Sentence Encoder model from Tensorflow HUB) Dense Baseline LSTM Bi-LSTM NN Model Serve Although this project covers classical ml methods and feature engineering techniques, it will mainly focus on using Tf-Hub pretrained model in text classification.

The final LSTM model achieved an accuracy of ~81% in Test Dataset