In this repo, I've trained some Machine Learning and Deep Learning models to classify amazon product reviews.
Machine Learning algorithms (sklearn) used in this projects,
- LogisticRegression
- LogisticRegressionCV
- MultinomialNB
- BernoulliNB
- SVC
- LinearSVC
- PassiveAggressiveClassifier
- SGDClassifier
- CalibratedClassifierCV
- KNeighborsClassifier
Checkout the ML based classifiers part-1 (Data Collection) here and part-2 (Training and Testing) here and part-3 (Start Predicting) here .
After Hyperparameters tuning, the best deep learning training score is 94.20% & testing score is 89.39%.
Checkout the DL based classifiers part-1 (Data Collection) here and part-2 (Training and Testing) here and part-3 (Start Predicting) here
After Hyperparameters tuning, the best deep learning (LSTM + Keras Embedding) training score is 97.20% & testing score is 88.39%.
Checkout the classifier part-1 (Data Preprocessing) here and part-2 (Training and Testing) here and part-3 (Start Predicting) here
Checkout the classifier part-1 (Data Preprocessing) here and part-2.1 (Training and Testing) here and part-3.1 (Start Predicting) here