Skip to content

Latest commit

 

History

History
7 lines (4 loc) · 408 Bytes

README.md

File metadata and controls

7 lines (4 loc) · 408 Bytes

review-based-recommender-system

In this model we use UserId ,ItemId and Reviews as Input and predict a Rate as Output. You can use Amazon Review Dataset for test it.

In this model we use Deep Learning algorithms like LSTM and CNN to have better performance, Attention Mechanism to extract important part of reviews, Factorization Matrix to predict Ratings and Word Embedding to convert words to matrix.