Skip to content

Latest commit

 

History

History
10 lines (9 loc) · 892 Bytes

README.md

File metadata and controls

10 lines (9 loc) · 892 Bytes

This Project is an NLP project. In which it predicts the rating of a book review, by analyzing the review. Dataset used in this project - https://drive.google.com/drive/folders/1e4UT9sebo7VzwuXMjlg0RdES7s30DSfx?usp=sharing.

  • Used relatively big data with 6,30,000 Reviews.
  • Text Preprocessing : Tokenization, Lowercasing, Removing Punctuation, Removing Stopwords, Stemming or Lemmatization
  • Feature Extraction : Bag-of-Words, TF-IDF, Word Embeddings ( Word2Vec )
  • Building Models: Logistic Regression, RandomForestClassifier, XGBoostClassifier, GradientBoostingClassifier, AdaBoostClassifier, DecisionTreeClassifier, KNeighborsClassifier -Model Evaluation and Tuning: Finalized XGBoostClassifier and performed hyperparameter tuninig using RandomSearchCV
  • Deployment: Webapp Using Streamlit library.
  • App is live - https://huggingface.co/spaces/Suyash1302/Book_Review_Rating_Predictor