frankenstein32 / Movie-Rating-Predicton

Predicting Movie rating corresponding to the Review.

Movie-Rating-Predicton

Predicting Movie rating corresponding to the Review Using Naive Bayes Algorithm.

• Data is Given in the form of the text, So NLP is used to clean the data set and then processing the data using the Bag of Words model.

Flow of the program

• Clean the data given by created a pipeline.

  • Tokenizing the review.
  • Removing the Stop words.
  • Stemming the words created.
  • Store the cleaned review in a new file.

• Training.
  Training of the data is done by the algorithm Naive - Bayes and usinf the bag of words Model.

• Prediction
  Prediction is made my comparing the prediction of Validation data and the true labels of the data.

How to run

• Run the clean_review.py file and give input file name and output file name along with the command in the terminal. for e.g
  

    rashid@rashid-ideapad: python3 clean_review.py imdb_trainX.txt x_train.txt<br>
  

With this command the clean_review file will clean the text present in the imdb_trainX.txt and write into the x_train.txt

• Do the same cleaning for the imdb_testX.txt file too.
• Now simply run the movie_prediction.py file
• It will print the training accuracy and Testing accuracy
• Keep the files and code in the same directory otherwise you need to change the path for reading and writing the files.