Skip to content

aknakshay/NLP_Sentiment_Classification_IMDB_Reviews

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 

Repository files navigation

NLP_Sentiment_Classification_IMDB_Reviews

Performing Sentiment Classification on IMDB Reviews using Deep Learning.

The task is to classify the sentiment of the text.

Cleaning

Set of Stop words is taken and the stop words which can have an impact on sentiment are removed from stop word set. Text is cleaned by removing html tags, removing limited stop words, and performing case normalization.

Vocabulary Filteration

The data is further filtered by restricting the vocabulary size to 10000 of most common words.

Encoding and Padding

Data is then encoded into numbers and padded to have the same length.

Model

A Bidirectional GRU model is trained over data with tanh activation. In two epochs only, model fits the data and gives an accuracy of around 87%.

About

Performing Sentiment Classification on IMDB Reviews using Deep Learning.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors