Sentiment analysis is the process of using natural language processing, text analysis, and statistics to analyze customer sentiment. Customer sentiment can be found in tweets, comments, reviews, or other places where people mention your brand. In this project, tweets of users have been used to predict their sentiment. Tweets were classified into 2 categories - positive and negative. LSTM-based deep learning model has been used to predict the sentiment of tweets.
Twitter Sentiment Dataset was used as the dataset in this project. This dataset consists of 2 files named train.csv that contains 31962 tweets along with their corresponding labels and test.csv that contains 17197 tweets.
- Clone this repository
git clone <URL of the repository>- Download glove embeddings file named "glove.6B.50d" from here
- Unzip the file
- Create a new directory in the parent folder named "glove"
mkdir glove- Move the unzipped glove.6B.50d file in this directory
- Run the sentiment_predictor.py file
python sentiment_predictor.pyFor any issues related to this project, you can contact me at:
- LinkedIn: Aaquib Asrar
- Gmail: [email protected]