About

Learn about the Project

Twitter Sentiment Analysis using Naive Bayes Algorithm

Below Technologies and frameworks have been used to develop the project:

  • Python: Pandas, Numpy, Nltk
  • Twitter API and Tweepy Library
  • Amazon Web Services(AWS): S3, EMR

Below Steps have been used to develop the project:

  • 1. Data Scraping: Collection of Tweets using Twitter API and Tweepy library
  • 2. Data Pre-processing: Multiple “.json” files concatenated to create datasets for Positive and Negative sentiments
  • 3. Model Building: Splitting the data in Training and Testing and building Naïve Bayes model
  • 4. Prediction: Predicting the sentiment of unseen tweets