Sentiment analysis of climate change using Twitter API and machine learning

Abstract

Climate change, also called global warming, is one of the most frequently discussed topics nowadays. Also, there has been a tremendous growth in the research for sentiment analysis on social network by using natural language processing (NLP). Sentiment analysis has been widely applied in various commercial and non-commercial areas. People have different opinions on various topics with positive, negative or neutral comments on social media. This thesis work performs sentiment analysis of content from Twitter with climate change hashtags, using Twitter API for authentication and tweepy libraries. Multinomial Naive Bayes Classifier is an optimal method which is selected for training model and detecting people’s opinions on climate change. The sentiment analysis has shown the result of 67% accuracy. The findings have indicated that the majority of the training group has negative sentiment on climate change, whereas the minority has an optimistic attitude towards climate change.