Sentiment Analysis of Tweets on Social Issues using Machine Learning Approach

Abstract

Sentiment analysis plays an important role in analyzing people's opinions, emotions, and feelings with the help of Natural Language Processing and Artificial Intelligence. This paper presents five major social issues of the world like Corruption, Women violence, Poverty, Child abuse, Illiteracy that are the most critical in the world. These evils are barriers to the social development of the peoples. Tweets from the years 2006 to July 2020 are collected to analyze the sentiments of peoples related to these social issues with Machine learning tools and techniques. In this proposed study different Classification algorithms, NLP techniques like tokenization, Stemming, stop words, Word Cloud are applied for effective enhancement of dataset. Sentiments are analyzed as positive, negative, and neutral from the twitter dataset. Naïve Bayes, Support Vector Machine, Logistic Regression, Random Forest Classifier, Decision Tree, and Stochastic Gradient Descent classifier applied to build the model and also measure the performance by precision, recall, and F1 score parameters.