Sentiment Analysis of Tweets Using Machine Learning Algorithms

Abstract

Social media is a popular platform for individuals to express their opinions on various topics. Sentiment analysis using Twitter datasets addresses important requirements such as conducting market research, performing political analysis, and gaining social insights. It empowers businesses to derive valuable insights from user-generated content on Twitter. The objective of this research is to create a sentiment analysis model that can be applied to social media data, specifically tweets from Twitter. The sentiment analysis model is built using machine learning (ML) techniques and natural language processing (NLP) to determine if each tweet is positive or negative. This allows organizations to gauge public opinion and make informed decisions. A vast collection of tweets is used to test the model's performance, with a focus on accuracy, precision, and recall metrics. The results show that the model is successful in analyzing sentiment across various topics and domains. Out of the various machine learning algorithms examined, the Support Vector Machine (SVM) algorithm yielded the highest accuracy of 94.65%. The study as a whole emphasizes the importance of sentiment analysis in comprehending public sentiment on social media platforms.