An intelligent sentiment analysis system based on recent trends in machine learning

Khan Awais Khan; Saadat Hanif Dar; Samina Fazilat; Muhammad Raees; Sehrish Manzoor; Muhammad Usama Ayub; Afzal Ahmed; Areeb Ahmed Mir

Social networks are producing data at an ever-growing rate. Social nets are the platforms for conveying opinions (or sentiments) on varying concerns. Sentiment analysis is a way of determining the subjectivity of the message/opinion. The sentiment analysis on social nets (like Twitter), assesses the sentiments represented by the users in their tweets with tweets expressing ideas, interests, and opinions in a range of settings. Consequently, Twitter sentiment analysis is a hot research area at present. We aim to concentrate on sentiment analysis on Twitter data, which is beneficial for discovering insight into the words used in the tweets, where opinions are strong, unstructured, and have either positive, negative, or neutral influence. Tweets are an amalgam of complex characters and complicated expressions requiring efficacious computation. In this study, we present various algorithms and methodologies employed on Twitter data for sentiment analysis. Here, the goal of sentiment analysis is to gauge or quantify the emotional perspective of the author against a certain issue.