A Systematic Review On Sentiment Analysis Approaches

With the rise of social media on the internet, one of the most important study fields is sentiment analysis. Millions of people trade their thoughts, ideas, expressions, sentiments, and opinions on social media sites such as Twitter, YouTube, Facebook, and others, infusing a lot of passion into human life these days. Sentiment analysis, often known as opinion mining, is primarily concerned withthe classification and prediction of people's attitudes toward a certain topic. It's also known as “emotion mining” or “feeling mining.” It entails categorizing written documents or sentences into positive or negative categories based on the expressed perspective on a certain issue. Although sentiment analysis appears to be comparable to text classification, it confronts a number of obstacles that have prompted a lot of research in this area. Different machine learning (ML) & lexicon-based algorithms have been developed in the narrative to automate the sentiment analysis task. We provide the findings of a tertiary study in this paper, which intends to investigate the current level of research. So that future investigators with better knowledge might create new automated approaches that solve all of the problems and provide the greatest results.