Sentiment analysis and opinion mining are the studies of interpretation by analyzing texts in which people express their opinions about a subject, group, product, brand, or situation with applications with natural language processing, artificial intelligence, or statistics. In recent years, with the increase in the use of social media and other platforms where users share ideas, the amount of data reaching the level of terabytes per hour has increased the importance given to sentiment analysis and opinion mining. Within the scope of this study, a literature review on current articles and articles on machine learning approaches, lexicon-based approaches, and hybrid approach in sentiment analysis is presented to the researchers. From the articles examined, a table containing the publication date of the article, research problem, approach, preprocessing and feature selection methods, classification algorithm, model success criterion, an algorithm with the highest success rate and success rate, data source information was created. Since machine learning-based methods are frequently preferred and the number of studies is higher than other methods, they are discussed separately under the titles of supervised, unsupervised, semi-supervised, and deep learning. At the end of the study, the study was completed by making a general evaluation in light of the articles examined.