Sentiment analysis methods, applications, and challenges: A systematic literature review

With the expansion of Internet-based applications, the number of comments shows explosive growth. Analyzing the attitudes and emotions behind comments provides powerful assistance for businesses, governments, and scholars. However, it is hard to effectively extract user's attitude from the massive amounts of comments. Sentiment analysis (SA) provides an automatic, fast and efficient tool to identify reviewers' opinions and sentiments. However, the existing literature reviews cover a limited number of studies or have a narrow field of studies for sentiment analysis. This paper provided a systematic literature review of sentiment analysis methods, applications, and challenges. This systematic literature review gives insights into the goal of the sentiment analysis task, offers comparisons of different techniques, investigates the application domains of sentiment analysis, highlights the challenges and limitations encountered by scholars, provides suggestions on possible solutions and explores the future research directions. The study's findings emphasize the significant role of artificial intelligence technologies in automatic text sentiment analysis and highlight the importance of sentiment analysis in people's production and life. This research not only contributes to the existing sentiment analysis knowledge body but also provides references to scholars and practitioners in choosing a suitable methodology and good practices to perform sentiment analysis.