Approaches, Applications, and Challenges of Using Sentiment Analysis for False Information Detection: A Systematic Literature Review

In today's digital age, the prevalence of false information on the internet, spanning from fake news to fake user accounts, poses a pressing concern that undermines the reliability of online information. Such misinformation can have harmful consequences, such as promoting biased ideologies, swaying public opinion, and inciting unrest. Implementing sentiment analysis technology in false information detection significantly impacts combating misinformation on digital platforms. It analyzes textual content in real time, which is crucial for effectively identifying and mitigating false information such as fake news, fake reviews, fake accounts, and fake websites on digital platforms. This paper explores sentiment analysis's approaches, applications, and challenges for false information detection, providing insights for future research and application. A systematic literature review was conducted using the PRISMA methodologies to identify and evaluate relevant studies. The results of this paper contribute to understanding the current state of sentiment analysis in combating misinformation. This paper provided the common and available applications of sentiment analysis, such as fake news detection, fake review detection, fake social media account detection, and more. Approaches that can be taken to achieve these applications include the machine learning approach, neural network approach, hybrid approach, and more. Moreover, this work highlights the need for more advanced algorithms and ongoing research to address challenges such as sarcasm detection, bot account identification, and effectively interpreting informal language and slang in texts.