Systematic Literature Review on Methods used in Classification and Fake News Detection in Indonesian

Currently, around 150 million Indonesians are online on social media, and around 70.7
^%
of fake news was received in the form of text or writing. Recent studies of fake news text detection classification have resulted in data sets and methods that allow them to explore and understand fake news detection and clasifications. So many techniques, algorithms, methods, and published data on the classification of fake news texts' classification are complex and different, making it difficult to get a comprehensive picture of the status of research on the classification of fake news. The literature search study based on Systematic Literature Review aims to identify and analyze trends, topics, datasets, and methods and answer research questions in classifying fake news text detection in Bahasa between 2017 and 2021. By exploring the literature, this research uses the SLR-based method to address the gap in the works on fake news and text classification in Indonesian that has received a little empirical investigation. Based on the data search, thirty related studies were obtained, which were further analyzed. However, there are still few studies that have a good ranking and reputation. Nineteen methods or algorithms are applied to classify fake news, and there are two methods or algorithms that are most often used namely Naïve Bayes and Term Frequency - Inverse Document Frequency.