Abstract: Online social networks have become increasingly popular as a means of social communication and sharing information in the digital era. The growing demand for social interaction has resulted in the spread of fake news on social media platforms. The depth and wide scope of the harm of fake news have made its detection one of the hot research topics in natural language processing. This paper first introduces the background, significance, related terms, and problem description of fake news detection. It then explains the process of data collection and feature extraction. Considering the relationship between features and news content, this paper reviews and categorizes fake news detection methods based on internal and external features. Finally, this paper highlights the research challenges that still exist in fake news detection and presents a systematic outlook on future research directions.