Abstract:

While in traditional news media, the content of the news is based on fake news detection, social context information in social media can be used to help detect fake news. Considering the social context, the distribution of data on social media and the interaction of online users with each other, it also explains the dissemination of news in the social environment and provides the necessary information to determine whether the news is true or not. Social media supports models based on news content. Developing these models provides some additional resources for researchers. Social context information represents three main topics: user detail, post and network analysis. In this study, a compilation study was conducted on the social context-based features and models of fake news from a data science perspective. Studies using these features and models in the literature have been examined with both machine learning and deep learning approaches. Analysis of 9 known data sets created for feature extraction and fake news detection was performed.

Keywords: