Provision a generalizable approach to the ranking of credible Twitter users

The abundance of information shared on social networks presents valuable opportunities, such as timely news coverage and user needs forecasting. However, the lack of oversight facilitates the spread of fake content across various fields. Therefore, evaluating user credibility is crucial for responsible social media usage. This paper proposes a topic-based user ranking system on Twitter. The system leverages machine learning algorithms to prioritize user credibility based on specific topics and introduces new features for comprehensive evaluation. Finally, users are rated with 5 models of machine learning, Linear Regression (LR), Support Vector Regression (SVR), k-nearest neighbors (KNN), Random Forest (RF) and Decision Tree (DT) that, DT has achieved the highest accuracy of 82%. This approach offers a generalizable solution for various user credibility assessment needs.