Fake News Detection on Social Media Through a Hybrid SVM-KNN Approach Leveraging Social Capital Variables

The spread of false information on social media platforms has turned into a serious problem, requiring the creation of cutting-edge detecting tools. Conventional approaches that just consider language and compositional analysis are frequently insufficient to combat the constantly changing strategies used by people who create and disseminate false material. In response to this difficulty, a novel hybrid model that successfully identifies bogus news has been unveiled. It combines the advantages of Support Vector Machines (SVM) and K-Nearest Neighbors (KNN). To identify the most important aspects that impact the identification of false news, this model prioritizes the complex relationships that exist between individuals, news content, and social networks, all within the context of social capital. The proposed approach incorporates a range of machine learning classifiers, comprising SVM, Random Forest (RF), Logistic Regression (LR), Classification and Regression Trees (CART), and Neural Networks (NNET). To ensure the robustness and generalizability of the results, a rigorous cross-validation process has been applied to these classifiers. This rigorous methodology provides a strong foundation for enhancing fake news detection systems, ultimately strengthening our defenses against the ever-evolving landscape of sophisticated fake news generation and dissemination on social media. By combining the power of SVM and KNN, this innovative hybrid model transcends the limitations of conventional techniques, allowing for a more nuanced understanding of fake news detection. It acknowledges the multifaceted nature of the problem, recognizing that fake news is not solely a linguistic or compositional issue but a complex interplay of various elements, including user behavior and social network dynamics. This holistic approach is a promising step toward effectively combating the pervasive spread of misinformation and disinformation in our digital age.