Performance Evaluation of Fake News Detection Using Artificial Intelligence Techniques

Introduction/Importance of Study:As the proliferation of fake news poses significant challenges to traditional fact-checking methods, there is a growing need for robust and automated approaches to combat misinformation.Novelty statement:This study presents a comprehensive evaluation of artificial models for fake news detection, offering insights into their effectiveness and applicability in addressing the contemporary issue of misinformation.Material and Method:The research employs various artificial algorithms, including logistic regression, gradient boosting, decision trees, random forest, AdaBoost, passive aggressive classification, XGBoost, naive Bayes, and support vector machines (SVM), to train datasets and evaluate the performance of each model.Result and Discussion:Through rigorous evaluation, the study finds that XGBoost and AdaBoost classifiers exhibit the highest accuracy rates of 99.83% and 99.77%, respectively, in detecting fake news. Decision Tree, Support Vector Machine, and Gradient Boosting classifiers also demonstrate commendable performance. Conversely, the Naive Bayes classifier exhibits the lowest accuracy, suggesting its limitations in fake news detection.Concluding Remarks:This research underscores the significance of ensemble methods such as XGBoost and AdaBoost in effectively identifying fake news, laying the groundwork for future advancements in combatting misinformation.