Analysis Fake News Detection Models Using Swarm Intelligence

The proliferation of fake news on social media platforms has become a significant societal challenge, undermining public discourse and democratic processes. This study proposes a novel approach to fake news detection by leveraging swarm intelligence techniques, specifically a Modified Ant Colony Optimization (ACO) algorithm, in combination with established machine learning models. Our method employs a four-stage process: data preparation, feature extraction using TF-IDF vectorization, feature selection via Modified ACO, and model evaluation. We compare the performance of Random Forest, XGBoost, and Support Vector Classifier (SVC) models before and after dimensionality reduction. The WELFake dataset, comprising 72,134 news articles, is used to train and evaluate the models. Results demonstrate that the proposed approach achieves high accuracy in fake news detection while significantly reducing computational time, particularly after dimensionality reduction. This research contributes to the development of more efficient and accurate fake news detection systems, addressing the critical need for robust tools to combat misinformation in the digital age.