Using machine learning techniques for detecting Fake News

Abstract

With the evolution of technology and the rise of social media, the proliferation of fake news has speed across a wide range of media platforms. This proliferation poses a significant challenge, exacerbated by the growing number of social media users and declining digital literacy. Existing solutions for detecting fake news have short comings, with the complexity of the task influenced by factors such as language type, news category and topic volatility. Machine Learned (ML) and Natural Language Processor (NLP) techniques offer viable means to address this issue by identifying patterns unique to fake news articles that are not present in authentic news content. This study deals with a classification-based approach for the automation of fake news detection. Several methods were employed, including experimentation with different features (Count-Vectorizer, Tf-Idf Vectorizer, Bag-of-Words) and machine learning models (SVM, KNN, Random Forest, Naive Bayes, Decision Tree) to construct accurate detectors. Experiments were conducted on a real-world dataset, LIAR, to evaluate the performance of the models. The results showed that the SVM model using Tf-Idf Vectorizer features achieved the highest accuracy at 92%. These findings highlight the potential of Machine Learning models in the field of fake news detection, with a promising trajectory for further advancements in the future.