FAKE NEWS DETECTION USING MACHINE LEARNING TECHNIQUES: A COMPARATIVE STUDY

Abstract

The increasing spread of misinformation across online platforms has made fake news detection an essential research area in data science and artificial intelligence. This study presents a comparative analysis of various machine learning algorithms for detecting fake news using Natural Language Processing (NLP) techniques. The dataset, collected from publicly available sources such as the Kaggle and LIAR datasets, underwent preprocessing steps including tokenization, stopword removal, and lemmatization. Text features were extracted using the Term Frequency–Inverse Document Frequency (TF-IDF) method to convert textual data into numerical representations. Multiple supervised algorithms—Naïve Bayes, Logistic Regression, Random Forest, Support Vector Machine (SVM), Decision Tree, Gradient Boosting, and K-Nearest Neighbors (KNN)—were implemented and evaluated using performance metrics such as accuracy, precision, recall, F1-score, and ROC-AUC. Among the tested models, Support Vector Machine (SVM) and Logistic Regression achieved the highest accuracy, demonstrating their efficiency in classifying high-dimensional textual data. The results confirm that appropriate preprocessing, feature extraction, and algorithm selection significantly enhance model performance, establishing a robust framework for automated fake news detection.