FAKE NEWS IDENTIFICATION AND CLASSIFICATION USING MACHINE LEARNING

Abstract

This paper investigates the application of traditional machine learning algorithms for the detection of fake news using the "Fake News Detection: The Battle Against Misinformation" dataset. The study implements and evaluates the performance of Naive Bayes, Support Vector Machine (SVM), Logistic Regression, K-Nearest Neighbors, and Support Vector Classifier (SVC) models on this binary classification task. The dataset comprises Fake - 5000, Real - 4900 news items labeled as either "Fake" or "Real." The performance of each model is assessed using key evaluation metrics including accuracy, precision, recall, and F1-score, based on experimental results obtained on a held-out test set. The findings reveal that SVC is the best performing models based on results, SVM, Logistic Regression, and SVC achieve high accuracy around 99%, demonstrating their effectiveness in distinguishing between real and fake news within this dataset. The study provides a comparative analysis of these classical machine learning approaches, highlighting their strengths and limitations for fake news detection and offering insights for future research in this critical area.