Fake News Detection Using Machine Learning and Deep Learning Algorithms

The rapid spread of misinformation and fake news across digital platforms poses significant challenges to social stability, political discourse, and cybersecurity. Traditional rule-based detection methods often fail to capture the linguistic complexity and evolving nature of fake news. This paper presents a comparative performance evaluation of multiple machine learning algorithms— including Logistic Regression, Naïve Bayes, Random Forest, Support Vector Machines (SVM), and Convolutional Neural Networks (CNN)—for fake news detection. Using benchmark datasets, the study analyzes preprocessing techniques such as TF-IDF (Term Frequency–Inverse Document Frequency) and word embeddings, and evaluates models based on accuracy, precision, recall, F1-score, and ROC-AUC. A survey of participants was conducted to understand perceptions of fake news detection tools and the importance of algorithmic transparency. Results indicate that ensemble and deep learning approaches can provide strong predictive performance, while highlighting trade-offs between interpretability and predictive power. The findings suggest that hybrid models combining statistical and deep learning techniques may enhance detection accuracy and resilience against misinformation.