The Development of a Multi-Strategy Fake News Detection System that Incorporates Source Trust Evaluation
Year: 2025
Type: article
Abstract: A multi-strategy fake news detection system is proposed, combining machine learning (ML) and natural language processing (NLP) techniques to address the growing spread of misinformation. The framework includes multiple models: XGBoost, Support Vector Machine (SVM), Naïve Bayes, Random Forest, and a CNN-LSTM hybrid. The framework adds sentiment analysis, fact-checking using BERT, semantic similarity using Word2Vec, and trustworthiness scoring. The system was implemented in a way to help with detection accuracy and trustworthiness. The results demonstrate that our fake news detection system is reliable, accurate and suitable for detecting and classifying fake news articles. Standard performance measures of accuracy, precision, recall and F1-score were used to evaluate the system and showed that our multi-way approach architecture provided reliable and accurate results and would be suitable for real-world usage.
Authors: Putta Tejaswini, Kagitha Sashidhar, Padamata Kavya, R. Mabu Basha
Language: English
Subfield: Sociology and Political Science
Field: Social Sciences
Domain: Social Sciences
Open Access status: gold
OpenAlex ID: W4410697349