International Journal of Computational and Experimental Science and Engineering, Journal Year: 2025, Volume and Issue: 11(2)

Published: April 6, 2025

The rapid proliferation of fake news on social media platforms has raised significant concerns about misinformation, particularly messaging applications like Telegram. This trend poses a severe threat to public trust and harmony. Detecting in such environments requires the development efficient machine learning (ML) models that can accurately identify misleading content while minimizing false positives negatives. research aims propose robust learning-based framework for detecting Telegram by analyzing text user interaction patterns. Data collection involved scraping dataset from publicly available channels, which include both genuine articles with relevant metadata as reactions engagement levels. To address problem detection, set algorithms, including XGBoost, K-Nearest Neighbors (KNN), Decision Trees, Naive Bayes, were explored. A novel ensemble-based approach, termed Ensemble Feature Fusion (EFF), is introduced, combining strengths multiple classifiers enhance predictive accuracy robustness against diverse characteristics. Performance metrics Accuracy, Engagement-Weighted Accuracy (EWA), False Positive Cost (FPC) , Contextual Precision (CP), Temporal Consistency Index (TCI) evaluated this research. Results indicate proposed model outperforms conventional ML techniques, demonstrating improved classification reduced error rates news. approach provides promising solution growing misinformation

Language: Английский