Document Type

Article

Publication Title

International Journal of Future Engineering Innovations

Abstract

The proliferation of fake news across digital platforms poses a significant threat to democratic processes, public health, and social cohesion in the United States. Manual fact-checking and traditional moderation approaches are increasingly inadequate due to the scale, speed, and sophistication of misinformation campaigns. This presents an AI-driven framework for real-time fake news detection, designed to serve as an intelligent filter for news platforms operating within the US media ecosystem. Leveraging natural language processing (NLP) techniques and machine learning algorithms—including supervised classifiers and deep learning models such as BERT and LSTM—the framework identifies deceptive content with high precision and low latency. The proposed architecture integrates news ingestion pipelines, contextual feature extraction, and classification modules capable of operating on streaming data. Publicly available labeled datasets such as LIAR, FakeNewsNet, and PolitiFact were utilized for training and evaluation, ensuring robustness and generalizability. The framework also includes a dynamic feedback loop to continuously improve performance through human-in-the-loop validation and real-time user engagement data. A systematic literature review guided by PRISMA methodology was conducted to inform model selection, dataset validation, and deployment strategies. Experimental results demonstrate that hybrid models combining linguistic, semantic, and social context features achieve superior performance over traditional baselines. Ethical, legal, and societal considerations—including transparency, free speech implications …

First Page

158

Last Page

169

DOI

10.54660/IJFEI.2025.2.4.158-169

Publication Date

7-23-2025

Download

Share

COinS