A Multimodal Machine Learning Framework for Fake News Detection Using Image–Text Semantic Similarity
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Globally, the quick dissemination of false information via digital channels has led to significant social, political, and economic problems.Inconsistencies between photographs and accompanying text are frequently overlooked by traditional false news detection techniques, which mostly concentrate on textual content.This paper suggests a multimodal machine learning framework for detecting false news based on image-text semantic similarity analysis in order to overcome this drawback.To find semantic discrepancies between written and visual information, the suggested approach combines similarity calculation, picture representation, and textual feature extraction.To differentiate between instances of fake news and true news, machine learning classifiers are developed using extracted multimodal data.The labelled news items with textual descriptions and related photos make up the dataset used in this work.The model is assessed using common performance metrics including accuracy, precision, recall, and F1-score, following preprocessing and feature engineering.When compared to text-only methods, experimental results show that adding multimodal similarity features enhances detection performance.A web application built with Django is used to deploy the system, allowing for real-time prediction of bogus news.The suggested approach offers a workable and scalable way to counteract false information in digital media settings.
Fields of Study & Subjects
Misinformation and Its Impacts
Spam and Phishing Detection
Multimodal Machine Learning Applications
Publication details
- Journal / source
- International Research Journal of Modernization in Engineering Technology and Science
- Publication date
- 31 March 2026
- Language
- EN