MCP-Orchestrated Multi-Agent System for Automated Disinformation Detection

TL;DR

This paper introduces a multi-agent system orchestrated by MCP for detecting disinformation in news, combining machine learning, knowledge checks, coherence detection, and relational analysis, achieving high accuracy and scalability.

Contribution

The paper presents a novel MCP-orchestrated multi-agent system integrating diverse agents for disinformation detection, with a focus on relation extraction and live learning.

Findings

Achieves 95.3% accuracy and 0.964 F1 score.

Weighted aggregation outperforms threshold optimization.

Modular architecture ensures scalability and transparency.

Abstract

The large spread of disinformation across digital platforms creates significant challenges to information integrity. This paper presents a multi-agent system that uses relation extraction to detect disinformation in news articles, focusing on titles and short text snippets. The proposed Agentic AI system combines four agents: (i) a machine learning agent (logistic regression), (ii) a Wikipedia knowledge check agent (which relies on named entity recognition), (iii) a coherence detection agent (using LLM prompt engineering), and (iv) a web-scraped data analyzer that extracts relational triplets for fact checking. The system is orchestrated via the Model Context Protocol (MCP), offering shared context and live learning across components. Results demonstrate that the multi-agent ensemble achieves 95.3% accuracy with an F1 score of 0.964, significantly outperforming individual agents and traditional approaches. The weighted aggregation method, mathematically derived from individual agent misclassification rates, proves superior to algorithmic threshold optimization. The modular architecture makes the system easily scalable, while also maintaining details of the decision processes.