BICT Honors Degree 2025

A Hybrid RAG-Based Framework for Real-Time Verification of Health Claims on Social Media Using Knowledge Graphs and Vector Embeddings

Using Knowledge Graphs and Vector Embeddings to improve the accuracy, scalability, and transparency of fact verification on social media.

Aims and Objectives

Our main goal is to design and evaluate a hybrid RAG framework that improves accuracy, scalability, and transparency in fact-checking.

Enhance Accuracy
Combining structured (Knowledge Graphs) and unstructured (Vector Embeddings) knowledge sources for robust claim verification.
Provide Transparency
Generating clear reasoning paths and evidence trails to build user trust in the verification process.
Ensure Scalability & Adaptability
Handling high-volume, fast-evolving data streams for practical deployment on social media platforms.

Our Methodology

Our hybrid RAG pipeline uses a multi-stage process to verify social media posts with high accuracy and transparency.

01

Intake

A post (text + optional links) is submitted and routed via WebSockets for processing.

02

Embedding & Categorization

The post is converted into a semantic embedding and classified into a domain (e.g., politics, health).

03

Retrieval & Ranking

A dual retrieval process from a Vector DB and Knowledge Graph, followed by a hybrid re-ranking of results.

04

Corrective Retrieval

An agentic loop creates focused searches on external sources for failed facets and re-ranks evidence.

05

Evidence Validation

User-submitted sources are validated for trust and entailment. An admin panel governs knowledge base growth.

06

RAG Generation

A RAG prompt is constructed with the post and evidence. The LLM outputs a truthfulness estimate and explanation.

07

Delivery & Feedback

Results are delivered back to the platform, and user feedback is collected for continuous improvement.

Meet Our Research Team

The driving force behind this research is a team of passionate and dedicated students and their supervisor.

Portrait of D.S.Athapattu
D.S.Athapattu
Student Researcher

Focus Area: Software Technology. University of Sri Jayewardenepura.

ICT/21/809

Portrait of N.M.N.L.N. Bandara
N.M.N.L.N. Bandara
Student Researcher

Focus Area: Software Technology. University of Sri Jayewardenepura.

ICT/21/813

Portrait of H.D.H. Chandrasiri
H.D.H. Chandrasiri
Student Researcher

Focus Area: Software Technology. University of Sri Jayewardenepura.

ICT/21/819

Portrait of Mrs. Sankani Heenkenda
Mrs. Sankani Heenkenda
Supervisor

Lecturer, Faculty of Technology, University of Sri Jayewardenepura.

System Architecture

An overview of the system architecture for our social media oriented agentic RAG verification system.

System Architecture for Social media oriented agentic RAG verification