Retrieval-Augmented Ontologic Graph Model with GNN Integration to Enhance Inductive Reasoning

Tracking #: 1029-2069

Flag : Review Assignment Stage

Authors: 

Hien Nguyen
Duc Truong
Sang Vu
Tai Huynh
Vuong Pham
Hung Nguyen

Responsible editor: 

Guest Editors X-NeSy

Submission Type: 

Article in Special Issue (note in cover letter)

Full PDF Version: 

Cover Letter: 

This paper is submitted to Special Issue on Explainable Neurosymbolic AI (X-NeSy) This paper introduces a novel e-learning knowledge retrieval method, significantly enhancing information access through the integration of ontology and graph-based reasoning. Ontology provides a structured framework for domain knowledge, while knowledge graphs visualize the intricate connections between concepts. By employing graph-based reasoning, the designed system can infer relations between interrelated topics to deliver deeper, more contextually relevant answers to user queries. This approach effectively addresses the challenge of knowledge gaps within e-learning materials. The model demonstrates the ability to deduce new information from existing data, bridging these gaps and enriching the learning experience. Experimental results confirm the superior performance of the proposed method compared to existing techniques, showcasing its capacity to provide more accurate and comprehensive information. To validate its practical application, it has been applied to develop an intelligent assistance system for the Database Systems course. This real-world implementation demonstrates the system's potential to revolutionize e-learning, offering personalized support and fostering a more effective and engaging learning environment. The integration of ontology and knowledge graphs, coupled with graph-based reasoning, presents a significant advancement in e-learning knowledge retrieval.

Tags: 

  • Under Review