Authors: Huiling Tu, Shuo Yu, Vidya Saikrishna, Feng Xia, Karin Verspoor

Published on: February 06, 2024

Impact Score: 8.15

Arxiv code: Arxiv:2402.03732

Summary

• What is new: DEAN introduces a novel deep learning framework for detecting outdated facts in knowledge graphs by leveraging implicit structural information.
• Why this is important: The challenge of outdated facts in knowledge graphs, which compromises their quality as real-world information changes.
• What the research proposes: The DEAN framework, which models entities and relations comprehensively and employs a contrastive approach for latent out-of-date information detection.
• Results: Experimental results show DEAN’s effectiveness and superiority over existing state-of-the-art baseline methods.

Technical Details

Technological frameworks used: DEAN (Deep outdatEd fAct detectioN)

Models used: Contrastive approach based on Relations-to-Nodes (R2N) graph

Data used: nan

Potential Impact

Companies relying on knowledge graphs for information retrieval, search engines, recommendation systems, and data analysis services might benefit or need to adapt.

Want to implement this idea in a business?

We have generated a startup concept here: TimeStream.