EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Towards Building a Multi-Source Heterogeneous Knowledge Graph for Complex Material Question Answering

Peize Li, Xi Guo, Nan Yin, Yiquan Deng, Lei Zhang, Jian Liu, Jie He

Electronics · 14.08.2026 · 15 (16) · 3615

DOI: 10.3390/electronics15163615 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 14.08.2026. Verfasst von Peize Li und 6 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Advanced Materials-National Science and Technology Major Project, Fundamental Research Funds for the Central Universities.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Large Language Models (LLMs) show considerable potential for materials-science question answering. However, LLM responses may still be affected by unsupported parametric associations, while dense Retrieval-Augmented Generation (RAG) can fragment relational evidence across text chunks. Moreover, general graph-based retrieval does not necessarily preserve the hierarchical relations and factual attributes required to resolve implicit material constraints. To address these limitations, we propose MCTD-KG, a multi-source heterogeneous knowledge graph integrated with a Knowledge-Enhanced RAG framework for complex material question answering. MCTD-KG adopts a Classification–Term–Data ontology to connect disciplinary taxonomies, domain concepts, semantic relations, and empirical records from toolbooks and the scientific literature. Through LLM-assisted knowledge extraction, entity normalization, and multi-source integration, the resulting graph contains more than 530,000 entities across three layers, including 61,768 text-extracted Term-layer entities. During inference, Dual-Channel Retrieval jointly retrieves query-relevant relational paths and associated material attributes, while an explicit semantic filtering stage screens candidate evidence against the query constraints. Evaluation on an expert-validated benchmark of 1577 questions shows that the proposed framework achieves an overall accuracy of 68.48%, compared with 17.40% for the zero-shot Pure LLM, 24.79% for the best Vanilla RAG setting, and 44.96% for GraphRAG. It also achieves 45.22% accuracy on four-hop questions, compared with 39.49% for GraphRAG. These results indicate that integrating multi-source domain knowledge with relation-preserved retrieval and attribute-supported filtering provides more focused and inspectable evidence, thereby supporting more accurate complex material question answering.

Förderung

  • Advanced Materials-National Science and Technology Major Project
  • Fundamental Research Funds for the Central Universities

Zitieren

APA

Li, P., Guo, X., Yin, N., Deng, Y., Zhang, L., Liu, J., & He, J. (2026). Towards Building a Multi-Source Heterogeneous Knowledge Graph for Complex Material Question Answering. Electronics, 15(16), 3615. https://doi.org/10.3390/electronics15163615

Vancouver

Li P, Guo X, Yin N, Deng Y, Zhang L, Liu J, et al. Towards Building a Multi-Source Heterogeneous Knowledge Graph for Complex Material Question Answering. Electronics. 2026;15(16):3615. doi:10.3390/electronics15163615

BibTeX
@article{Li2026,
  author = {Li, Peize and Guo, Xi and Yin, Nan and Deng, Yiquan and Zhang, Lei and Liu, Jian and He, Jie},
  title = {Towards Building a Multi-Source Heterogeneous Knowledge Graph for Complex Material Question Answering},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {16},
  pages = {3615},
  doi = {10.3390/electronics15163615}
}
Kostenlos starten →