EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

KGMP: Augmenting retrieval knowledge graph with multi-hop perceptron

Zhijie Yang iD, Liyuan Weng, Liang Zhang, Ruojia Tong, Jianyu Xie, Zhuo Zeng, Duanbing Chen iD

PLOS One · 06.10.2025 · 20 (10) · e0333037

DOI: 10.1371/journal.pone.0333037 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 06.10.2025. Verfasst von Zhijie Yang und 6 weiteren Autor:innen. Beteiligte Einrichtung: Big Data Research Center, University of Electronic Science and Technology of China, Chengdu, China. Hauptthemen laut MEDLINE: Neural Networks, Computer, Information Storage and Retrieval, Knowledge Bases. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 52 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Key Research and Development Program of Sichuan Province.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The core challenge of Knowledge Base Question Answering (KBQA), as a bridge between natural language and structured knowledge, is to accurately map complex semantic queries into Graph Query Language (GQL). Compared with the traditional Text-to-SQL task, KBQA faces a dual challenge: the structural differences between GQL and SQL and the lack of high-order subgraph information in multi-hop inference of knowledge graphs. While existing approaches such as ChatKBQA have made progress, the limitation of subgraph scalability severely constrains multi-hop query performance. To this end, this study proposes Knowledge Graph Multi-hop Perceptron (KGMP) - a retrieval-generation framework fine-tuned based on open-source large language models, whose innovativeness is reflected in three aspects: 1. Dynamic Graph Traversal Mechanism: Through an iterative subgraph expansion strategy, KGMP effectively achieves dynamic traversal of problem oriented graphs with progressive reasoning. 2. Structured Interaction Protocol: Based on SparQL syntax, KGMP designs a lightweight interaction instruction set to build an efficient communication interface between LLM and knowledge graph. 3. Graph Structure Optimization Technique: Develop subgraph reordering algorithms and pruning strategies based on the reranker model to ensure that the subgraphs input to the LLM are both compact and semantically complete. By integrating KGMP as a retrieval module into the ChatKBQA framework and providing it with optimised multi-hop subgraph input, the experimental results show a performance improvement of 6.2% and 5.3% on the WebQSP and CWQ datasets, respectively. This study provides a new technical paradigm for deep collaboration between LLM and knowledge graph.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Neural Networks, ComputerInformation Storage and RetrievalKnowledge BasesAlgorithmsSemanticsHumans

Autor:innen und Institutionen

  • Big Data Research Center, University of Electronic Science and Technology of China, Chengdu, China.
  • Chengdu Union Big Data Tech. Inc., Chengdu, China.
  • Cuiying Middle School, Shantou, China.
  • Suining Institute of Digital Economy, Suining, China.

ORCID-Profile: Zhijie Yang, Duanbing Chen

Förderung

  • Key Research and Development Program of Sichuan Province

Zitieren

APA

Yang, Z., Weng, L., Zhang, L., Tong, R., Xie, J., Zeng, Z., & Chen, D. (2025). KGMP: Augmenting retrieval knowledge graph with multi-hop perceptron. PLOS One, 20(10), e0333037. https://doi.org/10.1371/journal.pone.0333037

Vancouver

Yang Z, Weng L, Zhang L, Tong R, Xie J, Zeng Z, et al. KGMP: Augmenting retrieval knowledge graph with multi-hop perceptron. PLOS One. 2025;20(10):e0333037. doi:10.1371/journal.pone.0333037

BibTeX
@article{Yang2025,
  author = {Yang, Zhijie and Weng, Liyuan and Zhang, Liang and Tong, Ruojia and Xie, Jianyu and Zeng, Zhuo and Chen, Duanbing},
  title = {KGMP: Augmenting retrieval knowledge graph with multi-hop perceptron},
  journal = {PLOS One},
  year = {2025},
  volume = {20},
  number = {10},
  pages = {e0333037},
  doi = {10.1371/journal.pone.0333037}
}
Kostenlos starten →