Interpretable multi-hop knowledge reasoning for gastrointestinal disease

Dujuan Wang, Xinwei Wang, Mohammad Zoynul Abedin, Sutong Wang () and Yunqiang Yin
Additional contact information
Dujuan Wang: Sichuan University
Xinwei Wang: Sichuan University
Mohammad Zoynul Abedin: Swansea University, Bay Campus
Sutong Wang: Sichuan University
Yunqiang Yin: University of Electronic Science and Technology of China

Annals of Operations Research, 2025, vol. 347, issue 2, No 8, 959-990

Abstract: Abstract Gastrointestinal diseases cause a large number of outpatient treatments every year and cost a lot of medical and health budgets. This continuous rise in healthcare costs, along with the need for improved diagnostic and treatment strategies, has encouraged researchers to investigate advanced methods to address the issue. In this paper, we propose an Interpretable Multi-hop Knowledge Reasoning (IMKR) method for gastrointestinal diseases and design an intelligent question-answering system framework. Our proposed method is rooted in the Dual Process Theory (DPT), which argues that reasoning involves two processes, including intuitive and fast process, and reflective and comprehensive process. Drawing inspiration from DPT, we use the Tensorlog operator and recursive neural network to generate logic rules corresponding to the first process of DPT, and further introduce cognitive expansion and subgraph reasoning to enhance the performance of IMKR, which achieves the second process. Moreover, to better evaluate the predictions of one-to-many relationships in the medical context, we also propose a new model evaluation algorithm. Finally, we conduct experiments on the healthcare platform dataset and the proposed method achieves the performance of hit rate of ranking 1 (Hits@1) 0.9542, hit rate of ranking 10 (Hits@10) 0.9995, and Mean Reciprocal Ranking (MRR) 0.6976.

Keywords: Gastrointestinal disease; Knowledge graph; Multi-hop knowledge reasoning; Intelligent question answering (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s10479-023-05650-6 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:annopr:v:347:y:2025:i:2:d:10.1007_s10479-023-05650-6

Ordering information: This journal article can be ordered from
http://www.springer.com/journal/10479

DOI: 10.1007/s10479-023-05650-6

Access Statistics for this article

Annals of Operations Research is currently edited by Endre Boros

More articles in Annals of Operations Research from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().