{"title":"An auxiliary decision-support model for landslide treatment by integrating knowledge graph and case-based reasoning","authors":"Fei Ding, Daichao Li, Xinlei Jin, Minjiang Liu, Xiaohui Wang, Yuan Li","doi":"10.1007/s10064-025-04536-0","DOIUrl":null,"url":null,"abstract":"<div><p>The primary task in reducing the risk and hazards of landslides is to quickly identify and implement effective treatment measures. Currently, case-based reasoning (CBR) has shown significant results in landslide treatment decision-support applications, but this method still faces issues such as weak geographic spatial characteristics representation, unreasonable weight assignment, and interference from pseudo-similar cases. In response, this study proposes an auxiliary decision-support model for landslide treatment by integrating knowledge graph and case-based reasoning (KGCBR). First, a landslide treatment knowledge graph (KG) is constructed, and the TransH knowledge embedding model is used to obtain the geographical similarity of landslide attributes, which is then integrated into case representation to enhance its geographical spatial characteristics representation capability. Second, leveraging the grey wolf optimizer (GWO), an adaptive weight optimization assignment method is devised to obtain the optimal weights of landslide attributes, and the k-nearest neighbors (KNN) algorithm is introduced to retrieve k similar historical cases. Finally, the treatment measures of similar historical cases for different values of k are statistically analyzed and filtered to correct the recommended measures and eliminate the randomness of the results. Experimental results show that the model achieves a minimum recommendation error rate of 16.23%, lower than the case-based reasoning methods based on averaging weighting (24.42%) and entropy weighting (24.00%). It also demonstrates high engineering rationality and reliability in analyzing engineering treatment cases. Overall, the model can recommend widely applicable landslide treatment measures to decision-makers, reducing decision-making time costs and uncertainty, and meeting practical application needs.</p></div>","PeriodicalId":500,"journal":{"name":"Bulletin of Engineering Geology and the Environment","volume":"84 11","pages":""},"PeriodicalIF":4.2000,"publicationDate":"2025-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Bulletin of Engineering Geology and the Environment","FirstCategoryId":"5","ListUrlMain":"https://link.springer.com/article/10.1007/s10064-025-04536-0","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENGINEERING, ENVIRONMENTAL","Score":null,"Total":0}
引用次数: 0
Abstract
The primary task in reducing the risk and hazards of landslides is to quickly identify and implement effective treatment measures. Currently, case-based reasoning (CBR) has shown significant results in landslide treatment decision-support applications, but this method still faces issues such as weak geographic spatial characteristics representation, unreasonable weight assignment, and interference from pseudo-similar cases. In response, this study proposes an auxiliary decision-support model for landslide treatment by integrating knowledge graph and case-based reasoning (KGCBR). First, a landslide treatment knowledge graph (KG) is constructed, and the TransH knowledge embedding model is used to obtain the geographical similarity of landslide attributes, which is then integrated into case representation to enhance its geographical spatial characteristics representation capability. Second, leveraging the grey wolf optimizer (GWO), an adaptive weight optimization assignment method is devised to obtain the optimal weights of landslide attributes, and the k-nearest neighbors (KNN) algorithm is introduced to retrieve k similar historical cases. Finally, the treatment measures of similar historical cases for different values of k are statistically analyzed and filtered to correct the recommended measures and eliminate the randomness of the results. Experimental results show that the model achieves a minimum recommendation error rate of 16.23%, lower than the case-based reasoning methods based on averaging weighting (24.42%) and entropy weighting (24.00%). It also demonstrates high engineering rationality and reliability in analyzing engineering treatment cases. Overall, the model can recommend widely applicable landslide treatment measures to decision-makers, reducing decision-making time costs and uncertainty, and meeting practical application needs.
期刊介绍：
Engineering geology is defined in the statutes of the IAEG as the science devoted to the investigation, study and solution of engineering and environmental problems which may arise as the result of the interaction between geology and the works or activities of man, as well as of the prediction of and development of measures for the prevention or remediation of geological hazards. Engineering geology embraces:
• the applications/implications of the geomorphology, structural geology, and hydrogeological conditions of geological formations;
• the characterisation of the mineralogical, physico-geomechanical, chemical and hydraulic properties of all earth materials involved in construction, resource recovery and environmental change;
• the assessment of the mechanical and hydrological behaviour of soil and rock masses;
• the prediction of changes to the above properties with time;
• the determination of the parameters to be considered in the stability analysis of engineering works and earth masses.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
