张可,张政,金伟

• 1:河海大学商学院
• 2:河海大学项目管理研究所
• 3:杭州市南排工程建设管理服务中心

摘要(Abstract)：

【目的】已有的数据驱动的水利工程建设安全风险预测方法对领域知识的挖掘和利用不足，预测结果的准确性和可解释性有待进一步提高。为了构建数据-知识融合的水利工程建设安全风险预测模型，【方法】将灰色聚类与因子分解机相结合，提出了一种融合领域知识的灰色因子分解机。首先，引入基于可能度函数的灰色聚类表征水利工程建设领域专家有关安全风险的先验知识。然后，将先验知识以参数的形式嵌入到因子分解机中，构建出数据-知识融合的灰色因子分解机。最后，基于随机梯度下降构造模型参数的求解算法，并结合实例对模型的有效性进行验证。【结果】实例应用结果显示，与传统因子分解机相比，灰色因子分解机的准确率、精确率、召回率和F_1值均得到了不同程度的提升。与支持向量机、深度因子分解机等其他基准模型相比，灰色因子分解机同样具有更好的预测性能。【结论】这表明，数据-知识融合驱动的灰色因子分解机模型能够更加准确地预测出安全风险，从而为水利工程建设安全风险管控提供更好的决策支持。

关键词(KeyWords)： 因子分解机;风险交互;领域知识;可能度函数;灰色聚类;影响因素

Abstract:

Keywords:

基金项目(Foundation): 国家社会科学基金项目(17BGL156);; 江苏省建设科技项目(521021012);; 河海大学中央高校基本科研业务费项目(B220207039)

作者(Author): 张可,张政,金伟

DOI: 10.13928/j.cnki.wrahe.2024.01.012

参考文献(References)：

• [1] 李国英.深入贯彻落实党的二十大精神扎实推动新阶段水利高质量发展——在2023年全国水利工作会议上的讲话[J].水利发展研究，2023,23(1):1-11.LI Guoying.Deeply implementing the spirit of the 20th national congress of the communist party of china and solidly promoting the high quality development of water conservancy in the new stage-Speech at the 2023 National Water Conservancy Work Conference [J].Water Resources Development Research,2023,23 (1):1-11.
• [2] 杜婷，宋艳红，李智莹，等.基于模糊综合评价法的建筑项目施工安全评价[J].土木工程与管理学报，2019,36(6):61-66,78.DU Ting,SONG Yanhong,LI Zhiying,et al.Safety appraisal of construction project based on Fuzzy Comprehensive Evaluation[J].Journal of Civil Engineering and Management,2019,36(6):61-66,78.
• [3] ANSARI R,DEHGHANI P,MAHDIKHANI M,et al.A novel safety risk assessment based on Fuzzy Set Theory and Decision methods in high-rise buildings[J].Buildings,2022,12(12):2126.
• [4] 袁剑波，崔钢，符秋生，等.基于网络分析法的公路桥梁施工安全风险评价研究[J].科技进步与对策，2014,31(11):96-100.YUAN Jianbo,CUI Gang,FU Qiusheng,et al.Research on safety risk assessment of highway bridge construction based on Analytic Network Process [J].Science & Technology Progress and Policy,2014,31(11):96-100.
• [5] YAN H Y,GAO C,ELZARKA H,et al.Risk assessment for construction of urban rail transit projects[J].Safety Science,2019,118:583-594.
• [6] 兰博，关许为，肖庆华.基于FAHP与熵权融合法的堤防工程安全综合评价[J].中国农村水利水电，2019(6):131-133,137.LAN Bo,GUAN Xuwei,XIAO Qinghua.A comprehensive evaluation of dike engineering safety based on fusion method of FAHP and Entropy[J].China Rural Water and Hydropower,2019(6):131-133,137.
• [7] 黄黎明，朱军，张可.水利工程建设质量多阶段风险评价研究[J].水利水电技术(中英文),2017,48(9):117-125.HUANG Liming,ZHU Jun,ZHANG Ke.Study on multi-stage risk assessment of construction quality of water conservancy project[J].Water Resources and Hydropower Engineering,2017,48(9):117-125.
• [8] CHUA D K H,GOH Y M.Poisson model of construction incident occurrence[J].Journal of Construction Engineering and Management,2005,131(6):715-722.
• [9] LOVE P E D,TEO P.Statistical analysis of injury and nonconformance frequencies in construction:Negative Binomial Regression model[J].Journal of Construction Engineering and Management,2017,143(8):05017011.
• [10] GUO S Y,HE J L,LI J C,et al.Exploring the impact of unsafe behaviors on building construction accidents using a Bayesian Network[J].International Journal of Environmental Research and Public Health,2020,17(1):221.
• [11] ZHOU Y,LI S Q,ZHOU C,et al.Intelligent approach based on Random Forest for safety risk prediction of deep foundation pit in subway stations[J].Journal of Computing Civil Engineering,2019,33(1):05018004.
• [12] 王秀杰，孙瑀，苑希民，等.突变理论与BP神经网络相结合的堤防安全综合评价[J].水利水电技术(中英文),2018,49(7):167-173.WANG Xiujie,SUN Yu,YUAN Ximin,et al.Catastrophe Theory and BP Neural Network-jointed comprehensive evaluation on levee safety[J].Water Resources and Hydropower Engineering,2018,49(7):167-173.
• [13] LIU P,XIE M C,BIAN J,et al.A hybrid PSO-SVM model based on safety risk prediction for the design process in metro station construction[J].International Journal of Environmental Research and Public Health,2020,17(5):1714.
• [14] 冯继伟，孙开畅.基于因子分解机的水利工程事件风险状态预测[J].水利水电技术(中英文),2021,52(12):178-184.FENG Jiwei,SUN Kaichang,et al.Factorization Machine-based prediction of safety event risk status of water conservancy project[J].Water Resources and Hydropower Engineering,2021,52(12):178-184.
• [15] 尚宇炜，马钊，彭晨阳，等.内嵌专业知识和经验的机器学习方法探索(一):引导学习的应用与实践[J].中国电机工程学报，2017,37(20):5852-5861.SHANG Yuwei,MA Zhao,PENG Chenyang,et al.Study of a novel machine learning method embedding expertise Part I:Proposals and fundamentals of Guiding Learning[J].Proceedings of the CSEE,2017,37(20):5852-5861.
• [16] ZHOU L G,LU D,FUJITA H.The performance of corporate financial distress prediction models with features selection guided by domain knowledge and data mining approaches[J].Knowledge-Based Systems,2015,85:52-61.
• [17] 孙亚茹，杨莹，王永剑.基于知信图卷积神经网络的开放域知识图谱自动构建模型[J].计算机工程，2022,48(10):116-122.SUN Yaru,YANG Ying,WANG Yongjian.Knowledge graph automatic construction model in open domain based on knowledge-informed Graph Convolutional Neural Network[J].Computer Engineering,2022,48(10):116-122.
• [18] RENDLE S.Factorization Machines[C]//WEI F.ICDMW:The 10th IEEE International Conference on Data Mining Workshops.Sydney,NSW,Australia:IEEE Computer Society Conference Publishing,2010:995-1000.
• [19] YU Y H,JIAO L H,ZHOU N N,et al,Enhanced factorization machine via neural pairwise ranking and attention networks[J].Pattern Recognition Letters,2021,140:348-357.
• [20] YAN C R,CHEN Y Z,WAN Y Q,et al.Modeling low-and high-order feature interactions with FM and Self-Attention Network[J].Applied Intelligence,2020,51(6):3189-3201.
• [21] 刘思峰.灰色系统理论及其应用：第9版[M].北京：科学出版社，2021.LIU Sifeng.Grey system theory and its applications:9th Edition.Beijing:Science Press,2021.
• [22] 中华人民共和国水利部.水利部关于开展水利安全风险分级管控的指导意见[Z].北京：水利部，2018.Minstry of Water Resources of the People′s Republic.China Guiding Opinions and Policy Document of the Ministry of Water Resources on Carrying out Water Safety Risk Grading and Control[Z].Beijing:Ministry of Water Resources of the People′s Republic of China,2018.
• [23] LIU S F,YANG Y.Explanation of terms of Grey Clustering evaluation models[J].Grey Systems:Theory and Application,2017,7(1):129-135.