摘要
随着煤矿安全监测领域信息化水平的不断提高,所产生的大量数据资源需要愈加科学的知识管理方法,因此提出一种基于改进规则挖掘和溯源搜索算法的煤矿安全监测知识库构建方法。首先,在构建领域词典的基础上,通过改进的词频-逆文档频率算法提取领域概念,使用Protégé工具表示煤矿安全监测本体模型;其次,提出一种添加主客观约束和规则评价的改进Apriori算法实现关联规则挖掘,利用Jena推理机进行安全监测数据的规则推理,完成溯源知识库构建;最后,引入限制深度和剪枝技术改进广度优先搜索算法,并在溯源知识库中检索排除错误路径的报警原因。结果表明:改进的领域概念提取算法F1-score提升了36.54%;改进的关联规则挖掘算法减少了75.72%的频繁项集;改进的溯源搜索算法分别减少了84.14%和84.39%的平均节点数和边数,搜索时间缩短了69.54%。研究结果有效地提高了概念提取、规则挖掘和节点检索3个环节的效率,为煤矿安全监测事故溯源提供了有力的支持。
As the information level in coal mine safety monitoring keeps improving,there is a growing need for advanced knowledge management methods to handle a large amount of data.This paper presents a method for building a knowledge base of coal mine safety monitoring with an enhanced rule mining and traceability search algorithm in view.First,by establishing a domain dictionary,an improved term frequency-inverse document frequency algorithm is used to extract domain concepts,and Protege represents the ontology model of coal mine safety monitoring.Then,an improved Apriori algorithm with subjective&objective constraints and rule evaluation is proposed for rule mining.The Jena inference engine deduces rules from security monitoring data and completes the construction of the traceability knowledge base.Finally,the breadth-first search algorithm is enhanced by introducing depth limits and pruning techniques to find alarm causes and exclude wrong paths.The results show that the improved domain concept extraction algorithm’s F1-score rises by 36.54%;the improved rule mining algorithm reduces frequent item sets by 75.72%;the improved traceability search algorithm decreases the average number of nodes and edges by 84.14%and 84.39%respectively,and shortens the search time by 69.54%.The proposed construction method of the coal mine safety monitoring knowledge base effectively improves the efficiency of concept extraction,rule mining,and node retrieval,providing a strong guidance for accident tracing in coal mine safety monitoring.
作者
潘红光
董佳瑶
白佩云
李山
马彪
PAN Hongguang;DONG Jiayao;BAI Peiyun;LI Shan;MA Biao(College of Electrical and Control Engineering,Xi’an University of Science and Technology,Xi’an 710054,China;Wulanmulun Coal,CHN Energy Shendong Coal Group,Ordos 017205,China;Guoneng Ordos Engineering Design,Co.,Ltd.,Ordos 017209,China)
出处
《西安科技大学学报》
北大核心
2025年第3期524-533,共10页
Journal of Xi’an University of Science and Technology
基金
国家重点研发计划项目(2023YFC3009800)
国家自然科学基金项目(61603295)
陕西省秦创原“科学家+工程师”队伍建设项目(2022KXJ-38)
陕西省教育厅服务地方专项计划项目(23JC049)。
关键词
安全监测系统
本体模型
数据挖掘
知识库
规则推理
security monitoring system
ontology model
data mining
knowledge base
rule reasoning
