An approach for domain ontology construction based on association rules and K-means
摘要:
随着网络上服务资源的规模化增长,如何帮助用户找到所需服务是一个关键问题.研究发现领域本体的构建可帮助用户有效解决该问题,而已有的一些构建方法一般依靠人工,费时费力.针对该问题,提出一种基于关联规则和K-means的领域本体构建方法.该方法首先利用支持向量机进行面向领域的服务分类,从分类得到的领域知识中选取初始领域概念;然后根据关联规则和K-means算法挖掘概念间关系,以得到初始领域本体,并使用Wordnet对其进行语义丰富.最后,通过ProgrammableWeb网站提供的真实服务集进行实验验证.实验结果表明所提出的领域本体构建方法能够识别概念间关系,进而为Web服务语义查询提供相应支持.
With the scale growth of service resources on the network,how to help users find their required services is a key issue.Studies found that the construction of domain ontology can help users effectively solve the problem,but some of existing methods are built manually,which is time-consuming and laborious.In order to solve this problem,this paper proposed an ontology construction method based on association rules and K-means.Firstly,we use support vector machine to conduct domain-oriented services classification,and select initial domain concepts from the domain knowledge obtained by classification.Then according to association rules and K-means,we mine the relationships among concepts to obtain the initial domain ontology,and the obtained ontology is further enriched by Wordnet.Finally,the real services from ProgrammableWeb are used to conduct experiments.The experimental results show that the proposed domain ontology construction approach can identify the relationships among concepts,and then provide the corresponding support for Web services semantic query.
作者:
李征 李斌
Li Zheng;Li Bin(School of Computer and Information Engineering,Henan University,Kaifeng 475004,China;Key Laboratory of Intelligent Vision Monitoring for Hydropower Project of Hubei Province,China Three Gorges University,Yichang 443002,China)
机构地区:
河南大学计算机与信息工程学院 三峡大学湖北省水电工程智能视觉监测重点实验室
出处:
《betway官方app 学报:自然科学版》 CAS 北大核心 2020年第1期24-32,共9页
基金:
国家重点基础研究发展计划(973)(2014CB340404) 国家自然科学基金(61402150) 中国博士后科学基金资助项目(2016M592286) 河南省科技研发专项资助项目(182102410063) 三峡大学水电工程智能视觉监测湖北省重点实验室开放基金(2016KLA04).
关键词:
服务分类 关联规则 K-MEANS 领域本体
service classification association rule K-means domain ontology
分类号:
TP391 [自动化与计算机技术—计算机应用技术]