Research On Ontology-assisted Association Rule Mining Model For OGC Web Service Searching

With the rapid development of computer and internet technology,the frequency of using geographic information services is growing exponentially,and gradually massive geographic information service data is described in the form of sharing.It is very significant to find out the useful geographic information service data from the massive data containing huge semantic information.This thesis researches on ontology-assisted OGC geographic information service association rules model construction method taking OGC geographic information service as the research object,thus the association rules among OGC geographic information service data are mined and implemented.The experiments show that the algorithm combining ontology and traditional association rules can not only improve the speed of mining,but also realize the knowledge discovery of conceptual level.The main work of this thesis is as follows:(1)A geology mineral field ontology base is established for OGC geographic information services.Through collecting OGC geographic information service data and using DOM parsing technology,the topic vocabularies are extracted from the service metadata files.Then a geology mineral field ontology base is constructed through analyzing the topic vocabularies and combining with the framework of ontology construction process.(2)An ontology-assisted association rule model is proposed for geographic information service.In view of the shortcomings of association rules mining,this thesis takes the geology mineral field ontology to guide the association rules mining which can as a replacement for domain expert.The ontology-assisted association rules mining model is implemented for OGC geographic information services.(3)A prototype system based on ontology-assisted association rule mining model is developed.A recommendation system for OGC geographic information service is designed and developed by using the ontology-assisted Apriori algorithm model as the core algorithm of the system and combining open source tools such as MySQL,SpringMVC and others.A contrast experiment between the prototype system of ontology-assisted association rules mining and the traditional association rule mining algorithm system is conducted.The results show that: 1.In the case of same configuration environment,the operation efficiency of ontology-assisted association rule mining recommendation model is at least 2 seconds higher than the traditional model.2.The ontology-assisted association rule mining recommendation system ensures the recall rate is increased by 8%,and the precision rate is increased by 2% compared with the traditional association rule mining recommendation system.