- Citations Citations 11
- References References 11
- Then, similarity of services could be computed being an aggregate of the baby SoFs values. Current clustering approaches use several similarity calculation methods for example keyword-Match (KM) , information-retrieval (IR) techniques  and ontology based techniques . However, KM might not precisely find out the semantic similarity among terms due to the heterogeneity and independence and services information sources.
[Show abstract] [Hide abstract] ABSTRACT: Clustering Web services into functionally similar clusters is an extremely efficient method of service discovery. A principal problem for clustering is computing the semantic similarity between services. Current approaches use similarity-distance measurement methods for example keyword, information-retrieval or ontology based methods. These approaches have issues which include finding semantic characteristics, lack of semantic information and lack of high-quality ontologies. Further, current clustering approaches are thought have only simple data types in services’ input and output. However, services that printed on the internet have input/ output parameter of complex data type. Within this research, we advise clustering approach that views the straightforward type in addition to complex data key in calculating the service similarity. We use hybrid term similarity method which we suggested within our previous try to appraise the similarity. We capture the semantic pattern appear in complex data types and straightforward data types to enhance the ontology learning method. Experimental results show our clustering approach which utilizes complex data types in calculating similarity works efficiently.
Conference Paper · 12 , 2014
Samantha Kumara Incheon Paik Kowatte R. C. Koswatte Wuhui Chen
- Verified by actual test atmosphere, this process can shorten the service discovery response time on condition of not affecting the service discovery precision, therefore improving the performance of unmixed semantic UDDI. In , Web Service ontology is determined. They will use Web Service function and progress to calculate similarity.
[Show abstract] [Hide abstract] ABSTRACT: Web Services are independent software systems that offer machine-to-machine interactions on the internet to attain well-described operations. Using the creation of Service-Oriented Architecture (SOA), Web Services have acquired tremendous recognition. As the amount of Web Services is elevated, finding the right service based on users needs turns into a challenge. The Semantic Web Service discovery is the procedure to find probably the most appropriate service that satisfies the consumer request. Numerous methods to Web Service discovery happen to be suggested. Within this paper, we classify them and see the pros and cons of every group, to assist researchers to apply a brand new or to decide on the most suitable existing method for Semantic Web Service discovery. We, also, give a taxonomy which categorizes Web Service discovery systems from various perspectives. You will find three different views, namely, architectural view, automation view and matchmaking view.
We concentrate on the matchmaking view that is further split into semantic-based, syntax-based and context-aware. We explain each subgroup from it at length, after which subsequently compare the subgroups when it comes to their merits and downsides.
Full-text · Article · February 2014
Soodeh Pakari Esmaeel Kheirkhah Mehrdad Jalali
[Show abstract] [Hide abstract] ABSTRACT: Using the rapid development of Web services and the necessity of rapidly choosing the best services, instantly clustering Web services becomes exceedingly important and challenging. The primary disadvantage to current clustering algorithms is must get total services or perhaps a service training set to mine the service attribute. This doesn’t suit for that dynamic distributed atmosphere. To be able to address this issue, this paper first raises the idea of service ability and provides a means to compute the service ability. By way of ontology, this paper presents a manuscript service clustering formula according to service ability. Without priori understanding or similarity computation between any two services, this clustering formula can cluster the help concentrating on the same function and repair ability. Facilitated with this formula, something searching formula continues to be suggested. Experimental and theoretical results reveal that the clustering formula can reflect the help' function clustering feature effectively and looking out formula can filter lots of unrelated service.
Conference Paper · November 2012