Semantic Web Service Discovery Based on LDA Clustering

• Published in: Web Information Systems and Applications Vol. 11817; pp. 239 - 250
• Main Authors: Zhao, Heng, Chen, Jing, Xu, Lei
• Format: Book Chapter
• Language: English
• Published: Switzerland Springer International Publishing AG 2019; Springer International Publishing
• Series: Lecture Notes in Computer Science
• Subjects: Clustering; Document parsing; Latent Dirichlet Allocation; Semantic Web service; Web service discovery
• ISBN: 9783030309510; 3030309517
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/978-3-030-30952-7_25

In recent years, with the exponential growth of Web services, how to find the best Web services quickly, accurately and efficiently from the large Web services becomes an urgent problem in Web service discovery. Based on the previous work, we propose a semantic Web service discovery method based on LDA clustering. Firstly, the OWL-S Web service documents are parsed to obtain the document word vectors. Then these vectors are extended to make the documents more abundant of semantic information. Moreover, these vectors are modeled, trained and inferred to get the Document-Topic distribution, and the Web service documents are clustered. Finally, we search the Web service request records or the Web services clusters to find Web services that meet the requirements. Based on the data sets of OWLS-TC4 and hRESTS-TC3_release2, the experimental results show that our method (LDA plus semantic) has higher accuracy (13.48% and 9.97%), recall (37.39% and 24.26%), F-value (30.46% and 23.58%) when compared with VSM method and LDA method.