Paper Info
Title
Using Relational Topic Model and Factorization Machines to Recommend Web APIs for Mashup Creation.
Abstract
The rapid growth in the number of Web APIs, coupled with the myriad of functionally similar Web APIs, makes it difficult to find suitable Web APIs to develop Mashup applications. Even if the existing Web APIs recommendation methods show improvements in service discovery, the accuracy of them can be significantly improved due to overlooking the impact of sparsity and dimension of relationships between Mashup and Web APIs on recommendation accuracy. In this paper, we propose a Web APIs recommendation method for Mashup creation by combining relational topic model and factorization machines technique. This method firstly uses relational topic model to characterize the relationships among Mashup, Web APIs, and their links, and mine the latent topics derived by the relationships. Secondly, it exploits factorization machines to train the latent topics for predicting the link relationship among Mashup and Web APIs to recommend adequate relevant top-k Web APIs for target Mashup creation. Finally, we conduct a comprehensive evaluation to measure performance of our method. Compared with other existing recommendation approaches, experimental results show that our approach achieves a significant improvement in terms of precision, recall, and F-measure.
Year
DOI
Venue
2016
10.1007/978-3-319-49178-3_30
ADVANCES IN SERVICES COMPUTING
Keywords
Field
DocType
Relational topic model,Factorization machines,Web APIs recommendation,Mashup creation,Service discovery
Web API,Mashup,World Wide Web,Computer science,Exploit,Factorization,Topic model,Service discovery
Conference
Volume
ISSN
Citations 
10065
0302-9743
1
PageRank 
References 
Authors
0.36
17
5
Name
Order
Citations
PageRank
Buqing Cao120023.96
Min Shi2353.53
Xiaoqing (Frank) Liu335427.76
Jianxun Liu464067.12
Tang Mingdong555739.35