The current Web service architecture addresses the service discovery problem, but not service selection. If we treat services as a special type of products from the service providers, existing techniques used for product selection can be applied to service selection. Among many existing techniques for product selection, one dominant approach is to use recommender systems. For example many ecommerce Web site, such as Ebay, Amazon and Epinions all have recommender system support to ease the burden of product selection. Recommender systems are classified into different types, which include content-based, collaborative and hybrid based systems which perform recommendation based on user/item data set. There is also multidimensional recommender system where the system supports multiple dimensions, such as user,item, and its quality attributes. This research investigates the fundamentals of different types of recommender systems, in particular, the collaborative filtering based approach using both two dimensional and multidimensional data. This study is to facilitate the development of a framework that supports service selection. In this project we are interested in applying a multidimensional collaborative recommender system for Web service selection. But before we propose the framework, we first perform experiments on the similarity measures used in collaborative filtering to fully understand the working process of normal collaborative filtering system. Pearson correlation and Vector Similarity are tested and compared using the MovieLens data set. This experiment is important to decide which of the similarity measures performs better in a two dimensional data set. Because the Quality of Service is often characterized by multiple criteria, different people/users may put stronger emphasis on certain criteria than others, two dimensional approach is over-simplified. Therefore we looked at the multi dimensional approach on product selection as well. An experiment using the multidimensional data collected in this research from Epinions.com is performed to study how it works on product selection. Finally, we draw from the experience of both two dimensional and multidimensional product selection and propose a framework for Web service selection which uses the multidimensional recommender system approach.
[1]
E. Michael Maximilien,et al.
Toward autonomic web services trust and selection
,
2004,
ICSOC '04.
[2]
David Heckerman,et al.
Empirical Analysis of Predictive Algorithms for Collaborative Filtering
,
1998,
UAI.
[3]
John Riedl,et al.
GroupLens: an open architecture for collaborative filtering of netnews
,
1994,
CSCW '94.
[4]
John Riedl,et al.
Item-based collaborative filtering recommendation algorithms
,
2001,
WWW '01.
[5]
Munindar P. Singh,et al.
Agent-based service selection
,
2004,
J. Web Semant..
[6]
Daniel A. Menascé,et al.
QoS Issues in Web Services
,
2002,
IEEE Internet Comput..
[7]
Mark Claypool,et al.
Combining Content-Based and Collaborative Filters in an Online Newspaper
,
1999,
SIGIR 1999.
[8]
Matthew MacDonald,et al.
Web Services Architecture
,
2004
.
[9]
Hsinchun Chen,et al.
Applying associative retrieval techniques to alleviate the sparsity problem in collaborative filtering
,
2004,
TOIS.
[10]
E. Michael Maximilien,et al.
A framework and ontology for dynamic Web services selection
,
2004,
IEEE Internet Computing.
[11]
Gediminas Adomavicius,et al.
Incorporating contextual information in recommender systems using a multidimensional approach
,
2005,
TOIS.