Towards an e-Science Environment for Collaborative Filtering Researchers

Towards an e-Science Environment for Collaborative Filtering Researchers

Nikos Manouselis (Agro-Know Technologies, Athens, Greece) and Giannis Stoitsis (Agro-Know Technologies, Athens, Greece)
Copyright: © 2014 |Pages: 32
DOI: 10.4018/ijdls.2014010104
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

A plethora of collaborative filtering algorithms have been proposed in related literature. Due to the dynamic and changing parameters of the various application contexts, careful testing and parameterization has to be carried out before an algorithm is finally deployed in a real setting. This paper investigates how a previously proposed tool for simulated testing of collaborative filtering algorithms, called the Collaborative Filtering Simulator (CollaFiS), can be expanded to an e-science environment for researchers so that it runs over a digital research infrastructure. More specifically, a survey of design options for neighborhood-based collaborative filtering systems is carried out in order to illustrate the variety of requirements that need to be met. Then, a number of usage scenarios that could be supported by an e-science environment for collaborative filtering research are presented. Three example dataset cases are used to illustrate how the new version of CollaFiS can support research and experimentation on collaborative filtering algorithms using different data and various parameter options. Overall, this paper showcases how e-science environments and infrastructures may facilitate the research activities of people working on recommender systems.
Article Preview

Background

More than two decades ago, Malone et al. (1987) provided an overview of intelligent information sharing systems, identifying a fundamental categorization of systems that support access to highly dynamic information resources (Belkin & Croft, 1992; Baudisch, 2001; Hanani et al., 2001). More specifically, they distinguished:

Complete Article List

Search this Journal:
Reset
Volume 5: 2 Issues (2015)
Volume 4: 2 Issues (2014)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing