User Modeling and The Adaptive Semantic Web

Submitted by Pascal Hitzler on 05/06/2010 - 08:15
Paper Title: 
User Modeling and The Adaptive Semantic Web
Authors: 
Lora Aroyo, Geert-Jan Houben
Abstract: 
Historically, personalization and adaptation have been important factors for the success of the Web and therefore they have been important topics in Web research. Many research efforts in adaptive hypermedia and adaptive Web-based systems have resulted in solutions for user-adapted access to Web content, often in terms of systems that provide an adaptive hypermedia structure of content pages and hyperlinks. With pages and links that depend on the user, it is feasible to offer a high degree of personalization. For example, in educational, commercial or entertainment scenarios this kind of personalization has often featured in adaptation research and tool development. Next to research into engineering and realizing adaptation, research into user modeling has been crucial for the success of adaptation. To apply the right adaptation it is necessary to know the user and her relevant properties and the research field of user modeling has focused on theories and techniques for eliciting knowledge about the user. Naturally, the research fields of adaptive web-based systems and user modeling have always lived in close harmony. In order to create a similar success with personalization and adaptation on the Semantic Web, adaptation and user modeling have to be redefined, with consequences for the research into these topics. In particular, the nature of user modeling changes significantly with the extended distribution and openness that we encounter on the Web of Data, with implications from problems studied in Web science. In this vision paper we outline the evolution of user modeling and adaptation in the Semantic Web and list research questions and challenges for the relevant research fields.
Full PDF Version: 
PDF icon swj23.pdf
Submission type: 
Other
Responsible editor: 
Krzysztof Janowicz
Decision/Status: 
Accept
Reviews: 

Review 1 by Michel Dumontier:

This paper discusses the nature of user-adaptation and user modeling with a forward thinking projection to the Semantic Web. Of note is how user identification and properties could benefit from having a shared conceptualization, such that adaptive applications could make use of these in such a way that users benefit from this information. While cataloging the user's background knowledge will likely prove challenging, so will aspects of openness and scalability.

My underlying concern with this vision paper is that it provides little in terms of specifics with respect to linked data, ontologies and reasoning-capable knowledge bases. What aspects of personalization or adaptation seem particularly salient in the context of a global knowledge base? I would encourage the authors to promote some promising work in this direction.

Review 2 by Jie Tang:
The paper review the relevant research on user modeling and the adaptive semantic web. The paper is generally well written and the most important issues have been introduced and challenges in this field are discussed.

I have the following suggestions:
1. The concept of user-adaptation is not clearly introduced. Although, I can guess what it means, it would be better to give an explicit definition or an informal explanation.
2. What is the different of user-adaptation on the Web and that on the Semantic Web? The author introduces a lot about user-adaptation and user modeling in section 2; however, how they are related to semantic web is still unclear.
3. Section 2, 3, 4 are not organized very well. After reading, a general feeling is that the three sections are independent. The correlation between them and why the three problems should be addressed under the vision of user modeling and user adaptation is not clear.

Tags: 
Reviewed