ABSTRACT
Personalization in e-commerce has so far been server-centric, requiring users to create a separate individual profile on each server that they like to access. As product information is increasingly coming from multiple and heterogeneous sources, the number of profiles becomes unmanageably large. We present SmartClient, a technology based on constraint programming where a thin but intelligent client provides personalized information access for its user. As the process can run on the user's side, it allows much stronger filtering and visualization support with a wider range of personalization options than existing tools. It also eliminates the need to personalize many sites individually with different parameters, and supports product configuration and integration of different information sources in the same framework. We illustrate the technology using an application in travel e-commerce, which is currently under commercial deployment.
- 1.Burke, D., Hammond, K.J., and Young, B.C. The FindMe Approach to Assisted Browsing. IEEE Expert, pp. 32-40, July-August 1997. Google ScholarDigital Library
- 2.Excite Shopping. Available at: http://www.jango.com/. Accessed September 27, 2001.Google Scholar
- 3.Expedia. Available at: http://www.expedia.com/.Google Scholar
- 4.Lieberman, H. Autonomous Interface Agents. In Proceedings of CHI'97, March 1997. Google ScholarDigital Library
- 5.Shearin, S. and Lieberman, H. Intelligent Profiling by Example, in Proceedings of Conference on Intelligent User Interfaces, ACM Press, 2001. Google ScholarDigital Library
- 6.Pu, P. and Faltings, B. Enriching buyers' experiences: the SmartClient approach, in Proceedings of ACM CHI'2000, ACM Press, 2000. Google ScholarDigital Library
- 7.Pu, P. and B. Faltings (2001). Contextual Maps for Personalized Navigation of Electronic Catalogs. Technical Report, Swiss Federal Institute of Technology (EPFL), Lausanne, Switzerland.Google Scholar
- 8.M. Torrens, B. Faltings, and P. Pu. (2001). "SmartClient: Constraint Satisfaction as a Paradigm for Scalable Intelligent Information Systems." Constraint Journal (Special Issue on Constraint Agents), in press. Google ScholarDigital Library
- 9.Travelocity. Available at: http://www.travelocity.com/Google Scholar
Index Terms
- Personalized navigation of heterogeneous product spaces using SmartClient
Recommendations
Integrating collaborative filtering and matching-based search for product recommendations
Currently, recommender systems (RS) have been widely applied in many commercial e-commerce sites to help users deal with the information overload problem. Recommender systems provide personalized recommendations to users and, thus, help in making good ...
Effects of Personalized and Aggregate Top-N Recommendation Lists on User Preference Ratings
Prior research has shown a robust effect of personalized product recommendations on user preference judgments for items. Specifically, the display of system-predicted preference ratings as item recommendations has been shown in multiple studies to bias ...
Learning to Generate Personalized Product Descriptions
CIKM '19: Proceedings of the 28th ACM International Conference on Information and Knowledge ManagementPersonalization plays a key role in electronic commerce, adjusting the products presented to users through search and recommendations according to their personality and tastes. Current personalization efforts focus on the adaptation of product ...
Comments