CONCLUSION


In this chapter, we have described a model-based knowledge acquisition tool for user profiling for electronic commerce applications. The tool aims to reduce the burden on the user's side while providing a sense of control and trust. The tool is based on a selected user model and is agent-mediated.

Based on the customer's shopping behavior, the user's personalization needs are identified, and an appropriate user model is described. The user model presented in this chapter consists of a directed acyclic graph of PIEs (Preference Indication by Example). This model is motivated by the perceived need to broaden the coverage of the domain of products while dealing with a virtual or electronic shopping mall.

Knowledge about the consumer is acquired using different techniques, ranging from fill-in forms and dialogue to the observation of user actions and machine learning. Analysis of user data is done by the processing Agent A1 and by the Web log mining module. The validation Agent A2 deals with conflict resolution and interacts with the user via dialogues .

Our tool is domain-dependent. If the domain is changed, the ontology and other related data have to be changed, but the overall structure of the system remains the same. The tool allows dynamic user profiling that goes through a constant monitoring, validation, and upgrade cycle. Our ongoing research is still focusing on this last issue.




(ed.) Intelligent Agents for Data Mining and Information Retrieval
(ed.) Intelligent Agents for Data Mining and Information Retrieval
ISBN: N/A
EAN: N/A
Year: 2004
Pages: 171

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net