Towards a Framework that Allows Using a Cognitive Architecture to Personalize Recommendations in e-Commerce.

In this paper we propose the use of a cognitive architecture as a mechanism to improve the level of personalization in a recommendation process. The main idea is to use the cognitive architecture as an advisor that lies between the user and a set of specialized recommenders. The cognitive architecture should know about the strengths and weaknesses of the recommenders, when they can be used and their degree of reliability given the amount of information available, among other things. With this information and the detailed knowledge about the user needs, it should be able to provide much more reliable and trustful recommendations by combining the strengths of each recommender given the specific context. The flexibility of a cognitive architecture in this place of the recommendation process will allow also to consider a much more sophisticated interaction with the user by allowing for example argumentation processes between the user and the system.