Use of Buying Behavioural Characteristics as General Stereotypes in User Models

In the literature, long-term user models are used to provide personalized interactions in ecommerce web sites. The growing numbers of online users and massive number of different products, demand not only user models, but also well-organized product information in order to find the products that an individual may like. Although such knowledge engineering is a time consuming task, its reusability reduces this disadvantage. Although products are relatively easy to model using their attributes, due to the complexity of human nature, different types of information about users are required to capture their needs and preferences. In this paper we introduce a novel user model architecture, with the ability to store such user information in layers. We demonstrate the information reuse within a user model and highlight the importance of using buying behaviour characteristics to understand the user.