KI-03-2002 mit web-Hinweisen.p65

In ML4UM user models are often represented by n–ary vectors. In the course of content-based document recommendation, the vectors represent significance of “key”–phrases for the user’s interest which is defined as the frequency (TFIDF and similar) of those phrases in the documents the user has rated (‘relevance feedback’) as interesting in the past. In order to decide whether to recommend a document or not, the document (i.e. the corresponding vector) is compared to the user model vector, see [2,1,9], [5] and [6]. Though the vectors represent a user’s interest, they do not explicitely describe a user’s interest: A sequence of word frequencies is not a user model that can be explained to the user in an intutive way. Thus, our motivation was to find a transparent formalism which is accessible to every user and which allows for an easy translation into the user’s ‘language’. Such a language could be the language of concept hierarchies. The choice of such an approach is supported by the fact, that humans in general (including unexperienced users in the web) are capable of describing and finding objects using concept hierarchies such as library taxonomies, classification systems or web directories. As a result, we obtain a user adaptive system, which is capable of describing internal user models and which can explain recommendations to the user, thereby increasing user acceptance because the user is able to map elements of taxonomies onto meaningful concepts.