Fuzzy Decision Tree for User Modeling from Human-Computer Interactions

Providing helpful analysis of cognitive process is one of the main concerns for user modeling and adaptive hypermedia conception. In this paper, a work is presented that looks very closely to the Human Computer Interactions and exploits these interactions to help novice computer users. We think that all the cognitive actions involved in a task resolution reflect the level of expertise that the user has with the interface [1]. In this paper, we describe an application that learns these cognitive processes by fuzzy decision tree technique and discriminates novice from experimented user automatically. Once the discrimination is done, the system can provide an advice to the novice computer user with a contextual help.