Adaptive search results personalized by a fuzzy recommendation approach

Abstract

The cognitive load of the driver in future vehicles will be increased by a larger number of integrated or connected devices and by a higher complexity of applications covering various vehicle-related tasks and other information. Aspects of perception relevant to interaction design of adaptive and multimodal interfaces are discussed. We propose an extension of a system architecture for user-adaptive information presentation in order to cope with high information load in a multimodal interface. Visual information and speech are used as output channels, and manual input and speech as input channels. In the system, current information from the Internet can be queried by natural language question-answering dialogs using frame-based dialog management. A fuzzy recommendation approach is described in detail and used in the system for structuring content items according to the user preferences. In addition to such a personalized visual list of topic areas for an initial choice, an adaptive ranking is applied to acoustical speech output. The item that relates best to the driver's preferences is read out first by text-to-speech. Therefore, the first items presented to the user are ought to be of high interest. Cognitive load can potentially be decreased by our approach for the task of choosing one item out of a sequence of alternatives in working memory. The choice can be immediately performed by using the barge-in feature. © 2008 Springer-Verlag Berlin Heidelberg.