I-know my users: user-centric profiling based on the perceptual preference questionnaire (PPQ)

A substantial proportion of user modeling research deals with detection, classification and use of as well as adaptation to the users' preferences. In this paper, we develop a questionnaire to assess aspects of perceptual preferences in regard to information processing, knowledge gain and learning, named Perceptual Preferences Questionnaire (PPQ). The PPQ is validated based on the results of an online survey with forum users (n=76). Given the perceptual preferences of a user, we explore first implications on how to present information to that specific user in order to ease understanding, which can be achieved by matching his/her perceptual preferences. Further research aims to combine the PPQ with implicit testing based on the users' textual utterances and the users' interests.<br> That poses a problem to existing search algorithms: Such compounds could be of high interest for a search request, but how can be examined whether a compound comprises a given lexeme? A string match can be considered as an indication, but does not prove semantic relation. The same problem is faced when using lexicon based approaches where signal words are defined as lexemes only and need to be identified in all forms of appearance, and hence also as component of a compound. This paper explores the characteristics of compounds and their constituent elements for German, and compares seven algorithms with regard to runtime and error rates. The results of this study are relevant to query analysis and term weighting approaches in information retrieval system design.