Computer-aided thinking by mapping text-objects into metric spaces

摘要

This paper presents a system for computer-aided thinking. We propose the idea of reflecting the mental world indirectly into a metric space to support such human thinking activities as externalizing and forming new ideas. We use a method that maps text-objects into metric spaces for visualizing a user's thought space structure. Text-objects imply fragments of a user's idea, which have several keywords given by him/her. Spaces composed of text-objects are configured in the way “the higher the mutual relevance between a pair of text-objects is, the closer the text-objects are mapped”. The relevance values among text-objects are calculated due to cooccurrence of their keywords. Results of experiments with our implemented system, named CAT1 (computer-aided thinking, version 1), show that users of the system can get effective stimuli for further thinking in creative concept formation. The paper also discusses the potential application of CAT1 to collaborative work by groups of people.