Individual difference for HCI systems: Examining the probability of thinking style signature in online interaction

Researches on human thinking styles have received wide attention, especially in online education; an aspect of human computer interaction. However, the deciphering of specific thinking styles of an online user has suffered setbacks due to the limited exposition on the probability of online thinking style signature of users. This study explored this limitation with the assumption that human daily behavioral pattern can be translated into the electronic media of communication frequently used. To ascertain this assumption, server-side web data of 43-respondents were collected for 10-months as well as a self-report thinking style measurement instrument for each respondent. Cluster dichotomies from two thinking styles were extracted. Various supervised machine learning techniques were then applied to distinguish individuals on each dichotomy. The result showed that thinking styles of individuals on different dichotomies can be reliably distinguished on the internet using Logistic model tree with Bagging technique. The result demonstrates a high probability of the existence of digital thinking style. The findings from this study find relevance and application in human-centered graphical user interface design for recommender system as well as in e-commerce services. It also finds application in online profiling processes especially in e-learning systems.