Psychological model based attitude prediction for context-aware services

Context-aware computing is useful in providing individualized and automated services; so far it has focused on acquiring extra-personal and observable context such as location, weather and activity. By comparison, only very little research has been completed acquiring intra-personal and perceived context, despite its usefulness and diverse applications such as healthcare, marketing and smart homes. In particular, one of the most important psychological factors in personalized service selection and provision is individual attitude. Hence, the purpose of this paper is to propose a novel methodology of acquiring individual attitude context data for a more sophisticated context-aware service. We developed a model based approach to estimate the positive feedback of target services and products based on individual stimuli, which are regarded as external context. Models and corresponding modules are collected from research models on attitudes. To show the feasibility of the ideas proposed in this paper, we developed and examined a prototype system through laboratory experiments.