Resource recommender system based on psychological user type indicator

摘要

In an Internet of Things (IoT) environment, service composition and collaboration among heterogeneous resources are required. Thus, an infrastructure that supports these requirements is an essential factor in a seamless service delivery. For these requirements, mobile devices should have multiple functions. However, the miniaturization of mobile devices is another requirement, and a trade-off between the two requirements is naturally generated. My previous study proposed the resource collaboration system that provides a service consisting of shareable resources in the surrounding area to solve the resource limitations of devices. Reducing the processing time for generating the recommendation and improving user satisfaction about the results are important factors, particularly for a small mobile device with limited resources. This study analyzes and classifies personal user preferences from resource usage history based on the Myers-Briggs type indicator. The study also proposes a method to recommend customized resources for classified user types. Results show that the proposed method reduces the recommendation time and increases user satisfaction.