RFID-based user profiling of fashion preferences: blueprint for a smart wardrobe

In this paper, we explore the possibility of generating user profiles of fashion preferences from information captured by RFID technology. Proposing a design of a smart wardrobe, we investigate the appropriateness of the technology as an identification tool of real objects and to aid in detecting and tracking real objects movements. We then present a model of user fashion profile, which is generated through queries and data mining techniques. In order to illustrate the usefulness and real world feasibility of our proposed model, we build a working prototype as proof of concept. For evaluation purposes, we have created a random generator, which is able to generate random clothing items and dressing events that serve as input to our model for creating user profiles. Our experimental results clearly indicate that RFID technology is suitable to aid in creating smart systems.