VISCORS : a visual contents recommender system on the mobile web

Abstract

As mobile Web technology becomes more increasingly applicable, the mobile contents market, especially character image downloading for mobile phones, has recorded remarkable growth. In spite of this rapid growth, customers experience high levels of frustration in the process of searching for desired character images. We proposed a visual contents recommender system (VISCORS) to enhance the customers’ search efficiency by combining collaborative filtering with a content-based image retrieval technique. An experiment was undertaken with the participation of real mobile Web customers using the 230 character images that Korea Telecom Freetel (KTF), a leading Korean CDMA carrier, is currently offering. The experiment result confirmed that VISCORS is a significantly better searching aid than the two other benchmarked searching systems - best-seller-based and pure CF systems - and it could be used as a realistic solution for character image recommendation within the large-scale image database in the mobile web evironment.