Context-Aware Recommendation System Using Content Based Image Retrieval with Dynamic Context Considered

Abstract

This paper conceptually proposes a context-aware recommendation system that gives/recommends optimal information for users based on 1) the content of an image using content-based image retrieval engines, 2) the contextual information of its similar images on the Web, and 3) user context, namely users' situations such as their location, we are aiming to extract the detailed information to the text-unpresentable images. It is expected to increase the precision of content-based image matching by combining the partial context obtained there from other types of data such as contextual information.