Automatic Adaptation Method in Intelligent Image Retrieval System

초록

Information overload in modern electronic life is inevitable problem. It is more difficult for users to find information what they need. For resolving this problem, in this paper, we present the framework design and implementation issues of image retrieval system called IIRS with the capability of user adaptation. Even though many researches have been performed on the development of efficient image retrieval engines, most of them have focused on the performance of the system, but not friendly and efficient user interface. The satisfaction of users is at least as important as the functionality and performance in image retrieval systems. Intelligent user interface with the adaptation method enables the IIRS to make more intelligent, natural, and efficient. We address the adaptation method that consists of decision tree and back-propagation neural network. They have been employed for long-term and short-term adaptations, respectively. Experimental results show automatic adaptation method can improve the performance of the IIRS