An intelligent classification algorithm for lifelog multimedia applications

Abstract

LifeLog can be used as a stand-alone consumer device to serve as a powerful automated multimedia diary and scrapbook. By using a search engine interface, the user can easily retrieval a specific thread of past transactions, or recall a few seconds ago or from many years earlier in as much detail as is desired, including imagery , audio, or video event. In this paper, we have studied and implemented three audio classification algorithms. The testing results show that KNN is the best classifier to classify the multimedia data. © 2008 IEEE.