By Topic

Content-based music retrieval with nonlinear feature space transformation using relevance feedback

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

2 Author(s)
Sakai, S. ; Grad. Sch. of Syst. & Inf. Eng., Univ. of Tsukuba, Tsukuba ; Kameyama, K.

In recent years, studies of similar music retrieval have been conducted actively. However, because the similarity of music is based on subjective measures, the systems need to be adaptive to user preference. In this paper, we propose an effective method for adaptive similar music retrieval reflecting the user preference by nonlinear feature space transformation based on relevance feedback. The user's evaluation to initial retrieval ranking is used to train a neural network feature space transformation. Also, as the initial stage of retrieval, a coarse division is made to the set of music in the database, to reduce the retrieval computation. In the experiments, it was observed that in the retrieval after feature space transformation, the proposed method gave preferred retrievals according to objective measures when compared with the case without transformation.

Published in:

Systems, Man and Cybernetics, 2008. SMC 2008. IEEE International Conference on

Date of Conference:

12-15 Oct. 2008