2003 Articles
Toward Evaluation Techniques for Music Similarity
We describe and discuss our recent work developing a database, methodology and ground truth for the evaluation of automatic techniques for music similarity. Our database consists of acoustic and textual 'Web-mined' data covering 400 popular artists. Of note is our technique of sharing acoustic features rather than raw audio to avoid copyright problems. Our evaluation methodology allows any data source to be regarded as ground truth and can highlight which measure forms the best collective ground truth. We additionally describe an evaluation methodology that is useful for data collected from people in the form of a survey about music similarity. We have successfully used our database and techniques to evaluate a number of music similarity algorithms.
Files
-
sigir03-wp.pdf application/pdf 71.2 KB Download File
Also Published In
- Title
- SIGIR 2003: Workshop on the Evaluation of Music Information Retrieval Systems, 1 August 2003, Toronto, Canada
More About This Work
- Academic Units
- Electrical Engineering
- Published Here
- July 2, 2012