Evaluation of a simple and effective music information retrieval method

We developed, and then evaluated, a music information retrieval (MIR) system based upon the intervals found within the melodies of a collection of 9354 folksongs. The songs were converted to an interval-only representation of monophonic melodies and then fragmented t into length-n subsections called n-grams. The length of these n-grams and the degree to which we precisely represent the intervals are variables analyzed in this paper. We constructed a collection of “musical word” databases using the text-based, SMART information retrieval system. A group of simulated queries, some of which contained simulated errors, was run against these databases. The results were evaluated using the normalized precision and normalized recall measures. Our concept of “musical words” shows great merit thus implying that useful MIR systems can be constructed simply and efficiently using pre-existing text-based information retrieval software. Second, this study is a formal and comprehensive evaluation of a MIR system using rigorous statistical analyses to determine retrieval effectiveness.

[1]  Ian H. Witten,et al.  Towards the digital music library: tune retrieval from acoustic input , 1996, DL '96.

[2]  J. Stephen Downie,et al.  Music Retrieval as Text Retrieval: Simple Yet Effective (poster abstract). , 1999, SIGIR 1999.

[3]  Michael McGill,et al.  Introduction to Modern Information Retrieval , 1983 .

[4]  Marsha Berman,et al.  The directory of tunes and musical themes , 1975 .

[5]  Eleanor Selfridge-Field,et al.  Conceptual and representational issues in melodic comparison , 1998 .

[6]  Ian H. Witten,et al.  The New Zealand Digital Library MELody inDEX , 1997, D Lib Mag..

[7]  Donna K. Harman,et al.  Ranking Algorithms , 1992, Information Retrieval: Data Structures & Algorithms.

[8]  Carolyn Rabson,et al.  The National Tune Index: 18th-Century Secular Music , 1982 .

[9]  W. Dowling.  Scale and contour: Two components of a theory of memory for melodies. , 1978 .

[10]  Michael J. Nelson,et al.  Statistical methods in information science research , 1987 .

[11]  Joseph Dyer,et al.  Stanley Sadie, ed., The New Grove Dictionary of Music and Musicians . 20 vols. London: Macmillan, 1980. $1,900. , 1983 .

[12]  D. Randel,et al.  The New Harvard Dictionary of Music , 1988 .

[13]  J. Stephen Downie,et al.  Music retrieval as text retrieval (poster abstract): simple yet effective , 1999, SIGIR '99.

[14]  J. Stephen Downie,et al.  Evaluating a simple approach to music information retrieval : conceiving melodic n-grams as text , 1999 .

[15]  Rudolf Stephan,et al.  The New Harvard Dictionary of Music , 1988 .

[16]  Jean Tague-Sutcliffe,et al.  The Pragmatics of Information Retrieval Experimentation Revisited , 1997, Inf. Process. Manag..

[17]  Alexander McLane.  Music as information , 1996 .

[18]  Lutz Prechelt,et al.  An interface for melody input , 2001, TCHI.

[19]  Eleanor Selfridge-Field,et al.  Melodic Similarity : concepts, procedures, and applications , 1998 .

[20]  Mary Kay Duggan,et al.  Electronic Information and Applications in Musicology and Music Theory , 1992, Libr. Trends.

[21]  Yaşar Tonta.  Analysis of Search Failures in Document Retrieval Systems: A Review. , 1992 .