A data model for music information retrieval

Authors:
Tamar Berman
Affiliations:
Graduate School of Library and Information Science, University of Illinois at Urbana-Champaign, Champaign, IL
Venue:
NGITS'06 Proceedings of the 6th international conference on Next Generation Information Technologies and Systems
Year:
2006

Citing 4
Cited 0

Towards the digital music library: tune retrieval from acoustic input

Proceedings of the first ACM international conference on Digital libraries
Evaluation of a simple and effective music information retrieval method

SIGIR '00 Proceedings of the 23rd annual international ACM SIGIR conference on Research and development in information retrieval
Mining Sequential Patterns

ICDE '95 Proceedings of the Eleventh International Conference on Data Engineering
Robust Polyphonic Music Retrieval with N-grams

Journal of Intelligent Information Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper describes a data model for the representation of tonal music. In this model, music is conceived as an equally-spaced time series of 12-dimensional vectors. The model has been successfully applied to the task of discovering frequently recurring patterns, and to the related task of retrieving user-defined musical patterns. This was accomplished by converting midi sequences of music by W.A. Mozart into the time series representation and analyzing these with data mining tools and SQL queries. The novelty of the pattern extraction capability supported by the model is in the potentially complex description of the sequences, which may contain both melodic and harmonic features, may be embedded within each other, or interspersed with other patterns or occurrences. A unique feature of the model is the use of time intervals as the basic representational unit, which fosters possibilities for future application to audio data.