Other Links

Indexed By

    •

Volume 7 Issue 2 - July 2016

  • 1. Genre cataloging and instrument classification in large databases using music mining approach

    Authors : Rebekha R, Joseph M

    Pages : 494-506

    DOI : http://dx.doi.org/10.21172/1.72.577

    Keywords : Harmony Search Algorithm, CMM-C (Collaborative Music- Mining Classifier), Wavelet Transformation, Mel-Frequency Cepstral Coefficients (MFCC), Music Genre and Instrumental Classification.

    Abstract :

    Mining of music information has dependably been always a developing area of interest where information’s are stored in large stash. A challenge in music information management is how to utilize information indexing systems in view of various aspects of the information itself. For music data, content and different acoustic components can be connected to encourage efficient music management. In Mining of music information, characterization tasks are sub-divided into genre cataloging, instrument distinguishing and artist classification. Apart from classification, other information mining tasks are comparable for music applications. In this Paper a collaborative music-mining classifier (CMM-C) system which is based on self-adaptive harmony search algorithm. In the pre-processing stage, fundamental acoustic qualities are separated and stored as original feature set. CMC is created with altered wavelet transformation (for processing acoustic signals). Conventional frameworks utilize Fourier transform for the classifications, but from the survey it is observed that remaking high frequency coefficients are not possible from the modulus complex convolution. The classification through wavelet infuses mining instantaneous frequency of music information, mounting highlights with sub-band coefficients and spatial-temporal energy location curve. To progress the accuracy in classification to take high incidence coefficients into the processing and by solving the modernization issue with altered wavelet transform to organize the genre with higher achievement rate. As the result of evaluation it is able to complete the genre cataloging and instrument characterization with CMM classifier. The approval of this mining procedure is reasonably compared with other harmony search algorithms, multi-layer perceptrons and SVM techniques.

    Citing this Journal Article :

    Rebekha R, Joseph M, "Genre cataloging and instrument classification in large databases using music mining approach", Volume 7 Issue 2 - July 2016, 494-506

    Click here to Submit Copyright Takedown Notice for this article.

    View Paper