{"title":"Optimization of Feature Selection and Classification of Oriental Music Instruments Identification","authors":"P. Uruthiran, L. Ranathunga","doi":"10.1109/AiDAS47888.2019.8970974","DOIUrl":null,"url":null,"abstract":"Classification of music instrument is a challenging but important problem in music information retrieval. In music instrument identification, multimedia signal processing is heavily utilized. In this work, we have presented a sequential forward feature selection method to select a suitable feature set for the classification. We have used a reduced number of input data for the classification. Spectral domain and Time domain features are used for this study. Music instrument signals are identified as belonging to one of the three families namely string, brass, and woodwimt Decision tree, k-Nearest Neighbor (kNN) and Support Vector Machines (SVM) have been used as classifiers. The average accuracy achieved from SVM classifier has recorded the highest value as 93.37%. Therefore, it is concluded that the SVM classifier is the best classifier among the other classifiers for the derived feature vector.","PeriodicalId":227508,"journal":{"name":"2019 1st International Conference on Artificial Intelligence and Data Sciences (AiDAS)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2019-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 1st International Conference on Artificial Intelligence and Data Sciences (AiDAS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/AiDAS47888.2019.8970974","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Classification of music instrument is a challenging but important problem in music information retrieval. In music instrument identification, multimedia signal processing is heavily utilized. In this work, we have presented a sequential forward feature selection method to select a suitable feature set for the classification. We have used a reduced number of input data for the classification. Spectral domain and Time domain features are used for this study. Music instrument signals are identified as belonging to one of the three families namely string, brass, and woodwimt Decision tree, k-Nearest Neighbor (kNN) and Support Vector Machines (SVM) have been used as classifiers. The average accuracy achieved from SVM classifier has recorded the highest value as 93.37%. Therefore, it is concluded that the SVM classifier is the best classifier among the other classifiers for the derived feature vector.