UTKARSH UPADHYAY, Sonu Kumar, Prateek Dubey, Aman Singh, Miss Geetanjali
Keywords:
K-nearest neighbor (k-NN), Support Vector Machine (SVM),music genre, Mel Frequency Cepstral Coefficients (MFCC)
Abstract
Classification of music genre has always been an interest in the area of music and musical data. Classification of genre can be very important to explain some interesting problems such as creating song references, exploring related songs, finding groups which will like that specific song. The aim of our project is to find the machine learning algorithm that predicts the genre of songs using k-nearest neighbor (k-NN) and Support Vector Machine (SVM). This paper also gives the difference between k-nearest neighbor (k-NN) and Support Vector Machine (SVM) with the help of principal component analysis (PCA). The Mel Frequency Cepstral Coefficients (MFCC) is used to get the information for the data set. Also, the MFCC features are used for a particular track. From the outcome of the project, we found that without the dimensionality reduction both k-nearest neighbor and Support Vector Machine (SVM) gave more accurate results than dimensionality reduction. Overall the Support Vector Machine (SVM) is a much more effective classifier for classification of music genres. It had an accuracy of around 75%.
Article Details
Unique Paper ID: 151838
Publication Volume & Issue: Volume 8, Issue 1
Page(s): 725 - 728
Article Preview & Download
Share This Article
Join our RMS
Conference Alert
NCSEM 2024
National Conference on Sustainable Engineering and Management - 2024