Randomized algorithms
Neural Networks for Pattern Recognition
Neural Networks for Pattern Recognition
Alternatives to the k-means algorithm that find better clusterings
Proceedings of the eleventh international conference on Information and knowledge management
Pattern Classification (2nd Edition)
Pattern Classification (2nd Edition)
International Journal of Computer Vision - Special Issue on Content-Based Image Retrieval
MARSYAS: a framework for audio analysis
Organised Sound
MARSYAS: a framework for audio analysis
Organised Sound
An experimental comparison of several clustering and initialization methods
UAI'98 Proceedings of the Fourteenth conference on Uncertainty in artificial intelligence
IEEE Transactions on Neural Networks
Music genre classification using MIDI and audio features
EURASIP Journal on Applied Signal Processing
Classification of audio signals using Fuzzy c-Means with divergence-based Kernel
Pattern Recognition Letters
A novel approach to musical genre classification using probabilistic latent semantic analysis model
ICME'09 Proceedings of the 2009 IEEE international conference on Multimedia and Expo
Effective initialization of k-means for color quantization
ICIP'09 Proceedings of the 16th IEEE international conference on Image processing
A very fast neural learning for classification using only new incoming datum
IEEE Transactions on Neural Networks
Music genre classification based on ensemble of signals produced by source separation methods
Intelligent Decision Technologies
Improving the performance of k-means for color quantization
Image and Vision Computing
Classification of audio signals using gradient-based fuzzy c-means algorithm with divergence measure
PCM'05 Proceedings of the 6th Pacific-Rim conference on Advances in Multimedia Information Processing - Volume Part I
Information Sciences: an International Journal
An analysis of content-based classification of audio signals using a fuzzy c-means algorithm
Multimedia Tools and Applications
Classification accuracy is not enough
Journal of Intelligent Information Systems
| Hi-index | 0.00 |
This paper explores the automatic classification of audio tracks into musical genres. Our goal is to achieve human-level accuracy with fast training and classification. This goal is achieved with radial basis function (RBF) networks by using a combination of unsupervised and supervised initialization methods. These initialization methods yield classifiers that are as accurate as RBF networks trained with gradient descent (which is hundreds of times slower). In addition, feature subset selection further reduces training and classification time while preserving classification accuracy. Combined, our methods succeed in creating an RBF network that matches the musical classification accuracy of humans. The general algorithmic contribution of this paper is to show experimentally that RBF networks initialized with a combination of methods can yield good classification performance without relying on gradient descent. The simplicity and computational efficiency of our initialization methods produce classifiers that are fast to train as well as fast to apply to novel data. We also present an improved method for initializing the k{\hbox{-}}{\rm means} clustering algorithm which is useful for both unsupervised and supervised initialization methods.