Time series classification using k-Nearest neighbours, Multilayer Perceptron and Learning Vector Quantization algorithms

Abstract

We are presenting results comparison of three artificial intelligence algorithms in a classification of time series derived from musical excerpts in this paper. Algorithms were chosen to represent different principles of classification – statistic approach, neural networks and competitive learning. The first algorithm is a classical k-Nearest neighbours algorithm, the second algorithm is Multilayer Perceptron (MPL), an example of artificial neural network and the third one is a Learning Vector Quantization (LVQ) algorithm representing supervised counterpart to unsupervised Self Organizing Map (SOM).After our own former experiments with unlabelled data we moved forward to the data labels utilization, which generally led to a better accuracy of classification results. As we need huge data set of labelled time series (a priori knowledge of correct class which each time series instance belongs to), we used, with a good experience in former studies, musical excerpts as a source of real-world time series. We are using standard deviation of the sound signal as a descriptor of a musical excerpts volume level.We are describing principle of each algorithm as well as its implementation briefly, giving links for further research. Classification results of each algorithm are presented in a confusion matrix showing numbers of misclassifications and allowing to evaluate overall accuracy of the algorithm. Results are compared and particular misclassifications are discussed for each algorithm. Finally the best solution is chosen and further research goals are given.

Audio data classification by means of new algorithms

“…This method can be successfully used for such tasks clarinet vs. talking or heavy vs. silence Magnatagatune tags discrimination. The accuracy of such tasks can be as high as 98% [2] in case of heavy vs. silence or 92% in case of clarinet vs. talking recordings. These datasets represent well-separable data, explaining good classification results.…”

“…It can be used to improve clustering results of SOM by using data labels [8]. For adjusting the codebook vector algorithm moves the codebook vector according to (2) …”

“…These recordings are in addition labeled by humans with tags evaluating many aspect of recordings like volume, instrumentation, genre etc. We used this database also in our former experiments [4], [3] and [2].…”

Analysis of sound data streamed over the network

“…We use algorithm called Dynamic Time Warping (Müller, 2007) commonly used for time series comparison in this paper. Some other time series exploration approaches can be found in (Fejfar, 2011) and (Fejfar, 2012). The data was acquired in our network laboratory simulating network traffi c by downloading fi les, streaming audio and video simultaneously.…”