Investigation on the Impact of Preprocessing Methods and Parameter Selection in Acoustic Scene Classification Based on K-means Clustering Algorithm

DOI

10.2991/978-94-6463-300-9_30How to use a DOI?

Keywords

Acoustic Scene Classification; K-means Clustering Algorithm; Machine Learning Algorithms

Abstract

This research investigates the effectiveness of various preprocessing methods and parameters on Acoustic Scene Classification (ASC) using the K-means clustering algorithm. Utilizing the ESC-50 dataset, a combination of Principal Component Analysis (PCA) and StandardScaler was employed for preprocessing. The study's key findings include the identification of an optimal number of PCA components, around 30, which maximized the accuracy of the K-means algorithm. Additionally, the results revealed an unexpected phenomenon where increasing the number of clusters beyond the actual class count improved the model's accuracy, indicating potential nuanced sub-groupings within classes. These insights highlight the significance of preprocessing methods and the choice of parameters on the performance of ASC models. However, the findings may not be universally applicable across other datasets or feature sets. The study offers potential directions for future research, suggesting the exploration of other machine learning algorithms and further investigation into the potential sub-groupings within classes.

Copyright

© 2023 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Yuanyao Zuo
PY  - 2023
DA  - 2023/11/27
TI  - Investigation on the Impact of Preprocessing Methods and Parameter Selection in Acoustic Scene Classification Based on K-means Clustering Algorithm
BT  - Proceedings of the 2023 International Conference on Image, Algorithms and Artificial Intelligence (ICIAAI 2023)
PB  - Atlantis Press
SP  - 300
EP  - 306
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-300-9_30
DO  - 10.2991/978-94-6463-300-9_30
ID  - Zuo2023
ER  -