Semi-Supervised Active Learning for Sound Classification in Hybrid Learning Environments
Coping with scarcity of labeled data is a common problem in sound classification tasks. Approaches for classifying sounds are commonly based on supervised learning algorithms, which require labeled data which is often scarce and leads to models that do not generalize well. In this paper, we make an...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Online |
| Language: | English |
| Published: |
Public Library of Science
2016
|
| Online Access: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5023122/ |
| id |
pubmed-5023122 |
|---|---|
| recordtype |
oai_dc |
| spelling |
pubmed-50231222016-09-27 Semi-Supervised Active Learning for Sound Classification in Hybrid Learning Environments Han, Wenjing Coutinho, Eduardo Ruan, Huabin Li, Haifeng Schuller, Björn Yu, Xiaojie Zhu, Xuan Research Article Coping with scarcity of labeled data is a common problem in sound classification tasks. Approaches for classifying sounds are commonly based on supervised learning algorithms, which require labeled data which is often scarce and leads to models that do not generalize well. In this paper, we make an efficient combination of confidence-based Active Learning and Self-Training with the aim of minimizing the need for human annotation for sound classification model training. The proposed method pre-processes the instances that are ready for labeling by calculating their classifier confidence scores, and then delivers the candidates with lower scores to human annotators, and those with high scores are automatically labeled by the machine. We demonstrate the feasibility and efficacy of this method in two practical scenarios: pool-based and stream-based processing. Extensive experimental results indicate that our approach requires significantly less labeled instances to reach the same performance in both scenarios compared to Passive Learning, Active Learning and Self-Training. A reduction of 52.2% in human labeled instances is achieved in both of the pool-based and stream-based scenarios on a sound classification task considering 16,930 sound instances. Public Library of Science 2016-09-14 /pmc/articles/PMC5023122/ /pubmed/27627768 http://dx.doi.org/10.1371/journal.pone.0162075 Text en © 2016 Han et al http://creativecommons.org/licenses/by/4.0/ This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/) , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. |
| repository_type |
Open Access Journal |
| institution_category |
Foreign Institution |
| institution |
US National Center for Biotechnology Information |
| building |
NCBI PubMed |
| collection |
Online Access |
| language |
English |
| format |
Online |
| author |
Han, Wenjing Coutinho, Eduardo Ruan, Huabin Li, Haifeng Schuller, Björn Yu, Xiaojie Zhu, Xuan |
| spellingShingle |
Han, Wenjing Coutinho, Eduardo Ruan, Huabin Li, Haifeng Schuller, Björn Yu, Xiaojie Zhu, Xuan Semi-Supervised Active Learning for Sound Classification in Hybrid Learning Environments |
| author_facet |
Han, Wenjing Coutinho, Eduardo Ruan, Huabin Li, Haifeng Schuller, Björn Yu, Xiaojie Zhu, Xuan |
| author_sort |
Han, Wenjing |
| title |
Semi-Supervised Active Learning for Sound Classification in Hybrid Learning Environments |
| title_short |
Semi-Supervised Active Learning for Sound Classification in Hybrid Learning Environments |
| title_full |
Semi-Supervised Active Learning for Sound Classification in Hybrid Learning Environments |
| title_fullStr |
Semi-Supervised Active Learning for Sound Classification in Hybrid Learning Environments |
| title_full_unstemmed |
Semi-Supervised Active Learning for Sound Classification in Hybrid Learning Environments |
| title_sort |
semi-supervised active learning for sound classification in hybrid learning environments |
| description |
Coping with scarcity of labeled data is a common problem in sound classification tasks. Approaches for classifying sounds are commonly based on supervised learning algorithms, which require labeled data which is often scarce and leads to models that do not generalize well. In this paper, we make an efficient combination of confidence-based Active Learning and Self-Training with the aim of minimizing the need for human annotation for sound classification model training. The proposed method pre-processes the instances that are ready for labeling by calculating their classifier confidence scores, and then delivers the candidates with lower scores to human annotators, and those with high scores are automatically labeled by the machine. We demonstrate the feasibility and efficacy of this method in two practical scenarios: pool-based and stream-based processing. Extensive experimental results indicate that our approach requires significantly less labeled instances to reach the same performance in both scenarios compared to Passive Learning, Active Learning and Self-Training. A reduction of 52.2% in human labeled instances is achieved in both of the pool-based and stream-based scenarios on a sound classification task considering 16,930 sound instances. |
| publisher |
Public Library of Science |
| publishDate |
2016 |
| url |
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5023122/ |
| _version_ |
1613651665355800576 |