Hyperparameter Optimization Method Based on Harmony Search Algorithm to Improve Performance of 1D CNN Human Respiration Pattern Recognition System
In this study, we propose a method to find an optimal combination of hyperparameters to improve the accuracy of respiration pattern recognition in a 1D (Dimensional) convolutional neural network (CNN). The proposed method is designed to integrate with a 1D CNN using the harmony search algorithm. In...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-07-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/20/13/3697 |
| _version_ | 1797563523038969856 |
|---|---|
| author | Seong-Hoon Kim Zong Woo Geem Gi-Tae Han |
| author_facet | Seong-Hoon Kim Zong Woo Geem Gi-Tae Han |
| author_sort | Seong-Hoon Kim |
| collection | DOAJ |
| description | In this study, we propose a method to find an optimal combination of hyperparameters to improve the accuracy of respiration pattern recognition in a 1D (Dimensional) convolutional neural network (CNN). The proposed method is designed to integrate with a 1D CNN using the harmony search algorithm. In an experiment, we used the depth of the convolutional layer of the 1D CNN, the number and size of kernels in each layer, and the number of neurons in the dense layer as hyperparameters for optimization. The experimental results demonstrate that the proposed method provided a recognition rate for five respiration patterns of approximately 96.7% on average, which is an approximately 2.8% improvement over an existing method. In addition, the number of iterations required to derive the optimal combination of hyperparameters was 2,000,000 in the previous study. In contrast, the proposed method required only 3652 iterations. |
| first_indexed | 2024-03-10T18:44:49Z |
| format | Article |
| id | doaj.art-bc2de501f07849bb965d1b8ce97095ad |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T18:44:49Z |
| publishDate | 2020-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-bc2de501f07849bb965d1b8ce97095ad2023-11-20T05:36:22ZengMDPI AGSensors1424-82202020-07-012013369710.3390/s20133697Hyperparameter Optimization Method Based on Harmony Search Algorithm to Improve Performance of 1D CNN Human Respiration Pattern Recognition SystemSeong-Hoon Kim0Zong Woo Geem1Gi-Tae Han2Department of Computer Engineering, Gachon University, Seongnam 13120, KoreaDepartment of Energy IT, Gachon University, Seoongnam 13120, KoreaDepartment of Computer Engineering, Gachon University, Seongnam 13120, KoreaIn this study, we propose a method to find an optimal combination of hyperparameters to improve the accuracy of respiration pattern recognition in a 1D (Dimensional) convolutional neural network (CNN). The proposed method is designed to integrate with a 1D CNN using the harmony search algorithm. In an experiment, we used the depth of the convolutional layer of the 1D CNN, the number and size of kernels in each layer, and the number of neurons in the dense layer as hyperparameters for optimization. The experimental results demonstrate that the proposed method provided a recognition rate for five respiration patterns of approximately 96.7% on average, which is an approximately 2.8% improvement over an existing method. In addition, the number of iterations required to derive the optimal combination of hyperparameters was 2,000,000 in the previous study. In contrast, the proposed method required only 3652 iterations.https://www.mdpi.com/1424-8220/20/13/36971D convolutional neural networkultra-wideband radarrespiration patternsharmony search algorithmhyperparameter optimization |
| spellingShingle | Seong-Hoon Kim Zong Woo Geem Gi-Tae Han Hyperparameter Optimization Method Based on Harmony Search Algorithm to Improve Performance of 1D CNN Human Respiration Pattern Recognition System Sensors 1D convolutional neural network ultra-wideband radar respiration patterns harmony search algorithm hyperparameter optimization |
| title | Hyperparameter Optimization Method Based on Harmony Search Algorithm to Improve Performance of 1D CNN Human Respiration Pattern Recognition System |
| title_full | Hyperparameter Optimization Method Based on Harmony Search Algorithm to Improve Performance of 1D CNN Human Respiration Pattern Recognition System |
| title_fullStr | Hyperparameter Optimization Method Based on Harmony Search Algorithm to Improve Performance of 1D CNN Human Respiration Pattern Recognition System |
| title_full_unstemmed | Hyperparameter Optimization Method Based on Harmony Search Algorithm to Improve Performance of 1D CNN Human Respiration Pattern Recognition System |
| title_short | Hyperparameter Optimization Method Based on Harmony Search Algorithm to Improve Performance of 1D CNN Human Respiration Pattern Recognition System |
| title_sort | hyperparameter optimization method based on harmony search algorithm to improve performance of 1d cnn human respiration pattern recognition system |
| topic | 1D convolutional neural network ultra-wideband radar respiration patterns harmony search algorithm hyperparameter optimization |
| url | https://www.mdpi.com/1424-8220/20/13/3697 |
| work_keys_str_mv | AT seonghoonkim hyperparameteroptimizationmethodbasedonharmonysearchalgorithmtoimproveperformanceof1dcnnhumanrespirationpatternrecognitionsystem AT zongwoogeem hyperparameteroptimizationmethodbasedonharmonysearchalgorithmtoimproveperformanceof1dcnnhumanrespirationpatternrecognitionsystem AT gitaehan hyperparameteroptimizationmethodbasedonharmonysearchalgorithmtoimproveperformanceof1dcnnhumanrespirationpatternrecognitionsystem |