A Human Activity Recognition Algorithm Based on Stacking Denoising Autoencoder and LightGBM
Recently, the demand for human activity recognition has become more and more urgent. It is widely used in indoor positioning, medical monitoring, safe driving, etc. Existing activity recognition approaches require either the location information of the sensors or the specific domain knowledge, which...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-02-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/19/4/947 |
| _version_ | 1828137301790687232 |
|---|---|
| author | Xile Gao Haiyong Luo Qu Wang Fang Zhao Langlang Ye Yuexia Zhang |
| author_facet | Xile Gao Haiyong Luo Qu Wang Fang Zhao Langlang Ye Yuexia Zhang |
| author_sort | Xile Gao |
| collection | DOAJ |
| description | Recently, the demand for human activity recognition has become more and more urgent. It is widely used in indoor positioning, medical monitoring, safe driving, etc. Existing activity recognition approaches require either the location information of the sensors or the specific domain knowledge, which are expensive, intrusive, and inconvenient for pervasive implementation. In this paper, a human activity recognition algorithm based on SDAE (Stacking Denoising Autoencoder) and LightGBM (LGB) is proposed. The SDAE is adopted to sanitize the noise in raw sensor data and extract the most effective characteristic expression with unsupervised learning. The LGB reveals the inherent feature dependencies among categories for accurate human activity recognition. Extensive experiments are conducted on four datasets of distinct sensor combinations collected by different devices in three typical application scenarios, which are human moving modes, current static, and dynamic behaviors of users. The experimental results demonstrate that our proposed algorithm achieves an average accuracy of 95.99%, outperforming other comparative algorithms using XGBoost, CNN (Convolutional Neural Network), CNN + Statistical features, or single SDAE. |
| first_indexed | 2024-04-11T18:16:12Z |
| format | Article |
| id | doaj.art-58796fabe8824caba2676a0ab1962e83 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-11T18:16:12Z |
| publishDate | 2019-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-58796fabe8824caba2676a0ab1962e832022-12-22T04:09:52ZengMDPI AGSensors1424-82202019-02-0119494710.3390/s19040947s19040947A Human Activity Recognition Algorithm Based on Stacking Denoising Autoencoder and LightGBMXile Gao0Haiyong Luo1Qu Wang2Fang Zhao3Langlang Ye4Yuexia Zhang5Beijing Key Laboratory of Mobile Computing and Pervasive Device, Institute of Computing Technology Chinese Academy of Sciences, Beijing 100190, ChinaBeijing Key Laboratory of Mobile Computing and Pervasive Device, Institute of Computing Technology Chinese Academy of Sciences, Beijing 100190, ChinaSchool of Information and Communication Engineering, Beijing University of Posts and Telecommunication, Beijing 100876, ChinaSchool of Software Engineering, Beijing University of Posts and Telecommunication, Beijing 100876, ChinaBeijing Key Laboratory of Mobile Computing and Pervasive Device, Institute of Computing Technology Chinese Academy of Sciences, Beijing 100190, ChinaSchool of Information and Communication Engineering, Beijing Information Science and Technology University, Beijing 100876, ChinaRecently, the demand for human activity recognition has become more and more urgent. It is widely used in indoor positioning, medical monitoring, safe driving, etc. Existing activity recognition approaches require either the location information of the sensors or the specific domain knowledge, which are expensive, intrusive, and inconvenient for pervasive implementation. In this paper, a human activity recognition algorithm based on SDAE (Stacking Denoising Autoencoder) and LightGBM (LGB) is proposed. The SDAE is adopted to sanitize the noise in raw sensor data and extract the most effective characteristic expression with unsupervised learning. The LGB reveals the inherent feature dependencies among categories for accurate human activity recognition. Extensive experiments are conducted on four datasets of distinct sensor combinations collected by different devices in three typical application scenarios, which are human moving modes, current static, and dynamic behaviors of users. The experimental results demonstrate that our proposed algorithm achieves an average accuracy of 95.99%, outperforming other comparative algorithms using XGBoost, CNN (Convolutional Neural Network), CNN + Statistical features, or single SDAE.https://www.mdpi.com/1424-8220/19/4/947human activity recognitionindoor positioningdeep learningStacking Denoising AutoencoderLightGBM |
| spellingShingle | Xile Gao Haiyong Luo Qu Wang Fang Zhao Langlang Ye Yuexia Zhang A Human Activity Recognition Algorithm Based on Stacking Denoising Autoencoder and LightGBM Sensors human activity recognition indoor positioning deep learning Stacking Denoising Autoencoder LightGBM |
| title | A Human Activity Recognition Algorithm Based on Stacking Denoising Autoencoder and LightGBM |
| title_full | A Human Activity Recognition Algorithm Based on Stacking Denoising Autoencoder and LightGBM |
| title_fullStr | A Human Activity Recognition Algorithm Based on Stacking Denoising Autoencoder and LightGBM |
| title_full_unstemmed | A Human Activity Recognition Algorithm Based on Stacking Denoising Autoencoder and LightGBM |
| title_short | A Human Activity Recognition Algorithm Based on Stacking Denoising Autoencoder and LightGBM |
| title_sort | human activity recognition algorithm based on stacking denoising autoencoder and lightgbm |
| topic | human activity recognition indoor positioning deep learning Stacking Denoising Autoencoder LightGBM |
| url | https://www.mdpi.com/1424-8220/19/4/947 |
| work_keys_str_mv | AT xilegao ahumanactivityrecognitionalgorithmbasedonstackingdenoisingautoencoderandlightgbm AT haiyongluo ahumanactivityrecognitionalgorithmbasedonstackingdenoisingautoencoderandlightgbm AT quwang ahumanactivityrecognitionalgorithmbasedonstackingdenoisingautoencoderandlightgbm AT fangzhao ahumanactivityrecognitionalgorithmbasedonstackingdenoisingautoencoderandlightgbm AT langlangye ahumanactivityrecognitionalgorithmbasedonstackingdenoisingautoencoderandlightgbm AT yuexiazhang ahumanactivityrecognitionalgorithmbasedonstackingdenoisingautoencoderandlightgbm AT xilegao humanactivityrecognitionalgorithmbasedonstackingdenoisingautoencoderandlightgbm AT haiyongluo humanactivityrecognitionalgorithmbasedonstackingdenoisingautoencoderandlightgbm AT quwang humanactivityrecognitionalgorithmbasedonstackingdenoisingautoencoderandlightgbm AT fangzhao humanactivityrecognitionalgorithmbasedonstackingdenoisingautoencoderandlightgbm AT langlangye humanactivityrecognitionalgorithmbasedonstackingdenoisingautoencoderandlightgbm AT yuexiazhang humanactivityrecognitionalgorithmbasedonstackingdenoisingautoencoderandlightgbm |