Human action recognition method based on multi-view semi-supervised ensemble learning
Mass labeled data are hard to get in mobile devices.Inadequate training leads to bad performance of classifiers in human action recognition.To tackle this problem, a multi-view semi-supervised ensemble learning method was proposed.First, data of two different inertial sensors was used to construct t...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
POSTS&TELECOM PRESS Co., LTD
2021-06-01
|
| Series: | 网络与信息安全学报 |
| Subjects: | |
| Online Access: | http://www.cjnis.com.cn/thesisDetails#10.11959/j.issn.2096-109x.2021061 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841529873863016448 |
|---|---|
| author | Shengnan CHEN Xinmin FAN |
| author_facet | Shengnan CHEN Xinmin FAN |
| author_sort | Shengnan CHEN |
| collection | DOAJ |
| description | Mass labeled data are hard to get in mobile devices.Inadequate training leads to bad performance of classifiers in human action recognition.To tackle this problem, a multi-view semi-supervised ensemble learning method was proposed.First, data of two different inertial sensors was used to construct two feature views.Two feature views and two base classifiers were combined to construct co-training framework.Then, the confidence degree was redefined in multi-class task and was combined with active learning method to control predict pseudo-label result in each iteration.Finally, extended training data was used as input to train LightGBM.Experiments show that the method has good performance in precision rate, recall rate and F1 value, which can effectively detect different human action. |
| format | Article |
| id | doaj-art-455ded18d64a46a288aa21de86d973fa |
| institution | Kabale University |
| issn | 2096-109X |
| language | English |
| publishDate | 2021-06-01 |
| publisher | POSTS&TELECOM PRESS Co., LTD |
| record_format | Article |
| series | 网络与信息安全学报 |
| spelling | doaj-art-455ded18d64a46a288aa21de86d973fa2025-01-15T03:14:52ZengPOSTS&TELECOM PRESS Co., LTD网络与信息安全学报2096-109X2021-06-01714114859564105Human action recognition method based on multi-view semi-supervised ensemble learningShengnan CHENXinmin FANMass labeled data are hard to get in mobile devices.Inadequate training leads to bad performance of classifiers in human action recognition.To tackle this problem, a multi-view semi-supervised ensemble learning method was proposed.First, data of two different inertial sensors was used to construct two feature views.Two feature views and two base classifiers were combined to construct co-training framework.Then, the confidence degree was redefined in multi-class task and was combined with active learning method to control predict pseudo-label result in each iteration.Finally, extended training data was used as input to train LightGBM.Experiments show that the method has good performance in precision rate, recall rate and F1 value, which can effectively detect different human action.http://www.cjnis.com.cn/thesisDetails#10.11959/j.issn.2096-109x.2021061human action recognitionsemi-supervised learningfew-shotLightGBM |
| spellingShingle | Shengnan CHEN Xinmin FAN Human action recognition method based on multi-view semi-supervised ensemble learning 网络与信息安全学报 human action recognition semi-supervised learning few-shot LightGBM |
| title | Human action recognition method based on multi-view semi-supervised ensemble learning |
| title_full | Human action recognition method based on multi-view semi-supervised ensemble learning |
| title_fullStr | Human action recognition method based on multi-view semi-supervised ensemble learning |
| title_full_unstemmed | Human action recognition method based on multi-view semi-supervised ensemble learning |
| title_short | Human action recognition method based on multi-view semi-supervised ensemble learning |
| title_sort | human action recognition method based on multi view semi supervised ensemble learning |
| topic | human action recognition semi-supervised learning few-shot LightGBM |
| url | http://www.cjnis.com.cn/thesisDetails#10.11959/j.issn.2096-109x.2021061 |
| work_keys_str_mv | AT shengnanchen humanactionrecognitionmethodbasedonmultiviewsemisupervisedensemblelearning AT xinminfan humanactionrecognitionmethodbasedonmultiviewsemisupervisedensemblelearning |