TY - GEN
T1 - Hierarchical classification and visualization with multiple feature ranking criteria
AU - Wang, Qun
AU - Wang, Xuegang
AU - Zhou, Zhiguo
AU - Sheng, Di
AU - Sheng, Duozheng
N1 - Publisher Copyright:
© 2021 IEEE.
PY - 2021
Y1 - 2021
N2 - Hierarchical classification is consistent with human cognitive thinking mode and easy to understand. In the process of classification, the feature ranking method has a great impact on the classification result and convergence rate. In this paper, a hierarchical classification method with multiple feature ranking criteria is proposed, including Least Number of Overlapping Interval Samples (LNOIS) method, Maximum Average Distance (MAD) method, Minimum OTSU-MSE (MOTSU-MSE) method, etc. The proposed method is intuitive and concise without adjusting the specific super parameter. To enhance the interpretability of this method, a visual system is designed based on JavaScript programming language. The method is applied to the recognition of human daily behavior, and effective features are extracted and filtered according to the characteristics of signals. The hierarchical classification model is trained based on OTSU-MSE method, and 93.06% F1 score is obtained.
AB - Hierarchical classification is consistent with human cognitive thinking mode and easy to understand. In the process of classification, the feature ranking method has a great impact on the classification result and convergence rate. In this paper, a hierarchical classification method with multiple feature ranking criteria is proposed, including Least Number of Overlapping Interval Samples (LNOIS) method, Maximum Average Distance (MAD) method, Minimum OTSU-MSE (MOTSU-MSE) method, etc. The proposed method is intuitive and concise without adjusting the specific super parameter. To enhance the interpretability of this method, a visual system is designed based on JavaScript programming language. The method is applied to the recognition of human daily behavior, and effective features are extracted and filtered according to the characteristics of signals. The hierarchical classification model is trained based on OTSU-MSE method, and 93.06% F1 score is obtained.
KW - Feature sorting criteria
KW - Hierarchical classification
KW - Human daily behavior recognition
KW - MOTSU-MSE
KW - Visualization system
UR - http://www.scopus.com/inward/record.url?scp=85123500427&partnerID=8YFLogxK
U2 - 10.1109/CISP-BMEI53629.2021.9624378
DO - 10.1109/CISP-BMEI53629.2021.9624378
M3 - Conference contribution
AN - SCOPUS:85123500427
T3 - Proceedings - 2021 14th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2021
BT - Proceedings - 2021 14th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2021
A2 - Li, Qingli
A2 - Wang, Lipo
A2 - Wang, Yan
A2 - Li, Wenwu
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 14th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, CISP-BMEI 2021
Y2 - 23 October 2021 through 25 October 2021
ER -