TY - JOUR
T1 - A probability and integrated learning based classification algorithm for high-level human emotion recognition problems
AU - Jiang, Dazhi
AU - Wu, Kaichao
AU - Chen, Dicheng
AU - Tu, Geng
AU - Zhou, Teng
AU - Garg, Akhil
AU - Gao, Liang
N1 - Funding Information:
The authors would like to thank anonymous reviewers for their very detailed and helpful review. This work was supported by National Natural Science Foundation of China ( 61902232 , 61902231 ), Post-Funded Projects of National Natural Science Foundation of China ( 140-11319011 ), Key Project of Basic and Applied Basic Research of Colleges and Universities in Guangdong Province (Natural Science) ( 2018KZDXM035 ).
Funding Information:
The authors would like to thank anonymous reviewers for their very detailed and helpful review. This work was supported by National Natural Science Foundation of China (61902232, 61902231), Post-Funded Projects of National Natural Science Foundation of China (140-11319011), Key Project of Basic and Applied Basic Research of Colleges and Universities in Guangdong Province (Natural Science) (2018KZDXM035).
Publisher Copyright:
© 2019 Elsevier Ltd
PY - 2020/1
Y1 - 2020/1
N2 - In this paper, a probability and integrated learning (PIL) based classification algorithm is proposed for solving high-level human emotion recognition problems. Firstly, by simulating human thinking mode and construction, a novel topology of integrated learning is proposed to obtain the essential material basis for analyzing the complex human emotions. Secondly, classification algorithm based on PIL is presented to adapt the emotion classification fuzziness caused by the emotional uncertainty, which is realized by calculating the confidence interval of the classification probability. This paper also presented three new analyses methods based on classification probability including the emotional sensitivity, emotional decision preference and emotional tube. Our study expects that the proposed method could be used in the affective computing for video, and may play a reference role in artificial emotion established for robot with a natural and humanized way.
AB - In this paper, a probability and integrated learning (PIL) based classification algorithm is proposed for solving high-level human emotion recognition problems. Firstly, by simulating human thinking mode and construction, a novel topology of integrated learning is proposed to obtain the essential material basis for analyzing the complex human emotions. Secondly, classification algorithm based on PIL is presented to adapt the emotion classification fuzziness caused by the emotional uncertainty, which is realized by calculating the confidence interval of the classification probability. This paper also presented three new analyses methods based on classification probability including the emotional sensitivity, emotional decision preference and emotional tube. Our study expects that the proposed method could be used in the affective computing for video, and may play a reference role in artificial emotion established for robot with a natural and humanized way.
KW - Classification probability
KW - Emotion analysis problem
KW - Integrated learning
UR - http://www.scopus.com/inward/record.url?scp=85072570162&partnerID=8YFLogxK
U2 - 10.1016/j.measurement.2019.107049
DO - 10.1016/j.measurement.2019.107049
M3 - Article
AN - SCOPUS:85072570162
SN - 0263-2241
VL - 150
JO - Measurement
JF - Measurement
M1 - 107049
ER -