An enhanced fitness function to recognize unbalanced human emotions data

作者：

Highlights：

• An enhanced distance score (eD-score) fitness function is proposed to handle unbalanced dataset classification.

• A new EEG signal dataset for emotional clips is created with a portable, computer-efficient single-channel EEG headset.

• A novel eDGP framework is proposed for the classification of unbalanced emotion recognition data.

• Analysis of human behavior for age, genre, gender, and attention level is performed.

摘要

•An enhanced distance score (eD-score) fitness function is proposed to handle unbalanced dataset classification.•A new EEG signal dataset for emotional clips is created with a portable, computer-efficient single-channel EEG headset.•A novel eDGP framework is proposed for the classification of unbalanced emotion recognition data.•Analysis of human behavior for age, genre, gender, and attention level is performed.

论文关键词：Emotion recognition,Fitness function,EEG,Fast Fourier Transformation,Unbalanced dataset

论文评审过程：Received 25 September 2019, Revised 26 May 2020, Accepted 11 September 2020, Available online 30 September 2020, Version of Record 20 October 2020.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114011