TY - GEN
T1 - Classification of Subliminal Affective Priming Effect Based on AE and SVM
AU - Yin, Yongqiang
AU - Hu, Bin
AU - Li, Tiantian
AU - Zheng, Xiangwei
N1 - Publisher Copyright:
© 2019, Springer Nature Singapore Pte Ltd.
PY - 2019
Y1 - 2019
N2 - The study of the Subliminal Affective Priming Effect (SAPE) mainly uses event-related potential technology and mapping method. Many researches are only for the study of emotional classification, but there are few researches on the classification of the SAPE. That is, the SAPE is directly judged by the psychologist in most experiment. So, this paper designs a classifier based on Automatic Encoder (AE) and Support Vector Machine (SVM) for automatic recognition of SPAE. Initially, this paper collects EEG signal, and then extracts statistical features from EEG signal to form a data set. After that, the data set is dimension reduction by AE and then divided into training set and test set randomly. At last, the already designed model is trained with the training set and validated with the test set. In the experiment, we find that the designed classifier has the best performance compared with the classifiers based on BP neural network, Principal Component Analysis (PCA) and SVM. The experimental results show that the average classification accuracy is 95.31%. The classification results further indicate that the SAPE’s judgment is hopeful to reduce the labor with the machine.
AB - The study of the Subliminal Affective Priming Effect (SAPE) mainly uses event-related potential technology and mapping method. Many researches are only for the study of emotional classification, but there are few researches on the classification of the SAPE. That is, the SAPE is directly judged by the psychologist in most experiment. So, this paper designs a classifier based on Automatic Encoder (AE) and Support Vector Machine (SVM) for automatic recognition of SPAE. Initially, this paper collects EEG signal, and then extracts statistical features from EEG signal to form a data set. After that, the data set is dimension reduction by AE and then divided into training set and test set randomly. At last, the already designed model is trained with the training set and validated with the test set. In the experiment, we find that the designed classifier has the best performance compared with the classifiers based on BP neural network, Principal Component Analysis (PCA) and SVM. The experimental results show that the average classification accuracy is 95.31%. The classification results further indicate that the SAPE’s judgment is hopeful to reduce the labor with the machine.
KW - Automatic Encoder
KW - BP neural network
KW - Subliminal Affective Priming Effect
KW - Support vector machine
UR - http://www.scopus.com/inward/record.url?scp=85076233742&partnerID=8YFLogxK
U2 - 10.1007/978-981-15-1377-0_60
DO - 10.1007/978-981-15-1377-0_60
M3 - Conference contribution
AN - SCOPUS:85076233742
SN - 9789811513763
T3 - Communications in Computer and Information Science
SP - 778
EP - 788
BT - Computer Supported Cooperative Work and Social Computing - 14th CCF Conference, ChineseCSCW 2019, Revised Selected Papers
A2 - Sun, Yuqing
A2 - Lu, Tun
A2 - Yu, Zhengtao
A2 - Fan, Hongfei
A2 - Gao, Liping
PB - Springer
T2 - 14th CCF Conference on Computer Supported Cooperative Work and Social Computing, ChineseCSCW 2019
Y2 - 16 August 2019 through 18 August 2019
ER -