A Wrapper Feature Selection Algorithm: An Emotional Assessment Using Physiological Recordings from Wearable Sensors

Assessing emotional state is an emerging application field boosting research activities on the topic of analysis of non-invasive biosignals to find effective markers to accurately determine the emotional state in real-time. Nowadays using wearable sensors, electrocardiogram and thoracic impedance measurements can be recorded, facilitating analyzing cardiac and respiratory functions directly and autonomic nervous system function indirectly. Such analysis allows distinguishing between different emotional states: neutral, sadness, and disgust. This work was specifically focused on the proposal of a k-fold approach for selecting features while training the classifier that reduces the loss of generalization. The performance of the proposed algorithm used as the selection criterion was compared to the commonly used standard error function. The proposed k-fold approach outperforms the conventional method with 4% hit success rate improvement, reaching an accuracy near to 78%. Moreover, the proposed selection criterion method allows the classifier to produce the best performance using a lower number of features at lower computational cost. A reduced number of features reduces the risk of overfitting while a lower computational cost contributes to implementing real-time systems using wearable electronics.

[1]  Jing Cai,et al.  The Research on Emotion Recognition from ECG Signal , 2009, 2009 International Conference on Information Technology and Computer Science.

[2]  J. Gorman,et al.  Effect of mental stress throughout the day on cardiac autonomic control , 1994, Biological Psychology.

[3]  Jacob Zahavi,et al.  Using simulated annealing to optimize the feature selection problem in marketing applications , 2006, Eur. J. Oper. Res..

[4]  Carmen C. Y. Poon,et al.  Unobtrusive Sensing and Wearable Devices for Health Informatics , 2014, IEEE Transactions on Biomedical Engineering.

[5]  Ron Kohavi,et al.  Wrappers for Feature Subset Selection , 1997, Artif. Intell..

[6]  Roberto Gil-Pita,et al.  Synthetical Enlargement of MFCC Based Training Sets for Emotion Recognition , 2014, FOCS 2014.

[7]  Fernando Seoane,et al.  Wearable Biomedical Measurement Systems for Assessment of Mental Stress of Combatants in Real Time , 2014, Sensors.

[8]  Fernando Seoane,et al.  Activity Recognition Using Wearable Physiological Measurements: Selection of Features from a Comprehensive Literature Study , 2019, Sensors.

[9]  Li Zhuo,et al.  A genetic algorithm based wrapper feature selection method for classification of hyperspectral images using support vector machine , 2008, Geoinformatics.

[10]  Jozef Juhar,et al.  Real-time emotional state detection from facial expression on embedded devices , 2017, 2017 Seventh International Conference on Innovative Computing Technology (INTECH).

[11]  E. Gabazza,et al.  Smart watch-based coaching with tiotropium and olodaterol ameliorates physical activity in patients with chronic obstructive pulmonary disease , 2017, Experimental and therapeutic medicine.

[12]  Fernando Seoane,et al.  Assessment of Mental, Emotional and Physical Stress through Analysis of Physiological Signals Using Smartphones , 2015, Sensors.

[13]  A. Loewy,et al.  Central Command Neurons of the Sympathetic Nervous System: Basis of the Fight-or-Flight Response , 1995, Science.

[14]  Hari Singh Dhillon THE EFFECT OF EMOTIONS ON ELECTROCARDIOGRAM , 2011 .

[15]  C. Calhoun,et al.  What is an emotion? : classic readings in philosophical psychology , 1984 .

[16]  V Marigliano,et al.  Power spectral analysis of heart rate variability in obese subjects: evidence of decreased cardiac sympathetic responsiveness. , 1996, International journal of obesity and related metabolic disorders : journal of the International Association for the Study of Obesity.

[17]  Ron Kohavi,et al.  Feature Subset Selection Using the Wrapper Method: Overfitting and Dynamic Search Space Topology , 1995, KDD.

[18]  Alireza Mehrnia,et al.  On optimal IFIR filter design , 2004, 2004 IEEE International Symposium on Circuits and Systems (IEEE Cat. No.04CH37512).

[19]  D Gianola,et al.  Dimension reduction and variable selection for genomic selection: application to predicting milk yield in Holsteins. , 2011, Journal of animal breeding and genetics = Zeitschrift fur Tierzuchtung und Zuchtungsbiologie.

[20]  Andrew Kusiak,et al.  Data mining and genetic algorithm based gene/SNP selection , 2004, Artif. Intell. Medicine.

[21]  F Babiloni,et al.  Passive BCI beyond the lab: current trends and future directions , 2018, Physiological measurement.

[22]  Jin Wang,et al.  Weight smoothing to improve network generalization , 1994, IEEE Trans. Neural Networks.

[23]  A. Rasin,et al.  Using Real-Time Social Media Technologies to Monitor Levels of Perceived Stress and Emotional State in College Students: A Web-Based Questionnaire Study , 2017, JMIR mental health.

[24]  Guang-yuan Liu,et al.  Feature Extraction, Feature Selection and Classification from Electrocardiography to Emotions , 2009, 2009 International Conference on Computational Intelligence and Natural Computing.

[25]  Jennifer Healey,et al.  Detecting stress during real-world driving tasks using physiological sensors , 2005, IEEE Transactions on Intelligent Transportation Systems.

[26]  Yong-Jin Liu,et al.  Real-Time Movie-Induced Discrete Emotion Recognition from EEG Signals , 2018, IEEE Transactions on Affective Computing.

[27]  Xiyao Xie,et al.  Mindful attention to breath regulates emotions via increased amygdala–prefrontal cortex connectivity , 2016, NeuroImage.

[28]  Xiaoming Xu,et al.  A hybrid genetic algorithm for feature selection wrapper based on mutual information , 2007, Pattern Recognit. Lett..

[29]  De-Shuang Huang,et al.  A mended hybrid learning algorithm for radial basis function neural networks to improve generalization capability , 2007 .

[30]  R. Mccraty,et al.  The effects of emotions on short-term power spectrum analysis of heart rate variability . , 1995, The American journal of cardiology.

[31]  Hongwei Zhu,et al.  Recent advances in wearable tactile sensors: Materials, sensing mechanisms, and device performance , 2017 .

[32]  Branko Jakovljević,et al.  Thoracic bioelectrical impedance assessment of the hemodynamic reactions to recorded road-traffic noise in young adults. , 2014, Environmental research.

[33]  S. Yoo,et al.  Using Neural Network to Recognize Human Emotions from Heart Rate Variability and Skin Resistance , 2005, 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference.

[34]  G. Billman The LF/HF ratio does not accurately measure cardiac sympatho-vagal balance , 2013, Front. Physio..

[35]  I. Homma,et al.  Breathing rhythms and emotions , 2008, Experimental physiology.

[36]  R. Schandry,et al.  Heartbeat detection and the experience of emotions , 2009 .

[37]  D. Anguita,et al.  K-fold generalization capability assessment for support vector classifiers , 2005, Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005..

[38]  John A Updegraff,et al.  Biobehavioral responses to stress in females: tend-and-befriend, not fight-or-flight. , 2000, Psychological review.

[39]  Eamonn J. Keogh,et al.  Curse of Dimensionality , 2010, Encyclopedia of Machine Learning.