The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Vues en texte intégral
79
Cette étude se concentre sur la détection automatisée de la condition d'un opérateur de système complexe. Par exemple, dans cette étude, la réaction d'une personne lorsqu'elle écoutait de la musique (ou ne l'écoutait pas du tout) a été déterminée. À cette fin, divers outils d’exploration de données bien connus ainsi que ceux développés par les auteurs ont été utilisés. Pour être plus précis, les techniques suivantes ont été développées et appliquées aux problèmes mentionnés : les réseaux de neurones artificiels et les classificateurs flous basés sur des règles. Les réseaux de neurones ont été générés par deux modifications de l'algorithme Differential Evolution basé sur les schémas NSGA et MOEA/D, proposés pour résoudre des problèmes d'optimisation multi-objectifs. Les systèmes de logique floue ont été générés par l'algorithme basé sur la population appelé Co-Operation of Biology Related Algorithms ou COBRA. Cependant, l'état de chaque personne a d'abord été surveillé. Ainsi, les bases de données sur les problèmes décrits dans cette étude ont été obtenues en utilisant des capteurs Doppler sans contact. Les résultats expérimentaux ont démontré que les réseaux neuronaux générés automatiquement et les classificateurs flous basés sur des règles peuvent déterminer correctement la condition et la réaction humaines. En outre, les approches proposées ont surpassé les outils alternatifs d’exploration de données. Cependant, il a été établi que les classificateurs flous basés sur des règles sont plus précis et interprétables que les réseaux de neurones. Ainsi, ils peuvent être utilisés pour résoudre des problèmes plus complexes liés à la détection automatisée de l’état d’un opérateur.
Shakhnaz AKHMEDOVA
Reshetnev Siberian State University of Science and Technology
Vladimir STANOVOV
Reshetnev Siberian State University of Science and Technology
Sophia VISHNEVSKAYA
Reshetnev Siberian State University of Science and Technology
Chiori MIYAJIMA
Aichi Prefectural University
Yukihiro KAMIYA
Aichi Prefectural University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Shakhnaz AKHMEDOVA, Vladimir STANOVOV, Sophia VISHNEVSKAYA, Chiori MIYAJIMA, Yukihiro KAMIYA, "Automatically Generated Data Mining Tools for Complex System Operator's Condition Detection Using Non-Contact Vital Sensing" in IEICE TRANSACTIONS on Communications,
vol. E104-B, no. 6, pp. 571-579, June 2021, doi: 10.1587/transcom.2020HMI0001.
Abstract: This study is focused on the automated detection of a complex system operator's condition. For example, in this study a person's reaction while listening to music (or not listening at all) was determined. For this purpose various well-known data mining tools as well as ones developed by authors were used. To be more specific, the following techniques were developed and applied for the mentioned problems: artificial neural networks and fuzzy rule-based classifiers. The neural networks were generated by two modifications of the Differential Evolution algorithm based on the NSGA and MOEA/D schemes, proposed for solving multi-objective optimization problems. Fuzzy logic systems were generated by the population-based algorithm called Co-Operation of Biology Related Algorithms or COBRA. However, firstly each person's state was monitored. Thus, databases for problems described in this study were obtained by using non-contact Doppler sensors. Experimental results demonstrated that automatically generated neural networks and fuzzy rule-based classifiers can properly determine the human condition and reaction. Besides, proposed approaches outperformed alternative data mining tools. However, it was established that fuzzy rule-based classifiers are more accurate and interpretable than neural networks. Thus, they can be used for solving more complex problems related to the automated detection of an operator's condition.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.2020HMI0001/_p
Copier
@ARTICLE{e104-b_6_571,
author={Shakhnaz AKHMEDOVA, Vladimir STANOVOV, Sophia VISHNEVSKAYA, Chiori MIYAJIMA, Yukihiro KAMIYA, },
journal={IEICE TRANSACTIONS on Communications},
title={Automatically Generated Data Mining Tools for Complex System Operator's Condition Detection Using Non-Contact Vital Sensing},
year={2021},
volume={E104-B},
number={6},
pages={571-579},
abstract={This study is focused on the automated detection of a complex system operator's condition. For example, in this study a person's reaction while listening to music (or not listening at all) was determined. For this purpose various well-known data mining tools as well as ones developed by authors were used. To be more specific, the following techniques were developed and applied for the mentioned problems: artificial neural networks and fuzzy rule-based classifiers. The neural networks were generated by two modifications of the Differential Evolution algorithm based on the NSGA and MOEA/D schemes, proposed for solving multi-objective optimization problems. Fuzzy logic systems were generated by the population-based algorithm called Co-Operation of Biology Related Algorithms or COBRA. However, firstly each person's state was monitored. Thus, databases for problems described in this study were obtained by using non-contact Doppler sensors. Experimental results demonstrated that automatically generated neural networks and fuzzy rule-based classifiers can properly determine the human condition and reaction. Besides, proposed approaches outperformed alternative data mining tools. However, it was established that fuzzy rule-based classifiers are more accurate and interpretable than neural networks. Thus, they can be used for solving more complex problems related to the automated detection of an operator's condition.},
keywords={},
doi={10.1587/transcom.2020HMI0001},
ISSN={1745-1345},
month={June},}
Copier
TY - JOUR
TI - Automatically Generated Data Mining Tools for Complex System Operator's Condition Detection Using Non-Contact Vital Sensing
T2 - IEICE TRANSACTIONS on Communications
SP - 571
EP - 579
AU - Shakhnaz AKHMEDOVA
AU - Vladimir STANOVOV
AU - Sophia VISHNEVSKAYA
AU - Chiori MIYAJIMA
AU - Yukihiro KAMIYA
PY - 2021
DO - 10.1587/transcom.2020HMI0001
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E104-B
IS - 6
JA - IEICE TRANSACTIONS on Communications
Y1 - June 2021
AB - This study is focused on the automated detection of a complex system operator's condition. For example, in this study a person's reaction while listening to music (or not listening at all) was determined. For this purpose various well-known data mining tools as well as ones developed by authors were used. To be more specific, the following techniques were developed and applied for the mentioned problems: artificial neural networks and fuzzy rule-based classifiers. The neural networks were generated by two modifications of the Differential Evolution algorithm based on the NSGA and MOEA/D schemes, proposed for solving multi-objective optimization problems. Fuzzy logic systems were generated by the population-based algorithm called Co-Operation of Biology Related Algorithms or COBRA. However, firstly each person's state was monitored. Thus, databases for problems described in this study were obtained by using non-contact Doppler sensors. Experimental results demonstrated that automatically generated neural networks and fuzzy rule-based classifiers can properly determine the human condition and reaction. Besides, proposed approaches outperformed alternative data mining tools. However, it was established that fuzzy rule-based classifiers are more accurate and interpretable than neural networks. Thus, they can be used for solving more complex problems related to the automated detection of an operator's condition.
ER -