ICFICE 2018
Current Result Document :
ÇѱÛÁ¦¸ñ(Korean Title) |
Comparison of SVM and RF Classifiers for Stress and Fatigue Recognition |
¿µ¹®Á¦¸ñ(English Title) |
Comparison of SVM and RF Classifiers for Stress and Fatigue Recognition |
ÀúÀÚ(Author) |
Su-Jin Seong
Seong-Jae Park
Tae-Ho Park
Chang-Uk Shin
Da-Sol Park
Jeong-MooKim
Jeong-Won Cha
Yungi Park
Yongsoo Park
Youn-Sung Lee
Jeongwook Seo
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 10 NO. 01 PP. 0257 ~ 0259 (2018. 06) |
Çѱ۳»¿ë (Korean Abstract) |
|
¿µ¹®³»¿ë (English Abstract) |
If the stress and fatigue in industrial workers becomes high when they are doing work, this may lead to a decrease in productivity and industrial accidents. In order to prevent the decrease in productivity and industrial accidents, this paper presents a stress and fatigue recognition (SFR) system. The proposed SFR system collects the biosignal and environmental data related to an industrial worker through an one-M2M compliant platform and then it classifies three states of industrial workers: working, stress and fatigue by using the collected data. For the classification of the states, we exploited Support Vector Machine (SVM) and Random Forest (RF) algorithms and compared their accuracy performances. Experimental results showed that the SVM classifier achieved 89% accuracy and the RF classifier achieved 98% accuracy.
|
Å°¿öµå(Keyword) |
respiratory diseases
SVM
Random forest
DNN
Ensemble
Fatigue
Industrial IoT
Random Forest
Recognition
Stress
SVM
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|