ETRI-Knowledge Sharing Plaform

ENGLISH

성과물

논문 검색
구분 SCI
연도 ~ 키워드

상세정보

학술지 Label-preserving Data Augmentation for Mobile Sensor Data
Cited 19 time in scopus Download 56 time Share share facebook twitter linkedin kakaostory
저자
김무섭, 정치윤
발행일
202101
출처
Multidimensional Systems and Signal Processing, v.32 no.1, pp.115-129
ISSN
0923-6082
출판사
Springer
DOI
https://dx.doi.org/10.1007/s11045-020-00731-2
협약과제
20HS1300, 신체기능의 이상이나 저하를 극복하기 위한 휴먼 청각 및 근력 증강 원천 기술 개발, 신형철
초록
Data augmentation is important for training neural networks, especially when there is not enough data to train a network well. However, data augmentation that results in the loss of label information may reduce the performance of the model. Most conventional data augmentation methods have been developed for image- or sound-related tasks, in which case the label information of the augmented data is easily and intuitively verified by human observation. However, in the case of sensor signals, it is difficult to recognize whether there is a change in the label information of the augmented data. We propose a systematic data augmentation method to maximize the performance by automatically finding the range of augmentation that preserves the labels of the augmented data. The experimental results show that the proposed method to extract the label-preserving range is practical and that the retrained model using data augmented within this range improves the performance by at least 5% without the need to further optimize the model architecture.
KSP 제안 키워드
Augmentation method, Data Augmentation, Label information, Model architecture, mobile sensor data, neural network