http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Data Augmentation on Limited Biometric Data Set for M2M Authentication Model Testing
Rin Nadia,Dana Koshen,JaeSeung Song 한국통신학회 2021 한국통신학회 학술대회논문집 Vol.2021 No.6
Examining the performance of artificial intelligence (AI)-based model for machine to machine (M2M) authentication needs a large number of data. The accessibility of open biometric data is restricted by privacy regulations such as general data protection regulation (GDPR). This is because the data is commonly obtained by sensors embedded in personal wearable devices. Thus exploring and developing a system with biometric data as the main parameter is difficult to do. As data augmentation is a technique to increase the size of the data set by producing derived data from the original data, its usage in AI is popular especially for computer vision. Incorporating data augmentation in the development of AI-based authentication model could solve the shortage of database. Therefore this study proposes a data augmentation model and analyzes its training and augmenting performance.
Rin Nadia,Dana Koshen,JaeSeung Song 한국통신학회 2020 한국통신학회 학술대회논문집 Vol.2020 No.11
Internet of Things (IoT) user authentication is an essential part to keep and IoT ecosystem from an intruder. Due to this motivation, this authentication system has drawn researchers’ interest. Light, simple and pleasant in user experience are the ideal criteria for the authentication system. Compare to non-biometric security identifier, the biometric identifier naturally has characteristics to build a strong but convenient personal authentication system. However, the typical biometric system employs template matcher to verify user. For certain biometric data which is prone to deviation of biometric features caused by body changing, the template matcher can not learn the pattern alteration. As an application of Artificial Intelligence, machine learning is designed to get this shifting pattern. The usage of machine learning as template matcher substitute could create a flexible, secure but simple authentication system. Human impedance is one of the biometric data and its usability as authentication factor has been proven in several studies. Therefore, by conducting the experiment involving twenty five machine learning algorithms, this work proposes best binary classifier candidates for user verification in IoT ecosystem using human impedance.