http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
금융거래 정보 및 모바일 GPS를 활용한 모바일 이상 거래 탐지 시스템 구현
윤규성(Gyu-Seong Yun),박수환(Soo-Hwan Park),이광재(Kwangjae Lee) 한국디지털콘텐츠학회 2022 한국디지털콘텐츠학회논문지 Vol.23 No.10
Recently, as the number of users of mobile payment increases, cases of money theft based on mobile payment are also increasing. Currently, the anormal transaction detection system is optimized for general transactions, so it is insufficient to apply to mobile transactions. In this paper, we propose a mobile abnormal transaction detection system using machine learning techniques. In addition, since mobile payment can be used without space restrictions, location information analysis using this is proposed. The evaluation indicators were Accuracy, Recall, Precision, and F1 Score, which showed high performance with 0.99, 0.91, 0.95, and 0.93, respectively. And to evaluate the predictive power of the diagnostic method, AUC (area under the ROC curve) was used, and the result was 0.98. As a result of performing machine learning by the method proposed in this paper, all performance indicators showed a high detection rate of 0.93 or higher, and the reliability of anormal transaction detection was improved by using the location information of mobile devices.