Affiliation:
1. Kentkart Ege Elektronik
2. DOKUZ EYLÜL ÜNİVERSİTESİ, İKTİSADİ VE İDARİ BİLİMLER FAKÜLTESİ, EKONOMETRİ BÖLÜMÜ
Abstract
With the developing technology, mobile payment systems have become increasingly popular. In the public transport industry, this system has convenient to the sector in terms of purchasing, using, carrying and storing tickets. One of the greatest challenges encountered in the mobile payment system in this sector is fraud. Fraud reduces customer satisfaction, reduces snow margins and causes severe costs for the company. Therefore, it is very important to detect and prevent fraudsters. This study is based on users using a real mobile ticketing application in USA/Kansas, a customer of Kentkart, which has a smart public transportation system. An automatic and intelligent detection system was developed using a machine learning algorithm to detect whether the users in question are fraudulent or not. For this system, the historical profiles of the variables that represent a user that the risky behavior are created. These profiles are classified using Random Forest, Support Vector Machines, Logistic Regression, K-Nearest Neighbor and Naive Bayes machine learning techniques and results are combined with simple ensemble learning methods. Users classified as frauds are automatically blacklisted in accordance with the company's management policy. Thus, the fraud costs that these users caused the company have been reduced.
Publisher
Bulent Evcevit University
Reference44 articles.
1. A Liaw and M Wiener. Classification and regression... - Google Akademik. (n.d.). Retrieved July 28, 2021, from https://scholar.google.com.tr/scholar?hl=tr&as_sdt=0%2C5&q=A+Liaw++and+M+Wiener.+Classification+and+regression+by+randomForest.+R+news.+2002%3B+2%283%29%2C+18-22.&btnG=
2. Abdallah, A., Maarof, M. A., & Zainal, A. (2016). Fraud detection system: A survey. Journal of Network and Computer Applications, 68, 90–113.
3. Abe, S. (2005). Support vector machines for pattern classification (Vol. 2). Springer.
4. Adewumi, A. O., & Akinyelu, A. A. (2017). A survey of machine-learning and nature-inspired based credit card fraud detection techniques. International Journal of System Assurance Engineering and Management, 8(2), 937–953.
5. Aras, S., & Gulay, E. (2017). A new consensus between the mean and median combination methods to improve forecasting accuracy. Serbian Journal of Management, 12(2), 217–236.