Abstract: The huge numbers of users participating in online auctions and the low cost of creating accounts for such activities have increased the danger of fraud and other criminal endeavors in these environments and have pushed the monetary and financial institutions to seek efficient and quick solutions to detect such offenses. This issue has necessitated the use of fraud detection techniques to prevent fraudulent endeavors in banking and especially electronic banking systems. The objective of the present study was to propose a hybrid approach to detect the fraudulent accounts. This objective was pursued by analyzing the social networks to produce behavioral characteristics and then turning these characteristics into fuzzy rules. The fuzzy rules proposed by the genetic algorithm were then optimized for auction fraud detection model. The introduction of fuzzy crossover and mutation operators specifically modified for this objective was the other contribution of this research to the literature. The results obtained by the implementation of the proposed system showed that using these fuzzy crossover and mutation operators improved the algorithm performance and the speed by which algorithm obtained the optimal solution.
Mohammad Reza Parsaei, Reza Javidan and Mohammad Javad Sobouti, 2016. Optimization of Fuzzy Rules for Online Fraud Detection with the Use of Developed Genetic Algorithm and Fuzzy Operators. Asian Journal of Information Technology, 15: 1856-1864.