This paper investigates the structure of the payment card market, with consumers and merchants basing their subscription decisions on different information sets. We find that the market structure depends crucially on ...This paper investigates the structure of the payment card market, with consumers and merchants basing their subscription decisions on different information sets. We find that the market structure depends crucially on the information set on which consumers and merchants base their subscription decisions. In the studied case, we observe that a market with few cards dominating only emerges when decisions are based on very limited information. Under the same conditions using a complete information set, all cards survive in the long run. The use of an agent-based model, focusing on the interactions between merchants and consumers, as a basis for subscription decisions allows us to investigate the dynamics of the market and the effect of the indirect network externalities rather than investigating only equilibrium outcomes.展开更多
基于对语音、视频处理实时性和稳定性的高要求,以可编程数字信号处理器(Digital Signal Processor,DSP)为主体进行研究,通过分析DSP处理器的系统总体,介绍系统中不同环节的作用,设计出可以有效提高数据传输实时性和稳定性的数字信号处...基于对语音、视频处理实时性和稳定性的高要求,以可编程数字信号处理器(Digital Signal Processor,DSP)为主体进行研究,通过分析DSP处理器的系统总体,介绍系统中不同环节的作用,设计出可以有效提高数据传输实时性和稳定性的数字信号处理系统,以提高语音和视频处理效率,助力企业实现可持续发展。展开更多
Credit card fraud is a wide-ranging issue for financial institutions, involving theft and fraud committed using a payment card. In this paper, we explore the application of linear and nonlinear statistical modeling an...Credit card fraud is a wide-ranging issue for financial institutions, involving theft and fraud committed using a payment card. In this paper, we explore the application of linear and nonlinear statistical modeling and machine learning models on real credit card transaction data. The models built are supervised fraud models that attempt to identify which transactions are most likely fraudulent. We discuss the processes of data exploration, data cleaning, variable creation, feature selection, model algorithms, and results. Five different supervised models are explored and compared including logistic regression, neural networks, random forest, boosted tree and support vector machines. The boosted tree model shows the best fraud detection result (FDR = 49.83%) for this particular data set. The resulting model can be utilized in a credit card fraud detection system. A similar model development process can be performed in related business domains such as insurance and telecommunications, to avoid or detect fraudulent activity.展开更多
文摘This paper investigates the structure of the payment card market, with consumers and merchants basing their subscription decisions on different information sets. We find that the market structure depends crucially on the information set on which consumers and merchants base their subscription decisions. In the studied case, we observe that a market with few cards dominating only emerges when decisions are based on very limited information. Under the same conditions using a complete information set, all cards survive in the long run. The use of an agent-based model, focusing on the interactions between merchants and consumers, as a basis for subscription decisions allows us to investigate the dynamics of the market and the effect of the indirect network externalities rather than investigating only equilibrium outcomes.
文摘基于对语音、视频处理实时性和稳定性的高要求,以可编程数字信号处理器(Digital Signal Processor,DSP)为主体进行研究,通过分析DSP处理器的系统总体,介绍系统中不同环节的作用,设计出可以有效提高数据传输实时性和稳定性的数字信号处理系统,以提高语音和视频处理效率,助力企业实现可持续发展。
文摘Credit card fraud is a wide-ranging issue for financial institutions, involving theft and fraud committed using a payment card. In this paper, we explore the application of linear and nonlinear statistical modeling and machine learning models on real credit card transaction data. The models built are supervised fraud models that attempt to identify which transactions are most likely fraudulent. We discuss the processes of data exploration, data cleaning, variable creation, feature selection, model algorithms, and results. Five different supervised models are explored and compared including logistic regression, neural networks, random forest, boosted tree and support vector machines. The boosted tree model shows the best fraud detection result (FDR = 49.83%) for this particular data set. The resulting model can be utilized in a credit card fraud detection system. A similar model development process can be performed in related business domains such as insurance and telecommunications, to avoid or detect fraudulent activity.
