作者: Zakia Ferdousi , Akira Maeda
DOI:
关键词:
摘要: Fraud detection is of great importance to financial institutions. This paper concerned with the problem finding outliers in time series data using Peer Group Analysis (PGA), which an unsupervised technique for fraud detection. The objective PGA characterize expected pattern behavior around target sequence terms similar objects, and then detect any difference evolution between target. tool has been applied stock market data, collected from Bangladesh Stock Exchange assess its performance We observed can those brokers who suddenly start selling a different way other whom they were previously similar. also t-