A computer implementation method for detecting unfair transactions using local sharp hash and local outlier factor algorithm

摘要

A method and system for detecting an illegal data point in a database of computerized systems from a user device Includes receiving a request for detecting one or more illegal data points in a database that stores data points representing electronic transactions.The system constructs feature vectors and K random vectors for each data point.After the construction, the system performs a combination of the local feature type hash algorithm and the local outlier factor (LOF) algorithm for the constructed feature vector and the constructed K random vectors to calculate the outlier score of each data point.The system detects an illegal data point based on an outlier score indicating the degree of consistency.