METHOD FOR RAPID SCREENING OF OUTLIER DATA FROM LARGE-SCALE DATA

摘要

Provided is a method for rapid screening of outlier data from large-scale data; the features of the computation of time and space complexity in large-scale data outlier mining are fully taken into consideration; random sampling is used to reduce the quantity of samples participating in the computation, and parallel computing is used to increase computing speed, thereby effectively solving the problem of higher computing-time and memory space requirements when screening outliers of large-scale data, and thus accomplishing rapid and effective screening of outlier data.