METHOD, APPARATUS AND COMPUTER PROGRAM FOR DETECTING DEVIATIONS IN DATA SOURCES

摘要

A method and apparatus is disclosed for detecting deviations in data repositories, each data repository comprising a plurality of data posts, each data post comprising a number of data attribute values. The method comprises identifying (102) comparable data post pairs, each pair comprising a first data post in a first data repository and a second data post in a second data repository, wherein the first data post in a comparable data post pair is comparable with the second data post of the same data post pair. The method further comprises comparing (104) data attribute values of the first data post with data attribute values of the second data post within a plurality of comparable data post pairs in order to determine quantified similarities between the data attribute values of the first data post and the data attribute values of the second data post of each of the plurality of comparable data post pair. The method further comprises calculating (106) statistical values of the quantified similarities by comparing comparable determined quantified similarities for each of the plurality of comparable data post pairs, and using (108) the calculated statistical values of the quantified similarities for detecting deviations for a first comparable data post pair.