Method and system for discriminating abnormal data

Abstract

The embodiment of the invention discloses a method for discriminating abnormal data, which is realized on a plurality of sample data sets. The method comprises the following steps: acquiring a first sample data set and a plurality of corresponding types of service; setting screening rules in each corresponding type of service and obtaining screened data of each type of service according to the set screening rules; judging whether the screened data of each type of service exist in screened comparison data sets of the plurality of sample data sets except the first sample data set or not; if so, determining the screened data as abnormal data. According to the embodiment of the invention, the problem that analysis results are greatly different due to deviation existing in the sampling process can be corrected, the method can be used for sampling of a plurality of subtypes and the error rate of sampling results is reduced; meanwhile, all abnormal data in complex big data (data aggregate instead of sample sets only) can be quickly and accurately locked.

Description

technical field [0001] The invention relates to the technical field of power system marketing inspection, in particular to a method and system for identifying abnormal data. Background technique [0002] Power system marketing inspection is based on relevant policies, regulations and rules and regulations, and conducts internal professional inspection and supervision on the construction and implementation of marketing systems, marketing behavior norms, and marketing work quality. [0003] The existing normalized marketing inspection work system, based on the scientific sampling and evaluation model of statistical principles, first imports business data into statistical software, then conducts sampling through the sampling module of general statistical software, and finally imports the survey result data into statistical The software performs statistical inference. Therefore, when all data cannot be collected or analyzed, high-precision inferences can be made with less cost b...

AI Technical Summary

Problems solved by technology

[0003] The existing normalized marketing inspection work system, based on the scientific sampling and evaluation model of statistical principles, first imports business data into statistical software, then conducts sampling through the sampling module of general statistical software, and finally imports the survey result data into statistical The software performs statistical inference. Therefore, when all data cannot be collected or analyzed, high-precision inferences can be made with less cost by collecting random samples. The disadvantage is that once there is any deviation in the sampling process, the analysis results will be There will be a large error, and when random sampling is used for sampling of multiple subcategories, the error rate of random sampling results will greatly increase

[0004] At the same time, when the business data increases in a large amount, the method of finding abnormal data through sampling survey has the problem that all abnormal data cannot be found and the search efficiency is low, that is, abnormal data cannot be quickly locked in complex big data

Images