Big data mining method and device and data mining server

Abstract

The embodiment of the invention provides a big data mining method and device and a data mining server. Different interestingness measurement dimensions are comprehensively considered, after clusteringall the business big data of each dimension, it can be ensured that the performance of different interestingness measurement dimensions is more unified under different to-be-applied scenes of the data mining project. The big data mining capability is improved. According to the feature information of each clustering cluster, multiple data mining items of the to-be-mined service and the to-be-mined data dimension corresponding to each data mining item are dynamically determined, and subsequent data mining is performed according to the data mining items and the to-be-mined data dimension, so that the problem of relatively poor mining effect or low accuracy of a data mining result possibly caused by adoption of a fixed mining data dimension in the prior art can be avoided.

Description

technical field [0001] The present application relates to the technical field of big data, in particular, to a big data mining method, device and data mining server. Background technique [0002] At present, most of the big data mining schemes for various online businesses (such as order behavior business, browsing behavior business, etc.) only use one measurement dimension of interest degree, although some studies focus on the attributes and behaviors of different measurement dimensions of interest degree , but for a service to be mined, different interest degree measurement dimensions have different performances in different data mining project application scenarios, and its use limitations limit the ability in big data mining. Moreover, in the entire data mining process, most of them use fixed mining data dimensions for mining. However, when there are too many dimensions of mining data, these fixed mining data dimensions may not be able to achieve a better mining effect,...

AI Technical Summary

Problems solved by technology

[0003] In order to at least overcome the above-mentioned deficiencies in the prior art, one of the purposes of this application is to provide a big data mining method, device, and data mining server. By comprehensively considering different measurement dimensions of interest degree, the large After the data is clustered, it can ensure that the performance of different interest degree measurement dimensions is more uniform in different data mining project application scenarios, improve the ability of big data mining, and dynamically determine the data to be mined according to the characteristic information of each cluster cluster. Multiple data mining projects of the service and the data dimension to be mined corresponding to each data mining project, and subsequent data mining based on this can avoid poor mining effects or data mining results that may occur in the existing technology using fixed mining data dimensions. There is a problem of low accuracy

Images