Big data platform for constructing customer portrait and construction method

Abstract

The invention provides a big data platform for constructing a customer portrait. The platform comprises a data access layer for accessing client power grid data, a data storage layer used for providing corresponding storage technology according to the use characteristics of client power grid data; a data computing layer used for providing a data processing and computing engine for customer power grid data; a data analysis layer used for carrying out statistical retrieval analysis on the customer power grid data; and an data application layer used for providing data service for the data accesslayer, the data storage layer, the data computing layer and the data analysis layer. From the perspective of client power grid data, a large number of small files of monitoring data are combined intoa large sequence file and are compressed and stored, label information of the client power grid data can be combined, and a client portrait is formed through a large number of logical operations.

Description

Technical field [0001] The present invention relates to a big data platform, and more specifically, to a big data platform and a construction method for constructing customer portraits. Background technique [0002] In recent years, in the power industry, the rapid development of strong smart grids has enabled information and communication technology to integrate with grid production and enterprise management with unprecedented breadth and depth. Information and communication systems have become the "central nervous system" of smart grids, supporting the new generation of grids. Production and management development. [0003] At present, there are many types of power marketing customer data, and the fragmented data information is distributed in multiple information systems. Marketing personnel need to filter effective information from the complex business system, which requires a large workload and high requirements for the business personnel's business level. In addition, due to ...

AI Technical Summary

Problems solved by technology

[0003] At present, there are many types of customer data in electric power marketing, and the data information is fragmented and distributed in multiple information systems. Marketing personnel need to screen effective information from the complex business systems, with a heavy workload and high requirements for the professional level of business personnel.

In addition, due to the lack of theoretical guidance for classification methods, customer label categories are extensively expanded according to business needs, which brings more and more problems for the establishment of subsequent label models, label calculation efficiency, label management and use

Images