Method and system for generating and managing portrait of electric power customer

Abstract

The invention provides a method and a system for generating and managing a portrait of an electric power customer. The method mainly comprises the following steps: obtaining a refined feature mark through a big data processing technology and a machine learning algorithm, and establishing a dynamic label model; calling the dynamic label model to generate a dynamic label and starting the whole lifecycle management of the label; automatically adding new labels to existing labels, and recalculating and displaying the coverage condition of all the labels and visually displaying the distribution ofall the labels among the total customers; in combination with the existing labels, designing the customer clustering rules under the new business scenario, and carrying out multi-dimensional analysis; in combination with business experiences and the original relevant strategies, developing a differentiated service strategy; connecting the business system to the outside world, pushing the formulated differentiated information push strategies for different channels. The invention constructs a panoramic portrait of a customer in the form of a label, and realizes fine depiction of the characteristics of the electric power customer, and based on this, the customer clustering rules are designed to develop the corresponding differentiated marketing management strategy.

Description

Technical field [0001] The invention relates to the technical field of customer management in the electric power industry, in particular to a method and system for generating and managing electric customer portraits. Background technique [0002] At present, customer grouping in the business system of power grid companies is based on business experience or simple logical rules. For example, power users are divided into important customers, focused customers, large customers, residential customers, and other customers according to customer importance; For military unit customers, institutional customers, corporate customers, individual customers and group customers. This power customer grouping method has obvious shortcomings, such as low grouping accuracy and low efficiency of application of strategies formulated for grouping. [0003] In recent years, as the pressure of internal and external competition in the power grid has increased, business personnel and data analysts have ca...

AI Technical Summary

Problems solved by technology

This power customer grouping method has obvious shortcomings such as low grouping accuracy and low application efficiency of strategies formulated for grouping.

[0003] In recent years, with the increasing pressure of internal and external competition in the power grid, business personnel and data analysts have carried out a large amount of preliminary data exploration and analysis to assist management decision-making in the fields of customer precision marketing, customer demand forecasting, and customer service quality, and have achieved certain results. However, these analysis results are too modular and can only reflect the characteristics of a certain aspect of customers. There is still a certain distance from providing real-time, accurate and in-depth customer portraits and service strategies for business personnel.

Images