Power customer multi-incoming call service risk early warning method

Abstract

The invention relates to a power customer multi-incoming call service risk early warning method. The implementation process of the method comprises the following steps: firstly, acquiring 95598 service caller order data; analyzing the caller order data of each customer; by using XGBoost and Logistic regression modeling, the probability that a user calls again in the next three days is predicted in advance, and a client with a high probability of calling again is early warned in advance. Compared with the prior art, the multiple-incoming-call service risk early-warning method can reduce service risk events, accurately position risk reasons, improve the early-warning capability of an electric power customer service center for customer complaints, remarkably improve the service response speed and improve the service quality. And technical guarantee is provided for reducing service cost of an electric customer service center and improving enterprise management level.

Description

Technical field: [0001] The invention relates to a big data analysis technology, in particular to a method for early warning of service risks for power customers with repeated calls. Background technique: [0002] With the steady and rapid development of the national economy, the quality of residents has been significantly improved, and there are higher requirements for electricity services. In the current situation of diversified and convenient information exchange channels, electricity customers have become accustomed to using the 95598 service hotline To reflect customer demands. At the same time, residents' awareness of rights protection has increased. When customers have reported the same problem many times and have not met customer expectations, they will turn to solve their own problems through complaints, and there are greater service risks. After analysis, about 90% of the escalation complaints are multiple calls. Therefore, controlling multiple calls plays an imp...

AI Technical Summary

Problems solved by technology

[0004] Difficult to predict in advance: Service risks occur together and are dealt with together, without analyzing and judging the reasons for possible re-calls that may lead to service risks, it is difficult to avoid in advance, and the effect of improving customer perception is limited

[0005] Difficulty in locating during the event: Customer information with service risks has the characteristics of multi-channel and multi-dimensional. Only relying on information from a single channel and a single dimension is not enough to quickly locate the cause of the risk. It needs to use multiple channels and integrate multiple dimensions. to locate the problem

[0006] Difficult post-event analysis: The analysis of customer trajectories for multiple calls is relatively complicated, and screening out service risk events requires a very large workload, which makes this work impossible to carry out effectively

Images