Power customer demand prediction analysis method and system
A technology of customer demand, predictive analysis, applied in forecasting, instruments, data processing applications, etc., can solve the problems of customer service agents not being able to quickly, accurately and comprehensively understand customer needs, large customer base, and difficulty in responding to customer demands in a timely manner.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
example 1
[0139] Incoming user information is shown in Table 1-1, where 2 means high, 1 means medium, and 0 means low.
[0140] Table 1-1
[0141]
[0142] (1) First of all, based on the published planned power outage notice, it is judged that this user is a customer affected by a power outage event, meets the conditions of step 1, and the acceptance time is 6 minutes different from the time of the latest power outage event; according to different time points within 60 minutes of the power outage event obtained through statistical calculations The proportion of each business is shown in Table 1-2.
[0143] Table 1-2
[0144]
[0145] (2) Establish the feature set Z of the service demand prediction model for user Zhang San’s power outage affecting the customer’s incoming calls 1 ={x 11 ,x 12 ,x 13 ......, x 1n}, as shown in Table 1-3.
[0146] Table 1-3
[0147]
[0148]
[0149] (3) Use this feature set to predict customer Zhang San's demand for incoming calls by usi...
example 2
[0154] Incoming user information is shown in Table 2-1, where 2 means high, 1 means medium, and 0 means low.
[0155] table 2-1
[0156]
[0157] (1) First of all, it is judged by the issued planned power outage notice that this user is not affected by the power outage event; but according to the station area where Li Si is located and the acceptance time, it is judged that there were two reports of power outages in the area in the first 15 minutes;
[0158] According to the statistics and calculations, the proportion of incoming calls of each service at different time points within 60 minutes of the reported power outage in the same area is shown in Table 2-2.
[0159] Table 2-2
[0160]
[0161]
[0162] (2) According to the proportion table of incoming calls of various services at different time points within 60 minutes of reporting power outages in the same area, establish the feature set Z of the demand prediction model for user Li Si’s multiple reports of power...
example 3
[0170] Incoming user information is shown in Table 3-1, where 2 means high, 1 means medium, and 0 means low.
[0171] Table 3-1
[0172]
[0173]
[0174] (1) First of all, based on the published planned power outage notice, it is judged that the user Wang Wu was not affected by the power outage event and he did not report any calls from the power outage history in a short period of time in his station area, but Wang Wu has a personal history of calls to handle business history.
[0175] (2) Statistics of Wang Wu's historical accumulative incoming calls to handle various business frequencies, as shown in Table 3-2.
[0176] Table 3-2
[0177] user Power outage related electricity bill Customer Information other inquiries Electricity business …… Wang Wu 8 0 1 0 0 ……
[0178] (3) Calculated with the help of the business preference analysis model for incoming calls from customers, the proportion of Wang Wu's various business activities ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More