CN113344432A - Method and device for judging regional customer service risk - Google Patents
Method and device for judging regional customer service risk Download PDFInfo
- Publication number
- CN113344432A CN113344432A CN202110723755.4A CN202110723755A CN113344432A CN 113344432 A CN113344432 A CN 113344432A CN 202110723755 A CN202110723755 A CN 202110723755A CN 113344432 A CN113344432 A CN 113344432A
- Authority
- CN
- China
- Prior art keywords
- address
- customer service
- client
- customer
- appeal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/01—Customer relationship services
- G06Q30/015—Providing customer assistance, e.g. assisting a customer within a business location or via helpdesk
- G06Q30/016—After-sales
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Abstract
The application discloses a method and a device for judging regional customer service risks. Firstly, power utilization information of a power grid customer is obtained, and useless data are removed. And then forming an original address table according to the customer service work order and the customer address. And identifying the original address table, determining a five-level address library, and then collecting the customer service worksheets according to the five-level address library to form a customer appeal address relation table. And analyzing the client appeal to generate a client appeal collection model. And finally, generating a region risk level identification model according to the client appeal collection model and the client appeal address relation table. The method is based on a gridding service concept, analysis and monitoring of customer service risks are collected into a target area, risk identification is carried out on the target area every day through an area risk level identification model, the risk level of the target area is determined, basis of regional customer service problems is provided for customer service staff, and therefore regional customer service risks are effectively avoided.
Description
Technical Field
The application relates to the technical field of electric power big data, in particular to a method and a device for judging regional customer service risks.
Background
In recent years, new technologies such as cloud computing, big data, internet of things, mobile internet and artificial intelligence are emerging, new generation information technology is rapidly developed, the transformation and upgrading of traditional industry customer service is continuously promoted, customer service personnel are promoted to improve service modes, and service efficiency is improved. On the other hand, with the advancement of supply side reform and power system reform, consumer consumption modes, energy structure transformation and energy ecological situations in China present new development trends, consumers put forward higher differentiated requirements on 'good electricity utilization', limited service resources and traditional technical means cannot meet rapidly increasing business volume and continuously improved customer requirements, and the acceptance volume of consumer appeal in the power industry presents a trend of increasing year by year.
Currently, the service risk of a client is generally judged according to the appeal of the single client. Individual customer appeal can only be handled passively until a large number of customer appeal's are not widely collected and analyzed. However, only for a single client, the problem that the client appeal is not considered is the universality of the area where the client is located, and the appeal of one client may represent the regional group user appeal, namely, the regional client service risk exists.
Disclosure of Invention
The application discloses a method and a device for judging regional client service risks, which are used for solving the technical problems that in the prior art, the appeal of a single client can only be passively processed before a large number of client appeal is not widely collected and analyzed, and the problem that the customer appeal is not considered is the universality problem of the region where the client appeal is located.
The first aspect of the present application discloses a method for determining regional customer service risk, including:
the method comprises the steps of obtaining power grid customer electricity utilization information, and removing useless data in the power grid customer electricity utilization information, wherein the power grid customer electricity utilization information comprises a plurality of customer service worksheets and a plurality of customer addresses, any one of the customer service worksheets corresponds to a customer appeal, and each customer address comprises a five-level structured address and an unstructured address;
determining an address base basic table according to the plurality of customer service work orders and the five-level structured address;
determining an original address table according to the address base basic table and the unstructured address;
identifying the original address table and determining a five-level address library;
determining a client appeal address relation table according to the plurality of client service worksheets and the five-level address library;
analyzing the client appeal according to the client service worksheet to generate a client appeal collection model, wherein the client appeal collection model is used for corresponding the client service worksheet with a pre-established subject domain according to the client appeal;
and generating a region risk level identification model according to the client appeal collection model and the client appeal address relation table, wherein the region risk level identification model is used for identifying whether a preset target region has risks or not and determining the risk level of the target region.
Optionally, the removing useless data in the power consumption information of the grid customer includes:
removing the found error customer service worksheet;
removing the customer service worksheets which are manually operated as obsolete;
removing the customer service work orders which cannot define a specific power supply unit, wherein any customer service work order also corresponds to one specific power supply unit;
removing the customer service worksheets with invalid contents;
removing the customer service work orders which do not generate the work order numbers, wherein any customer service work order also corresponds to one work order number;
and judging whether the work order numbers in the plurality of customer service work orders are repeated, if so, only keeping the latest customer service work order.
Optionally, the identifying the address original table and determining a five-level address library include:
and identifying the content in the original address table by using word segmentation and named entity identification, and determining a five-level address library.
Optionally, the analyzing the customer appeal according to the customer service work order to generate a customer appeal aggregation model includes:
and analyzing the appeal points of the client appeal through an LDA (Linear discriminant analysis) model according to the pre-established theme domain and the client service worksheet to generate the client appeal collection model.
Optionally, the generating a region risk level identification model according to the client appeal collection model and the client appeal address relation table includes:
determining the target area according to the client appeal address relation table;
and generating a region risk level identification model by using a GBDT algorithm according to the client appeal collection model and the subject domain.
A second aspect of the present application discloses a device for determining a risk of regional customer service, where the device for determining a risk of regional customer service is applied to the method for determining a risk of regional customer service disclosed in the first aspect of the present application, and the device for determining a risk of regional customer service includes:
the information acquisition module is used for acquiring power grid customer power consumption information and removing useless data in the power grid customer power consumption information, wherein the power grid customer power consumption information comprises a plurality of customer service worksheets and a plurality of customer addresses, any one of the customer service worksheets corresponds to a customer appeal, and each customer address comprises a five-level structured address and an unstructured address;
the address base basic table determining module is used for determining an address base basic table according to the plurality of customer service work orders and the five-level structured address;
the original address table determining module is used for determining an original address table according to the address base basic table and the unstructured address;
the five-level address base determining module is used for identifying the original address table and determining a five-level address base;
the client appeal address relation table determining module is used for determining a client appeal address relation table according to the plurality of client service worksheets and the five-level address library;
the client appeal collection model generation module is used for analyzing the client appeal according to the client service worksheet to generate a client appeal collection model, and the client appeal collection model is used for corresponding the client service worksheet to a pre-established theme domain according to the client appeal;
and the regional client risk judgment module is used for generating a regional risk level identification model according to the client appeal collection model and the client appeal address relation table, and the regional risk level identification model is used for identifying whether a preset target region has risks and determining the risk level of the target region.
Optionally, the information obtaining module includes:
the first useless data processing unit is used for rejecting the found error customer service worksheet;
the second useless data processing unit is used for rejecting the customer service worksheet which is manually operated as a waste;
the third useless data processing unit is used for eliminating the customer service work orders which cannot define a specific power supply unit, and any one customer service work order also corresponds to one specific power supply unit;
the fourth useless data processing unit is used for eliminating the client service worksheets with invalid contents;
the fifth useless data processing unit is used for eliminating the customer service work orders without work order numbers, and any customer service work order also corresponds to a work order number;
and the sixth useless data processing unit is used for judging whether the work order numbers in the plurality of customer service work orders are repeated, and only keeping the latest customer service work order if the work order numbers are repeated.
Optionally, the five-level address bank determining module includes:
and the content identification unit is used for identifying the content in the address original table by utilizing word segmentation and named entity identification, and determining a five-level address library.
Optionally, the customer appeal collection model generating module includes:
and the LDA model processing unit is used for analyzing the appeal points of the client appeal through an LDA model according to the pre-established theme domain and the client service worksheet, and generating the client appeal collection model.
Optionally, the regional client risk judgment module includes:
the preset area acquisition unit is used for determining the target area according to the client appeal address relation table;
and the GBDT algorithm processing unit is used for generating a region risk level identification model by using a GBDT algorithm according to the client appeal collection model and the theme domain.
The application relates to the technical field of electric power big data, and discloses a method and a device for judging regional customer service risks. In the method, power consumption information of a power grid customer is obtained and useless data are removed. And then forming an address base table according to the customer service work order and the five-level structured address in the power grid customer power utilization information. And further supplementing the unstructured address into an address base table, thereby determining an original address table. And identifying the original address table, determining a five-level address library, and then collecting the customer service worksheets according to the five-level address library to form a customer appeal address relation table. And analyzing the client appeal according to the client service worksheet to generate a client appeal collection model. And finally, generating a region risk level identification model according to the client appeal collection model and the client appeal address relation table. The method is based on a gridding service concept, analysis and monitoring of customer service risks are collected into a target area, risk identification is carried out on the target area every day through an area risk level identification model, the risk level of the target area is determined, basis of regional customer service problems is provided for customer service staff, and therefore regional customer service risks are effectively avoided.
Drawings
In order to more clearly explain the technical solution of the present application, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious to those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic workflow diagram of a method for determining a regional customer service risk according to an embodiment of the present application;
fig. 2 is a schematic structural diagram of a device for determining a regional customer service risk disclosed in an embodiment of the present application.
Detailed Description
In order to solve the technical problems that in the prior art, only single customer appeal can be passively processed before a large number of customer appeal is widely collected and analyzed, and the problem that the customer appeal is a general problem of a region where the customer appeal is not considered, the application discloses a method and a device for judging regional customer service risk through the following two embodiments.
A first embodiment of the present application discloses a method for determining a regional customer service risk, which, with reference to a workflow diagram shown in fig. 1, includes:
step S101, power grid customer power consumption information is obtained, useless data in the power grid customer power consumption information are removed, the power grid customer power consumption information comprises a plurality of customer service worksheets and a plurality of customer addresses, any customer service worksheet corresponds to a customer appeal, and the customer addresses comprise five-level structured addresses and unstructured addresses.
In some embodiments of the present application, the power grid customer electricity consumption information includes data on the cloud, such as marketing, metering, customer service, and production. The system is a system of each domain of a power grid, is a system of a data source, and comprises power utilization customers, power utilization addresses, user power supply information, real power charge collection records, arrearage records, abnormal power charge information, transformer relations of a transformer area and the transformer area, customer service comprises customer service work orders, customer requests, basic information of business expansion work orders, power failure events, power failure equipment and power failure customer details, metering comprises power failure alarms, voltage and current curves and daily electric quantity freezing table codes, and production comprises lines, line transformer relations and transformers.
Further, the removing useless data in the power consumption information of the grid customer comprises:
and removing the found error customer service worksheet.
And removing the customer service worksheets which are manually operated to be invalid.
And removing the customer service work orders which cannot define a specific power supply unit, wherein any customer service work order also corresponds to a specific power supply unit.
And removing the customer service worksheets with invalid contents.
And removing the customer service work orders which do not generate the work order numbers, wherein any customer service work order also corresponds to one work order number.
And judging whether the work order numbers in the plurality of customer service work orders are repeated, if so, only keeping the latest customer service work order.
Specifically, the culling system identifies as a scrap work order in order to cull erroneous work order data information that has been found by the system application layer.
And the elimination of the work orders with the work order state as waste is to eliminate the business layer and manually operate the work orders as waste.
The work order with the power supply unit code of '05', '00', 'empty' or 'Chinese character' is removed in order to remove the power supply which cannot be defined to the specific power supply unit in service.
The rejection content is a 'test' or empty work order, and is used for rejecting the work order with system test or invalid content.
And the elimination of the work orders with empty work order numbers is to eliminate the work orders without work order numbers due to system layer errors.
The work order with repeated work order numbers only takes the latest work order, so that the error of a system layer is eliminated, and a plurality of data records are repeatedly generated by one work order.
And step S102, determining an address base basic table according to the plurality of customer service work orders and the five-level structured address.
After useless data are removed, the data are converted into highly readable and completely associated data, namely, data cleaning, processing and integration are carried out according to business requirement logic, for example, a customer service work order address is required to be associated with a work order of a five-level structured address, and the address base table is formed.
And step S103, determining an address original table according to the address base basic table and the unstructured address.
Part of the customer addresses do not conform to the structured addresses and also need to be supplemented for a part of the address base to form an original address table.
And step S104, identifying the original address table and determining a five-level address library.
Further, the identifying the original address table and determining a five-level address library include:
and identifying the content in the original address table by using word segmentation and named entity identification, and determining a five-level address library.
Specifically, the address in the original address table is only a string of characters at this time, and is not subjected to the ranking processing, so that the address is subjected to a ranking processing by the vocabulary. And recognizing the content in the address by utilizing word segmentation and named entity recognition, and merging to form a five-level address library. And the address library reserves the association relationship between the original user and the power grid equipment. At present, the power grid equipment has no association relation with the five-level address, but the customer service work order part can be associated with the equipment, so that the five-level address of the customer service work order is proved to belong to the associated power grid equipment as long as the customer service work order is associated with the equipment.
Illustratively, the word segmentation process is shown in Table 1:
TABLE 1
And step S105, determining a client appeal address relation table according to the plurality of client service worksheets and the five-level address library.
With the customer service work order and the five-level address library, the positions with the requirements can be determined. If the data statistics report forms are needed, statistics can be carried out from the address relation table requested by the client.
And S106, analyzing the client appeal according to the client service worksheet to generate a client appeal collection model, wherein the client appeal collection model is used for corresponding the client service worksheet with a pre-established subject domain according to the client appeal.
Further, the analyzing the customer appeal according to the customer service work order to generate a customer appeal aggregation model includes:
and analyzing the appeal points of the client appeal through an LDA (Linear discriminant analysis) model according to the pre-established theme domain and the client service worksheet to generate the client appeal collection model.
And analyzing the appeal points of the events through the LDA model through a pre-established theme domain, and establishing a client appeal collection model. And identifying whether the content of the customer service worksheet belongs to a certain theme, namely a certain appeal, through a predefined theme.
The LDA model can carry out collection analysis according to the content of the customer service work order, and the customer service work order is arranged in a pre-established theme domain in a number-to-number mode.
The following are exemplary: dividing the complain types of the customer service worksheet into the complain subjects of outage type, voltage quality type, reading, checking, accepting type and business expanding type, identifying the content of the customer service worksheet by using an LDA (latent dirichlet allocation) model for analysis and identification, generating a customer appeal collection model, automatically distinguishing the customer service worksheets by the model, and collecting the customer service worksheets into corresponding subjects.
And S107, generating a region risk level identification model according to the client appeal collection model and the client appeal address relation table, wherein the region risk level identification model is used for identifying whether a preset target region has risks or not and determining the risk level of the target region.
Further, the generating a region risk level identification model according to the client appeal collection model and the client appeal address relation table includes:
and determining the target area according to the client appeal address relation table.
And generating a region risk level identification model by using a GBDT algorithm according to the client appeal collection model and the subject domain. The GBDT algorithm, namely the gradient lifting tree algorithm, is mainly applied to data analysis and prediction, has a good effect, and is an integrated algorithm based on a decision tree.
Under different subjects, an area risk level identification model is established by using a GBDT algorithm, the risk reasons of the complaints occurring in the area historical data are analyzed, and the risk levels of the complaints of all the subjects of the area are divided by calculation.
And inputting the defined risk regions into an algorithm and then performing model training to obtain a region risk level identification model according to actual data. And inputting all the target areas into the algorithm model to identify which areas in the target areas have risks, obtaining a risk coefficient of each area, and judging the risk level of each area according to the risk coefficient. The present embodiment divides the risk level into first, second, third and fourth levels according to the severity.
The indexes of the risk coefficients of the specific influence areas are shown in table 2:
TABLE 2
According to the method for judging regional customer service risk disclosed by the embodiment of the application, the power utilization information of the power grid customer is obtained and useless data are removed. And then forming an address base table according to the customer service work order and the five-level structured address in the power grid customer power utilization information. And further supplementing the unstructured address into an address base table, thereby determining an original address table. And identifying the original address table, determining a five-level address library, and then collecting the customer service worksheets according to the five-level address library to form a customer appeal address relation table. And analyzing the client appeal according to the client service worksheet to generate a client appeal collection model. And finally, generating a region risk level identification model according to the client appeal collection model and the client appeal address relation table. The method is based on a gridding service concept, analysis and monitoring of customer service risks are collected into a target area, risk identification is carried out on the target area every day through an area risk level identification model, the risk level of the target area is determined, basis of regional customer service problems is provided for customer service staff, and therefore regional customer service risks are effectively avoided. The 'after the fact' treatment is changed into 'before' prevention, the regional client service risk is early warned, and under the condition that the complaint is not raised, the client service resource allocation is optimized in advance, and the response and the disposal are fast.
The following are embodiments of the apparatus of the present application that may be used to perform embodiments of the method of the present application. For details which are not disclosed in the embodiments of the apparatus of the present application, reference is made to the embodiments of the method of the present application.
A second embodiment of the present application discloses a device for determining a risk of regional customer service, where the device for determining a risk of regional customer service is applied to the method for determining a risk of regional customer service disclosed in the first embodiment of the present application, and referring to a schematic structural diagram shown in fig. 2, the device for determining a risk of regional customer service includes:
the information acquisition module 10 is configured to acquire power grid customer electricity consumption information and remove useless data in the power grid customer electricity consumption information, where the power grid customer electricity consumption information includes a plurality of customer service workflows and a plurality of customer addresses, any one of the customer service workflows corresponds to a customer appeal, and the customer addresses include five-level structured addresses and unstructured addresses.
Further, the information obtaining module 10 includes:
and the first useless data processing unit is used for eliminating the found error customer service worksheet.
And the second useless data processing unit is used for eliminating the customer service worksheets which are manually operated to be invalid.
And the third useless data processing unit is used for eliminating the customer service work orders which cannot define a specific power supply unit, and any one customer service work order also corresponds to one specific power supply unit.
And the fourth useless data processing unit is used for eliminating the client service worksheets with invalid contents.
And the fifth useless data processing unit is used for eliminating the customer service work orders without the work order numbers, and any customer service work order also corresponds to one work order number.
And the sixth useless data processing unit is used for judging whether the work order numbers in the plurality of customer service work orders are repeated, and only keeping the latest customer service work order if the work order numbers are repeated.
And an address base table determining module 20, configured to determine an address base table according to the multiple customer service work orders and the five-level structured address.
And an original address table determining module 30, configured to determine an original address table according to the address library base table and the unstructured address.
And a five-level address base determining module 40, configured to identify the original address table and determine a five-level address base.
Further, the five-level address bank determination module 40 includes:
and the content identification unit is used for identifying the content in the address original table by utilizing word segmentation and named entity identification, and determining a five-level address library.
And the client appeal address relation table determining module 50 is configured to determine a client appeal address relation table according to the plurality of client service work orders and the five-level address library.
And a client appeal collection model generation module 60, configured to analyze the client appeal according to the client service worksheet, and generate a client appeal collection model, where the client appeal collection model is configured to correspond the client service worksheet to a pre-established theme domain according to the client appeal.
Further, the customer appeal collection model generation module 60 includes:
and the LDA model processing unit is used for analyzing the appeal points of the client appeal through an LDA model according to the pre-established theme domain and the client service worksheet, and generating the client appeal collection model.
And a regional client risk judgment module 70, configured to generate a regional risk level identification model according to the client appeal collection model and the client appeal address relation table, where the regional risk level identification model is configured to identify whether a preset target region has a risk or not, and determine a risk level of the target region.
Further, the regional client risk judgment module 70 includes:
and the preset region acquisition unit is used for determining the target region according to the client appeal address relation table.
And the GBDT algorithm processing unit is used for generating a region risk level identification model by using a GBDT algorithm according to the client appeal collection model and the theme domain.
The present application has been described in detail with reference to specific embodiments and illustrative examples, but the description is not intended to limit the application. Those skilled in the art will appreciate that various equivalent substitutions, modifications or improvements may be made to the presently disclosed embodiments and implementations thereof without departing from the spirit and scope of the present disclosure, and these fall within the scope of the present disclosure. The protection scope of this application is subject to the appended claims.
Claims (10)
1. A method for judging regional customer service risk is characterized by comprising the following steps:
the method comprises the steps of obtaining power grid customer electricity utilization information, and removing useless data in the power grid customer electricity utilization information, wherein the power grid customer electricity utilization information comprises a plurality of customer service worksheets and a plurality of customer addresses, any one of the customer service worksheets corresponds to a customer appeal, and each customer address comprises a five-level structured address and an unstructured address;
determining an address base basic table according to the plurality of customer service work orders and the five-level structured address;
determining an original address table according to the address base basic table and the unstructured address;
identifying the original address table and determining a five-level address library;
determining a client appeal address relation table according to the plurality of client service worksheets and the five-level address library;
analyzing the client appeal according to the client service worksheet to generate a client appeal collection model, wherein the client appeal collection model is used for corresponding the client service worksheet with a pre-established subject domain according to the client appeal;
and generating a region risk level identification model according to the client appeal collection model and the client appeal address relation table, wherein the region risk level identification model is used for identifying whether a preset target region has risks or not and determining the risk level of the target region.
2. The method for determining regional customer service risk according to claim 1, wherein the removing useless data in the grid customer electricity consumption information comprises:
removing the found error customer service worksheet;
removing the customer service worksheets which are manually operated as obsolete;
removing the customer service work orders which cannot define a specific power supply unit, wherein any customer service work order also corresponds to one specific power supply unit;
removing the customer service worksheets with invalid contents;
removing the customer service work orders which do not generate the work order numbers, wherein any customer service work order also corresponds to one work order number;
and judging whether the work order numbers in the plurality of customer service work orders are repeated, if so, only keeping the latest customer service work order.
3. The method for determining regional customer service risk according to claim 1, wherein the identifying the original address table to determine a five-level address base comprises:
and identifying the content in the original address table by using word segmentation and named entity identification, and determining a five-level address library.
4. The method of claim 1, wherein the analyzing the customer appeal according to the customer service worksheet to generate a customer appeal aggregation model comprises:
and analyzing the appeal points of the client appeal through an LDA (Linear discriminant analysis) model according to the pre-established theme domain and the client service worksheet to generate the client appeal collection model.
5. The method for determining regional client service risk according to claim 1, wherein the generating a regional risk level identification model according to the client complaint collection model and the client complaint address relationship table includes:
determining the target area according to the client appeal address relation table;
and generating a region risk level identification model by using a GBDT algorithm according to the client appeal collection model and the subject domain.
6. A device for determining a risk of regional customer service, the device being applied to the method for determining a risk of regional customer service according to any one of claims 1 to 5, the device comprising:
the information acquisition module is used for acquiring power grid customer power consumption information and removing useless data in the power grid customer power consumption information, wherein the power grid customer power consumption information comprises a plurality of customer service worksheets and a plurality of customer addresses, any one of the customer service worksheets corresponds to a customer appeal, and each customer address comprises a five-level structured address and an unstructured address;
the address base basic table determining module is used for determining an address base basic table according to the plurality of customer service work orders and the five-level structured address;
the original address table determining module is used for determining an original address table according to the address base basic table and the unstructured address;
the five-level address base determining module is used for identifying the original address table and determining a five-level address base;
the client appeal address relation table determining module is used for determining a client appeal address relation table according to the plurality of client service worksheets and the five-level address library;
the client appeal collection model generation module is used for analyzing the client appeal according to the client service worksheet to generate a client appeal collection model, and the client appeal collection model is used for corresponding the client service worksheet to a pre-established theme domain according to the client appeal;
and the regional client risk judgment module is used for generating a regional risk level identification model according to the client appeal collection model and the client appeal address relation table, and the regional risk level identification model is used for identifying whether a preset target region has risks and determining the risk level of the target region.
7. The apparatus for determining regional customer service risk according to claim 6, wherein the information obtaining module comprises:
the first useless data processing unit is used for rejecting the found error customer service worksheet;
the second useless data processing unit is used for rejecting the customer service worksheet which is manually operated as a waste;
the third useless data processing unit is used for eliminating the customer service work orders which cannot define a specific power supply unit, and any one customer service work order also corresponds to one specific power supply unit;
the fourth useless data processing unit is used for eliminating the client service worksheets with invalid contents;
the fifth useless data processing unit is used for eliminating the customer service work orders without work order numbers, and any customer service work order also corresponds to a work order number;
and the sixth useless data processing unit is used for judging whether the work order numbers in the plurality of customer service work orders are repeated, and only keeping the latest customer service work order if the work order numbers are repeated.
8. The apparatus for determining regional customer service risk according to claim 6, wherein the five-level address base determination module comprises:
and the content identification unit is used for identifying the content in the address original table by utilizing word segmentation and named entity identification, and determining a five-level address library.
9. The apparatus for determining regional customer service risk of claim 6, wherein the customer appeal aggregation model generation module comprises:
and the LDA model processing unit is used for analyzing the appeal points of the client appeal through an LDA model according to the pre-established theme domain and the client service worksheet, and generating the client appeal collection model.
10. The apparatus for determining regional customer service risk according to claim 6, wherein the regional customer risk determining module comprises:
the preset area acquisition unit is used for determining the target area according to the client appeal address relation table;
and the GBDT algorithm processing unit is used for generating a region risk level identification model by using a GBDT algorithm according to the client appeal collection model and the theme domain.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110723755.4A CN113344432B (en) | 2021-06-29 | 2021-06-29 | Method and device for judging regional customer service risk |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110723755.4A CN113344432B (en) | 2021-06-29 | 2021-06-29 | Method and device for judging regional customer service risk |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113344432A true CN113344432A (en) | 2021-09-03 |
| CN113344432B CN113344432B (en) | 2022-05-13 |
Family
ID=77481167
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110723755.4A Active CN113344432B (en) | 2021-06-29 | 2021-06-29 | Method and device for judging regional customer service risk |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113344432B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108320086A (en) * | 2018-01-23 | 2018-07-24 | 国网江西省电力有限公司电力科学研究院 | A kind of power distribution network risk management method and device |
| CN109389418A (en) * | 2018-08-17 | 2019-02-26 | 国家电网有限公司客户服务中心 | Electric service client's demand recognition methods based on LDA model |
| CN109636329A (en) * | 2018-12-04 | 2019-04-16 | 广西电网有限责任公司电力科学研究院 | A kind of Electricity customers complaint analysis system and analysis method based on thermodynamic chart |
| US20200310874A1 (en) * | 2019-03-25 | 2020-10-01 | Fujitsu Limited | Job scheduler, job schedule control method, and storage medium |
| CN112084402A (en) * | 2020-08-24 | 2020-12-15 | 浙江云合数据科技有限责任公司 | Method for predicting user attribute by analyzing application program use data |
| CN112232662A (en) * | 2020-10-15 | 2021-01-15 | 国网冀北电力有限公司计量中心 | Service monitoring system and method |
| CN112529056A (en) * | 2020-12-02 | 2021-03-19 | 国网浙江省电力有限公司营销服务中心 | Research and judgment method for large-area power failure of power distribution network |
| CN112801801A (en) * | 2021-01-29 | 2021-05-14 | 中国建设银行股份有限公司 | Model training method, risk identification method, model, device, equipment and medium |
-
2021
- 2021-06-29 CN CN202110723755.4A patent/CN113344432B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108320086A (en) * | 2018-01-23 | 2018-07-24 | 国网江西省电力有限公司电力科学研究院 | A kind of power distribution network risk management method and device |
| CN109389418A (en) * | 2018-08-17 | 2019-02-26 | 国家电网有限公司客户服务中心 | Electric service client's demand recognition methods based on LDA model |
| CN109636329A (en) * | 2018-12-04 | 2019-04-16 | 广西电网有限责任公司电力科学研究院 | A kind of Electricity customers complaint analysis system and analysis method based on thermodynamic chart |
| US20200310874A1 (en) * | 2019-03-25 | 2020-10-01 | Fujitsu Limited | Job scheduler, job schedule control method, and storage medium |
| CN112084402A (en) * | 2020-08-24 | 2020-12-15 | 浙江云合数据科技有限责任公司 | Method for predicting user attribute by analyzing application program use data |
| CN112232662A (en) * | 2020-10-15 | 2021-01-15 | 国网冀北电力有限公司计量中心 | Service monitoring system and method |
| CN112529056A (en) * | 2020-12-02 | 2021-03-19 | 国网浙江省电力有限公司营销服务中心 | Research and judgment method for large-area power failure of power distribution network |
| CN112801801A (en) * | 2021-01-29 | 2021-05-14 | 中国建设银行股份有限公司 | Model training method, risk identification method, model, device, equipment and medium |
Non-Patent Citations (1)
| Title |
|---|
| 张梅，保富: "基于局部线性嵌入和深度森林算法的电力客户投诉预测模型", 《电测与仪表》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113344432B (en) | 2022-05-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106530121B (en) | Method and system for detecting safety protection compliance of power monitoring system | |
| CN105354616A (en) | Processing device and on-line processing method for electric power measurement asset data | |
| CN108345670B (en) | Service hotspot discovery method for 95598 power work order | |
| CN109687447B (en) | Electric power energy consumption prediction method and device | |
| CN112101971B (en) | Sensitive client identification method, system and storage medium | |
| CN112686491A (en) | Enterprise power data analysis method based on power consumption behavior | |
| CN113065580B (en) | Power plant equipment management method and system based on multi-information fusion | |
| CN113344432B (en) | Method and device for judging regional customer service risk | |
| Du et al. | Text similarity detection method of power customer service work order based on tfidf algorithm | |
| CN107194529B (en) | Power distribution network reliability economic benefit analysis method and device based on mining technology | |
| CN116542781A (en) | Task allocation method, device, computer equipment and storage medium | |
| CN108258802B (en) | Method and device for monitoring running condition of power distribution equipment in power distribution network | |
| CN115438663A (en) | Risk analysis method, device, equipment and medium for power grid work bill | |
| CN110727910B (en) | Method and system for realizing copying and collecting risk prediction and business linkage | |
| CN114626433A (en) | Fault prediction and classification method, device and system for intelligent electric energy meter | |
| CN114118672A (en) | Method and system for automatically generating project requirements of power system | |
| CN113159118A (en) | Logistics data index processing method, device, equipment and storage medium | |
| CN117094688B (en) | Digital control method and system for power supply station | |
| CN113537519A (en) | Method and device for identifying abnormal equipment | |
| Meng et al. | Research on the typical application of knowledge graph in power marketing | |
| CN113435701B (en) | Method and device for processing consumption quality information | |
| An et al. | Automatic Annotation of Metadata in Power System Databases Based on Correlation Feature Selection and Natural Language Processing | |
| CN115018366B (en) | Energy storage system working state monitoring method and device, storage medium and electronic equipment | |
| CN116757807B (en) | Intelligent auxiliary label evaluation method based on optical character recognition | |
| WO2024065776A1 (en) | Method for data processing, apparatus for data processing, electronic device, and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |