CN116681450A - Customer credit evaluation method and system supporting intelligent fee-forcing - Google Patents
Customer credit evaluation method and system supporting intelligent fee-forcing Download PDFInfo
- Publication number
- CN116681450A CN116681450A CN202310347847.6A CN202310347847A CN116681450A CN 116681450 A CN116681450 A CN 116681450A CN 202310347847 A CN202310347847 A CN 202310347847A CN 116681450 A CN116681450 A CN 116681450A
- Authority
- CN
- China
- Prior art keywords
- credit
- customer
- electric power
- client
- electric
- 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
- 238000011156 evaluation Methods 0.000 title claims abstract description 40
- 238000003745 diagnosis Methods 0.000 claims abstract description 34
- 238000012549 training Methods 0.000 claims abstract description 26
- 238000004458 analytical method Methods 0.000 claims abstract description 23
- 238000000034 method Methods 0.000 claims description 45
- 238000004422 calculation algorithm Methods 0.000 claims description 21
- 238000011084 recovery Methods 0.000 claims description 21
- 230000005611 electricity Effects 0.000 claims description 18
- 238000010276 construction Methods 0.000 claims description 9
- 238000004064 recycling Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 7
- 230000001133 acceleration Effects 0.000 claims description 2
- 238000003064 k means clustering Methods 0.000 claims 1
- 238000009472 formulation Methods 0.000 abstract description 4
- 239000000203 mixture Substances 0.000 abstract description 4
- 230000006399 behavior Effects 0.000 description 6
- 230000002776 aggregation Effects 0.000 description 5
- 238000004220 aggregation Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000007621 cluster analysis Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000013210 evaluation model Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000010835 comparative analysis Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000012502 risk assessment Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Finance (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Primary Health Care (AREA)
- Tourism & Hospitality (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a customer credit evaluation method and a system supporting intelligent fee-forcing, which relate to the field of electric fee reading and collecting business of electric power marketing work and construct a training data set based on electric power fee-paying historical data; taking the credit index characteristics of the customers as input variables and the credit grades of the customers as target variables, constructing a credit diagnosis model of the electric customers, and training the credit diagnosis model of the electric customers by using a training data set; acquiring electric power payment data of a customer to be evaluated, extracting credit index characteristics of the customer, inputting the characteristics into a credit diagnosis model of the power training customer, and outputting the credit rating of the customer to be evaluated; according to the invention, the electric power customer credit diagnosis model is constructed by collecting the electric power customer payment behavior data, the electric power customer dynamic credit evaluation demonstration analysis based on the payment behavior is carried out, the electric power customer credit rating is realized, the credit performance is comprehensively evaluated according to the credit rating, the electric power customer classification credit management is promoted, and the formulation of differential charge-accelerating strategies is supported.
Description
Technical Field
The invention belongs to the field of electric charge meter reading and collecting business of electric power marketing work, and particularly relates to a customer credit evaluation method and system supporting intelligent charge acceleration.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
Along with the formation of the 'buyer' market of the national power grid system, the national power grid is required to build leading edge consciousness, fully utilizes big data and informatization means, perfects the national power grid credit evaluation system, scientifically and accurately reads the law behind the big data, formulates accurate marketing strategies, avoids risks and furthest improves the social benefit of the national power grid.
The electric charge recycling service is a key service of power supply enterprise marketing, and a service mode of firstly using electricity and then paying charges is a plurality of potential risks caused by electric charge recycling, so that the economic benefit of the power supply enterprise is directly influenced; with the development of market economy, the electricity demand of customers is increased increasingly, and the prevention and control of the electric charge recycling risk face greater challenges.
The power supply enterprises take the following enhanced electric charge recovery measures: the propaganda strength is increased, and the electricity consumption payment awareness is improved; perfecting and popularizing a payment mode and a payment channel; establishing a special electric charge payment group; a scientific internal charging management mechanism is established, and means such as important attention of distribution personnel to key enterprises are provided. The method improves the electricity charge recovery condition of enterprises to a great extent, but the work mainly starts from the management level, and the effect depends on the management strength of the enterprises, the working experience of electric staff and other factors.
The customer's electric charge recovery risk identification mainly relies on the experience judgment of business personnel, so that the power supply company lacks an effective risk management mechanism to ensure the electric charge recovery rate at present when carrying out electric power marketing, and not only lacks scientific analysis on the payment capability of the user, but also only can carry out the payment in a manual mode when encountering the electric charge delinquent by the user, and the efficiency is greatly limited; meanwhile, the illegal electricity utilization behavior of some electricity utilization enterprises is also lack of effective control, so that the efficiency of electricity charge recovery work is seriously affected.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a customer credit evaluation method and a system for supporting intelligent fee-charging, which are characterized in that by collecting electric customer fee-charging data, an electric customer credit grading evaluation model is constructed, electric customer dynamic credit evaluation demonstration analysis based on fee-charging behavior is developed, electric customer credit grading is realized, credit performance is comprehensively evaluated according to credit grade, electric customer classification credit management is promoted, and the establishment of differential fee-charging strategies is supported.
To achieve the above object, one or more embodiments of the present invention provide the following technical solutions:
the first aspect of the invention provides a client credit evaluation method supporting intelligent fee-forcing;
a customer credit rating method supporting intelligent fee-forcing, comprising:
based on the acquired electric power payment historical data, constructing a training data set consisting of client credit index characteristics and corresponding client credit grades;
taking the credit index characteristics of the customers as input variables and the credit grades of the customers as target variables, constructing a credit diagnosis model of the electric customers, and training the credit diagnosis model of the electric customers by using a training data set;
and acquiring the electric power payment data of the clients to be evaluated, extracting the credit index characteristics of the clients, inputting the electric power payment data into a trained electric power client credit diagnosis model, and outputting the credit grade of the clients to be evaluated.
Further, the method for extracting the credit index features of the client comprises the following steps:
and extracting a client credit index characteristic data set from the acquired electric power payment data by using the constructed electric power client credit evaluation index system.
Further, the evaluation index of the power customer credit evaluation index system comprises: the rate of arrearage, the rate of arrearage amount, the average monthly fee-forcing number, the average recycling period and the overdue fee rate.
Further, the method for calculating the credit rating of the client comprises the following steps:
dividing a client credit index characteristic data set into a plurality of clusters by using a constructed combined clustering method;
and analyzing the credit characteristics of the clients in each cluster, and grading the credit of the clients according to the credit characteristics to obtain the credit grade of each client.
Further, the combined clustering method is based on hierarchical clustering and K-medoids clustering, and comprises the following specific processing steps:
determining the clustering number K by using an elbow method;
dividing data into K clusters by using a hierarchical clustering method;
respectively calculating the sum of the distances between each sample point and other points in each cluster, selecting the point with the smallest sum of the distances as the cluster center of the cluster, and further determining K cluster centers;
and (4) calling a K-medoids clustering algorithm to iterate until all the clustering centers are not changed, and finishing clustering.
Further, the credit rating for the client according to the credit characteristics specifically includes:
AA level client: the electric charge is paid on time in full in the whole analysis period, the phenomenon of electric charge payment is avoided, and the electric charge recovery period is shortest; such user credits are best, ranked first;
class a clients: the electric charge is paid in full time basically in the whole analysis period, the phenomenon of electric charge delineating is very little, and the electric charge recovery period is short; such users are second most credited, ranking second;
BB level client: the electric charge is paid in full time in most of the whole analysis period, the phenomenon of electric charge delineating is common, and the electric charge recovery period is slightly longer; such users are better credited, ranked third;
class B client: the whole analysis period is partially and fully paid with electricity fee on time, the phenomenon of delineating the electricity fee is more, and the electricity fee recovery period is longer; the credit of the users is poor, and the ranking is fourth;
class C clients: the electric charge cannot be paid in full on time basically in the whole analysis period, the phenomenon of electric charge delineating is very frequent, and the electric charge recycling period is longest; such users are the worst in credit and ranked fifth.
Further, the power customer credit diagnosis model is constructed based on an XGBoost algorithm.
The second aspect of the present invention provides a customer credit rating system supporting intelligent fee-based fees.
A client credit evaluation system supporting intelligent fee-forcing comprises a data set construction module, a model construction module and a grade evaluation module:
a dataset construction module configured to: based on the acquired electric power payment historical data, constructing a training data set consisting of client credit index characteristics and corresponding client credit grades;
a model building module configured to: taking the credit index characteristics of the customers as input variables and the credit grades of the customers as target variables, constructing a credit diagnosis model of the electric customers, and training the credit diagnosis model of the electric customers by using a training data set;
a rank evaluation module configured to: and acquiring the electric power payment data of the clients to be evaluated, extracting the credit index characteristics of the clients, inputting the electric power payment data into a trained electric power client credit diagnosis model, and outputting the credit grade of the clients to be evaluated.
A third aspect of the present invention provides a computer readable storage medium having stored thereon a program which when executed by a processor performs steps in a method of evaluating customer credit supporting intelligent tariffs in accordance with the first aspect of the present invention.
A fourth aspect of the invention provides an electronic device comprising a memory, a processor and a program stored on the memory and executable on the processor, the processor implementing the steps in a method for evaluating customer credit supporting intelligent charging according to the first aspect of the invention when the program is executed.
The one or more of the above technical solutions have the following beneficial effects:
(1) The invention designs the credit evaluation index system of the electric power customer based on the electric power payment data, extracts the uniformly quantized evaluation index from the electric power customer payment behavior data, reduces the workload of the subsequent data processing, and improves the instantaneity of the credit evaluation of the customer.
(2) The invention designs a combined clustering method based on a K-medoids algorithm and a hierarchical clustering algorithm, precisely divides the credit evaluation data of the users, realizes the separation of users with different characteristics, forms clusters representing different credit levels, and lays a foundation for the credit grading of different clients.
(3) According to the invention, different clusters formed after cluster analysis are accurately analyzed by using a statistical method, credit characteristic types of different clusters are formed according to characteristic expression of the different clusters, and credit classification is carried out on users of the different clusters based on the credit characteristic types.
(4) The invention builds the XGBoost power customer credit diagnosis model based on the credit index system and the credit grading result data, accurately diagnoses the power customer credit condition, provides a customer credit basis for the power charging activity, and ensures more accurate power charging.
Additional aspects of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention.
Fig. 1 is a flow chart of a method of a first embodiment.
FIG. 2 is a diagram showing an example of selecting cluster centers by the elbow method according to the first embodiment.
Fig. 3 is a system configuration diagram of a second embodiment.
Detailed Description
The invention will be further described with reference to the drawings and examples.
The invention provides a general idea:
in order to solve the problem that the identification of the recovery risk of the electric charge of the customer mainly depends on the experience judgment of service personnel, a power supply enterprise needs to study and construct an electric charge collection credit system, an effective electric charge collection risk management model is built on the basis of comprehensively and scientifically analyzing the self attribute, the electricity consumption historical data, the payment capability, the electricity consumption habit, the normalization and other related attributes of the customer, the electric charge collection credit grade of the customer is obtained, early warning is timely carried out on the high-risk customer, and accordingly a one-to-one personalized service strategy is adopted.
According to the invention, the electric power customer credit grading evaluation model is constructed by collecting the electric power customer payment behavior data, the electric power customer dynamic credit evaluation demonstration analysis based on the payment behavior is carried out, the electric power customer credit rating is realized, the credit performance is comprehensively evaluated according to the credit rating, the electric power customer classification credit management is promoted, and the formulation of differential charge-accelerating strategies is supported.
The invention relates to an electric charge urging process in the electric charge copying and collecting business field of electric charge marketing work, which belongs to an electric behavior analysis method of electric power customers, and constructs credit characteristic indexes of electric power customer payment by carrying out deep analysis on relevant data such as user arrearages, and classifies user groups through a cluster recognition algorithm, thereby realizing accurate rating of the user group payment credit conditions of different types by power supply enterprises, and providing powerful support for developing intelligent urging work based on the accurate rating.
Example 1
The embodiment discloses a client credit evaluation method supporting intelligent fee-forcing, which is used for carrying out credit grading on electric clients, and when a power supply enterprise carries out fee-forcing, the method provided by the embodiment is used for diagnosing the credit condition of the users, carrying out fee-forcing strategy formulation based on the credit grade of the clients, and realizing accurate fee-forcing.
As shown in fig. 1, a method for evaluating credit of a customer supporting intelligent charging includes:
step S1: based on the acquired electric power payment historical data, a training data set consisting of the credit index characteristics of the client and the corresponding credit grades of the client is constructed, and the specific steps are as follows:
step S101: data processing
The method comprises the steps of acquiring electric power payment historical data from an electric power marketing system, wherein main fields comprise a user name, a user number, an electric charge year and month to be paid, a payment channel, current arrearage, electric charge release time, payment amount and arrearage amount to form an original data set. In order to avoid the influence of the data ranges of different fields on the index characteristics and eliminate the influence of different units of different indexes, the acquired power payment historical data is subjected to standardized processing, each data in the original data set is transformed into the range of [0,1] to obtain the power payment basic data, and the standardized formula of the data is as follows:
wherein x is i Is any value of a certain index, x min Is the minimum value of the index, x max Is the minimum value of the index.
Step S102: and constructing a credit evaluation index system of the electric power customer, and measuring and calculating a credit index characteristic data set of the customer based on the acquired electric power payment basic data.
The credit evaluation index system for the electric power customer comprises five evaluation indexes, specifically:
(1) The rate of arrearage is calculated by the formula:
wherein i is the arrearage number, n is the total payment number, and the index reflects the probability of arrearage of the user in a certain time period.
(2) The arrearage amount is calculated by the following formula:
wherein M is the total payment amount, M is the total arrearage amount, and the index can reflect the arrearage condition of the user in a certain time period.
(3) The average monthly fee-promoting times are as follows:
the index reflects the payment frequency of the user, and the higher the payment frequency is, the smaller the probability of the user defaulting is likely to be.
(4) The average recovery period is specifically:
wherein, date max Date for payment expiration Date min Date for electric charge release Date T And the actual payment date is obtained. The index reflects the enthusiasm of user payment, and the smaller the average recovery period is, the more positive the user payment is.
(5) The overdue charge rate is specifically:
wherein N is the number of users paying after the payment expiration date, and N is the total number of users. The index reflects the rate of overdue payment for the user, and the higher the rate of overdue payment for the user is, the greater the probability of arrearage.
Based on the design of the 5 evaluation indexes, 5 evaluation index values of each customer are calculated from the acquired electric power payment data and serve as the credit index characteristics of the customer, so that a credit index characteristic data set of the customer is formed.
Step S103: power customer credit grouping based on combined clustering
Aiming at the defect that an initial clustering center is required to be selected randomly by a K-medoids clustering algorithm, and combining the advantages of a hierarchical clustering model, a combined clustering method based on hierarchical clustering and K-medoids clustering is provided, and the specific implementation process is as follows:
input: customer credit index feature data set
And (3) outputting: k grouping results
The processing steps are as follows:
(1) Determining the clustering number K by using an elbow method;
the core index of the elbow method is SSE (sum of the squared errors, sum of squares error), the formula is:
wherein C is i Is the ith cluster, p is C i Sample points m in (1) i Is C i SSE is the cluster error of all samples, representing the quality of the clustering effect.
The core idea of the elbow method is: as the number K of clusters increases, the sample division is finer, the aggregation degree of each cluster is gradually increased, and then the square error and SSE naturally become smaller gradually; and when K is smaller than the actual cluster number, the aggregation degree of each cluster can be greatly increased due to the increase of K, and when K reaches the actual cluster number, the aggregation degree return obtained by increasing K again can be rapidly reduced, so that the aggregation degree return of SSE can be rapidly reduced, and then the aggregation degree return of the SSE becomes gentle along with the continuous increase of the K value, namely the relation diagram of the SSE and the K is in the shape of an elbow, and the K value corresponding to the elbow is the actual cluster number of the data.
Fig. 2 is an exemplary graph of the elbow method for selecting cluster centers, with the abscissa K being the optimal number of clusters and the ordinate SSE being the sum of squares of the errors. As shown in fig. 2, the elbow corresponds to a K value of 5 (with the highest curvature), so for clustering of this dataset, the optimal cluster number should be chosen to be 5.
(2) Dividing data into K clusters by using a hierarchical clustering method;
(3) Respectively calculating the sum of the distances between each sample point and other points in each cluster, selecting the point with the smallest sum of the distances as the cluster center of the cluster, and further determining K cluster centers;
(4) And (4) calling a K-medoids clustering algorithm to iterate until all the clustering centers are not changed, and finishing clustering.
By using the combined clustering method, the clients are finally divided into 5 clusters according to the client credit index characteristic data set, and the clients can be also called categories.
Step S104: electric power customer credit rating
Based on the dividing result of the combined cluster, the users of each cluster are subjected to comparative analysis to form credit characteristics capable of reflecting the users of the cluster, and the users are subjected to credit grading based on the credit characteristics, specifically:
AA level client: the prepayment contract is signed in the whole analysis period, the electric charge is paid on time in full, the phenomenon of electric charge payment is avoided, and the electric charge recovery period is the shortest. Such user credits are preferably ranked first.
Class a clients: the electric charge is paid in full time basically in the whole analysis period, the phenomenon of electric charge delineating is very little, and the electric charge recovery period is short; such users are credited second best, ranked second.
BB level client: the electric charge is paid in full time in most of the whole analysis period, the phenomenon of electric charge delineating is common, and the electric charge recovery period is slightly longer; such users are better credited and ranked third.
Class B client: the whole analysis period is partially and fully paid with electricity fee on time, the phenomenon of delineating the electricity fee is more, and the electricity fee recovery period is longer; such users are less credited and ranked fourth.
Class C clients: the electric charge cannot be paid in full on time basically in the whole analysis period, the phenomenon of electric charge delineating is very frequent, and the electric charge recycling period is longest; such users are the worst in credit and ranked fifth.
After obtaining the credit rating of the customer, a training data set consisting of the credit index characteristics of the customer and the corresponding credit rating of the customer is constructed and used for training a credit diagnosis model of the electric customer.
Step S2: and constructing a power customer credit diagnosis model by taking the customer credit index characteristics as input variables and the customer credit grade as target variables, and training the power customer credit diagnosis model by using a training data set.
In order to diagnose the credit condition of the customer intelligently, efficiently and in real time, a credit diagnosis model of the electric power customer is constructed based on an XGBoost algorithm, credit index features of the customer are taken as input, and the credit grade of the customer is predicted, so that the problem that the identification of the recovery risk of the electric charge of the customer mainly depends on the experience judgment of service personnel is solved, and the method specifically comprises the following steps:
firstly, performing credit rating on an electric power customer related to electric power payment historical data by utilizing the step 1, and marking the electric power customer with a credit rating label; then, taking the client credit index feature as an input variable, taking a client credit label as a target variable, constructing a power client credit diagnosis model based on an XGBoost algorithm, diagnosing the power client credit condition, providing a client credit basis for the power charging activity, and enabling the power charging to be more accurate.
The XGBoost algorithm belongs to Boosting of an integrated algorithm, is a lifting algorithm of a GBDT model, and is the same as the GBDT algorithm, the goal of each training is to find a function which can reduce fitting residual errors, and different from GBDT, the XGBoost algorithm is improved in an objective function, so that the algorithm effect is improved, and the risk of overfitting is reduced; the XGBoost algorithm improves on the objective function by using taylor expansion to approximate the original objective function and also adds a penalty term to the objective function.
After the electric power customer credit diagnosis model is completed, the model is trained by using a training data set, so that the model has the capability of intelligently predicting the customer credit grade according to the customer credit index characteristics.
Step S3: acquiring electric power payment data of a customer to be evaluated, extracting credit index characteristics of the customer, inputting the characteristics into a trained electric power customer credit diagnosis model, and outputting the credit grade of the customer to be evaluated, wherein the specific steps are as follows:
firstly, acquiring electric power payment data of a customer to be evaluated; then, the electric power payment data are processed in the step S101 and the step S102, and credit index characteristics of the customer are obtained; and finally, processing the credit index characteristics of the client by using the trained power client credit diagnosis model to obtain the credit rating of the client to be evaluated.
The invention aims at: the user payment characteristic data is deeply analyzed, a user credit evaluation characteristic index system is constructed, the credit grades of different users are classified by utilizing a clustering algorithm, the classified credit management of electric customers is promoted, powerful credit evaluation support is provided for the differentiation and the accuracy of enterprise fee-forcing reminding, the accurate marketing is realized, the service quality of enterprises is improved, the service level of the customers is improved, and the electric fee-forcing cost is reduced.
In order to achieve the above objective, the present invention uses historical electric power payment data as a basis, and through deep analysis of the historical electric power payment data, a power customer credit evaluation index system is constructed, and through cluster analysis by a combined clustering algorithm, 5 different clusters representing different credit levels are formed, then through analysis of users of each cluster, credit grading of user groups represented by each cluster is achieved, finally, an XGBoost algorithm is utilized to construct a power customer credit diagnosis model, credit basis is provided for electric power charging work, and formulation and implementation of a differential charging method are supported, and the application of the present invention has important significance:
(1) And a theoretical basis is provided for the management of the customers after credit diagnosis.
Because the third party credit rating system in the current stage of China is not perfect enough, the credit rating of the electric charge paid by the electric power customer needs to be updated in time, the rating is tracked for a long time, and at present, the rating of the electric power customer by the electric power enterprise is still in the stage of evaluating by personal experience, so that errors are easily caused, and therefore, the customer credit diagnosis method studied by the invention provides theoretical basis for risk assessment of the electric power enterprise customer.
(2) The daily operation of the power enterprise is ensured.
Establishing a credit evaluation diagnosis model of a customer to ensure that the electric charge is effectively recovered, and adopting a targeted prompting and paying means according to the credit grade and a specific short board of the customer to fulfill the aim of timely and fully recovering the electric charge; the benefit of the power company can be improved, funds can be returned in time, the operation safety and stability of the power company can be ensured, and better service can be provided for customers.
Example two
The embodiment discloses a client credit evaluation system supporting intelligent fee-forcing;
as shown in fig. 3, a client credit evaluation system supporting intelligent fee-charging includes a data set construction module, a model construction module and a grade evaluation module:
a dataset construction module configured to: based on the acquired electric power payment historical data, constructing a training data set consisting of client credit index characteristics and corresponding client credit grades;
a model building module configured to: taking the credit index characteristics of the customers as input variables and the credit grades of the customers as target variables, constructing a credit diagnosis model of the electric customers, and training the credit diagnosis model of the electric customers by using a training data set;
a rank evaluation module configured to: and acquiring the electric power payment data of the clients to be evaluated, extracting the credit index characteristics of the clients, inputting the electric power payment data into a trained electric power client credit diagnosis model, and outputting the credit grade of the clients to be evaluated.
Example III
An object of the present embodiment is to provide a computer-readable storage medium.
A computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps in a client credit rating method supporting intelligent charging as described in one embodiment of the present disclosure.
Example IV
An object of the present embodiment is to provide an electronic apparatus.
An electronic device comprising a memory, a processor and a program stored on the memory and executable on the processor, wherein the processor, when executing the program, performs the steps in a client credit rating method supporting intelligent charging according to an embodiment of the present disclosure.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A method for evaluating credit of a customer supporting intelligent fee acceleration, comprising:
based on the acquired electric power payment historical data, constructing a training data set consisting of client credit index characteristics and corresponding client credit grades;
taking the credit index characteristics of the customers as input variables and the credit grades of the customers as target variables, constructing a credit diagnosis model of the electric customers, and training the credit diagnosis model of the electric customers by using a training data set;
and acquiring the electric power payment data of the clients to be evaluated, extracting the credit index characteristics of the clients, inputting the electric power payment data into a trained electric power client credit diagnosis model, and outputting the credit grade of the clients to be evaluated.
2. The method for evaluating the credit of the client supporting intelligent fee-charging as claimed in claim 1, wherein the method for extracting the characteristics of the credit index of the client is as follows:
and extracting a client credit index characteristic data set from the acquired electric power payment data by using the constructed electric power client credit evaluation index system.
3. The method for evaluating credit of a customer supporting intelligent charging as claimed in claim 2, wherein the evaluation index of the electric power customer credit evaluation index system comprises: the rate of arrearage, the rate of arrearage amount, the average monthly fee-forcing number, the average recycling period and the overdue fee rate.
4. The method for evaluating the credit of the client supporting the intelligent charging as claimed in claim 2, wherein the calculating method of the credit rating of the client is as follows:
dividing a client credit index characteristic data set into a plurality of clusters by using a constructed combined clustering method;
and analyzing the credit characteristics of the clients in each cluster, and grading the credit of the clients according to the credit characteristics to obtain the credit grade of each client.
5. The method for evaluating the credit of the client supporting intelligent fee-forcing according to claim 4, wherein the combined clustering method is based on hierarchical clustering and K-means clustering, and comprises the following specific processing steps:
determining the clustering number K by using an elbow method;
dividing data into K clusters by using a hierarchical clustering method;
respectively calculating the sum of the distances between each sample point and other points in each cluster, selecting the point with the smallest sum of the distances as the cluster center of the cluster, and further determining K cluster centers;
and (4) calling a K-medoids clustering algorithm to iterate until all the clustering centers are not changed, and finishing clustering.
6. The method for evaluating credit of a customer supporting intelligent charging as claimed in claim 4, wherein the credit rating of the customer according to the credit characteristics is as follows:
AA level client: the prepayment contract is signed in the whole analysis period, the electric charge is paid on time in full, the phenomenon of electric charge payment is avoided, and the electric charge recovery period is shortest; such user credits are best, ranked first;
class a clients: the electric charge is paid in full time basically in the whole analysis period, the phenomenon of electric charge delineating is very little, and the electric charge recovery period is short; such users are second most credited, ranking second;
BB level client: the electric charge is paid in full time in most of the whole analysis period, the phenomenon of electric charge delineating is common, and the electric charge recovery period is slightly longer; such users are better credited, ranked third;
class B client: the whole analysis period is partially and fully paid with electricity fee on time, the phenomenon of delineating the electricity fee is more, and the electricity fee recovery period is longer; the credit of the users is poor, and the ranking is fourth;
class C clients: the electric charge cannot be paid in full on time basically in the whole analysis period, the phenomenon of electric charge delineating is very frequent, and the electric charge recycling period is longest; such users are the worst in credit and ranked fifth.
7. The method for evaluating credit of a customer supporting intelligent charging as recited in claim 1, wherein the power customer credit diagnosis model is constructed based on XGBoost algorithm.
8. The customer credit evaluation system supporting intelligent fee-forcing is characterized by comprising a data set construction module, a model construction module and a grade evaluation module:
the dataset construction module is configured to: based on the acquired electric power payment historical data, constructing a training data set consisting of client credit index characteristics and corresponding client credit grades;
the model building module is configured to: taking the credit index characteristics of the customers as input variables and the credit grades of the customers as target variables, constructing a credit diagnosis model of the electric customers, and training the credit diagnosis model of the electric customers by using a training data set;
the rank evaluation module is configured to: and acquiring the electric power payment data of the clients to be evaluated, extracting the credit index characteristics of the clients, inputting the electric power payment data into a trained electric power client credit diagnosis model, and outputting the credit grade of the clients to be evaluated.
9. An electronic device, comprising:
a memory for non-transitory storage of computer readable instructions; and
a processor for executing the computer-readable instructions,
wherein the computer readable instructions, when executed by the processor, perform the method of any of the preceding claims 1-7.
10. A storage medium, characterized by non-transitory storing computer-readable instructions, wherein the instructions of the method of any one of claims 1-7 are performed when the non-transitory computer-readable instructions are executed by a computer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310347847.6A CN116681450B (en) | 2023-03-30 | 2023-03-30 | Customer credit evaluation method and system supporting intelligent fee-forcing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310347847.6A CN116681450B (en) | 2023-03-30 | 2023-03-30 | Customer credit evaluation method and system supporting intelligent fee-forcing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116681450A true CN116681450A (en) | 2023-09-01 |
| CN116681450B CN116681450B (en) | 2024-06-21 |
Family
ID=87789700
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310347847.6A Active CN116681450B (en) | 2023-03-30 | 2023-03-30 | Customer credit evaluation method and system supporting intelligent fee-forcing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116681450B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117557288A (en) * | 2023-10-18 | 2024-02-13 | 广州电力交易中心有限责任公司 | Market subject transaction behavior standardization method |
| CN117973879A (en) * | 2024-04-02 | 2024-05-03 | 国网山东省电力公司营销服务中心(计量中心) | Power payment risk identification method and system based on multi-source data joint analysis |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160225073A1 (en) * | 2015-01-30 | 2016-08-04 | Wal-Mart Stores, Inc. | System, method, and non-transitory computer-readable storage media for predicting a customer's credit score |
| CN106920110A (en) * | 2017-02-23 | 2017-07-04 | 国网安徽省电力公司合肥供电公司 | A kind of evaluation method of power customer credit and arrears risk |
| CN108665184A (en) * | 2018-05-21 | 2018-10-16 | 国网陕西省电力公司咸阳供电公司 | A kind of power customer credit assessment method based on big data reference |
| CN110580268A (en) * | 2019-08-05 | 2019-12-17 | 西北大学 | Credit scoring integrated classification system and method based on deep learning |
| CN110796354A (en) * | 2019-10-21 | 2020-02-14 | 国网湖南省电力有限公司 | Enterprise electric charge recovery risk portrait method and system |
| CN113743678A (en) * | 2021-09-18 | 2021-12-03 | 中国银行股份有限公司 | User credit score prediction method and related equipment |
| CN115841250A (en) * | 2022-11-30 | 2023-03-24 | 国网山东省电力公司营销服务中心(计量中心) | Electricity charge delinquent risk early warning method and system based on ensemble learning |
-
2023
- 2023-03-30 CN CN202310347847.6A patent/CN116681450B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160225073A1 (en) * | 2015-01-30 | 2016-08-04 | Wal-Mart Stores, Inc. | System, method, and non-transitory computer-readable storage media for predicting a customer's credit score |
| CN106920110A (en) * | 2017-02-23 | 2017-07-04 | 国网安徽省电力公司合肥供电公司 | A kind of evaluation method of power customer credit and arrears risk |
| CN108665184A (en) * | 2018-05-21 | 2018-10-16 | 国网陕西省电力公司咸阳供电公司 | A kind of power customer credit assessment method based on big data reference |
| CN110580268A (en) * | 2019-08-05 | 2019-12-17 | 西北大学 | Credit scoring integrated classification system and method based on deep learning |
| CN110796354A (en) * | 2019-10-21 | 2020-02-14 | 国网湖南省电力有限公司 | Enterprise electric charge recovery risk portrait method and system |
| CN113743678A (en) * | 2021-09-18 | 2021-12-03 | 中国银行股份有限公司 | User credit score prediction method and related equipment |
| CN115841250A (en) * | 2022-11-30 | 2023-03-24 | 国网山东省电力公司营销服务中心(计量中心) | Electricity charge delinquent risk early warning method and system based on ensemble learning |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117557288A (en) * | 2023-10-18 | 2024-02-13 | 广州电力交易中心有限责任公司 | Market subject transaction behavior standardization method |
| CN117973879A (en) * | 2024-04-02 | 2024-05-03 | 国网山东省电力公司营销服务中心(计量中心) | Power payment risk identification method and system based on multi-source data joint analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116681450B (en) | 2024-06-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116681450B (en) | Customer credit evaluation method and system supporting intelligent fee-forcing | |
| CN108665184A (en) | A kind of power customer credit assessment method based on big data reference | |
| CN110119948B (en) | Power consumer credit evaluation method and system based on time-varying weight dynamic combination | |
| CN108596443A (en) | A kind of Electricity customers method for evaluating credit rating based on multi-dimensional data | |
| CN107507038A (en) | A kind of electricity charge sensitive users analysis method based on stacking and bagging algorithms | |
| CN115375205B (en) | Method, device and equipment for determining water user portrait | |
| CN102081781A (en) | Finance modeling optimization method based on information self-circulation | |
| CN114048436A (en) | Construction method and construction device for forecasting enterprise financial data model | |
| CN111798301A (en) | Resource transfer quota determining method, device and system | |
| CN109102396A (en) | A kind of user credit ranking method, computer equipment and readable medium | |
| CN114398941A (en) | User payment behavior classification method, device, equipment and readable storage medium | |
| CN117952740A (en) | Method, device and server for determining resource amount processing strategy of client | |
| CN115130811A (en) | Method and device for establishing power user portrait and electronic equipment | |
| CN113450004A (en) | Power credit report generation method and device, electronic equipment and readable storage medium | |
| CN112529712A (en) | Modeling method and system for user operation analysis RFM | |
| CN111080130A (en) | Power consumer credit evaluation method and system integrating power consumer payment indexes and industry disclosure indexes | |
| CN115905319B (en) | Automatic identification method and system for abnormal electricity fees of massive users | |
| CN117436688A (en) | Electricity fee recycling risk early warning method and system based on electricity marketing data | |
| CN116644956A (en) | Electricity utilization abnormal grade classification method based on multi-scheme decision and fuzzy judgment | |
| CN115983898A (en) | Electricity charge income prediction method and system based on Markov state transition matrix | |
| CN115237970A (en) | Data prediction method, device, equipment, storage medium and program product | |
| CN113627821A (en) | Method and system for identifying abnormal electricity utilization based on electricity utilization behavior characteristics | |
| Fan et al. | Digital Marketing Technology Based on Electricity User Portrait and K-means Clustering Algorithm | |
| Zeng | A comparison study on the era of internet finance China construction of credit scoring system model | |
| CN117236532B (en) | Load data-based electricity consumption peak load prediction method and system |
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 |