CN111198907A

CN111198907A - Method and device for identifying potential defaulting user, computer equipment and storage medium

Info

Publication number: CN111198907A
Application number: CN201911345201.4A
Authority: CN
Inventors: 黄腾; 邱方驰; 孙晓佳; 陈华仙; 郑晨露
Original assignee: Shenzhen Power Supply Bureau Co Ltd
Current assignee: Shenzhen Power Supply Bureau Co Ltd
Priority date: 2019-12-24
Filing date: 2019-12-24
Publication date: 2020-05-26

Abstract

The application relates to a method and a device for identifying a potential arrearage user, computer equipment and a storage medium. The method comprises the following steps: acquiring arrearage record data, electricity utilization correlation information and credit representation information of each sample user from a database; carrying out statistical analysis processing on arrearage recorded data, electricity utilization correlation information and credit representation information of each sample user to obtain a training data set corresponding to each sample user; performing machine learning training according to the training data set to generate an arrearage user identification model; acquiring model input data corresponding to a user to be identified; the model input data corresponding to the user to be identified is determined according to the arrearage record data, the electricity utilization associated information and the credit representation information of the user to be identified; and inputting model input data corresponding to the user to be identified into the arrearage user identification model, and predicting to obtain a potential arrearage identification result of the user to be identified. By adopting the method, the information amount of power supply processing is improved.

Description

Method and device for identifying potential defaulting user, computer equipment and storage medium

Technical Field

The present application relates to the field of information processing technologies, and in particular, to a method and an apparatus for identifying a potential arrearage user, a computer device, and a storage medium.

Background

With the opening of the power market and the transformation and upgrading faced by the power supply service, the application of the computer processing technology in the power supply scene is more and more extensive. For example, the relevant data of the electricity consumption of the user is stored on line in an electronic mode, so that the unified management of the relevant data of the electricity consumption is realized.

In the traditional method, only simple storage and query are carried out on the electricity consumption related data, and information mining and research are not further carried out, so that the quantity of information which can be expressed by the large quantity of electricity consumption related data is low.

Disclosure of Invention

In view of the above, it is desirable to provide a method, an apparatus, a computer device, and a storage medium for identifying a potential defaulting user, which can improve the amount of information expressed by power consumption data.

A method of identifying a potential owed user, the method comprising:

acquiring arrearage record data, electricity utilization correlation information and credit representation information of each sample user from a database;

carrying out statistical analysis processing on the arrearage recorded data, the electricity utilization correlation information and the credit representation information of each sample user to obtain a training data set corresponding to each sample user;

performing machine learning training according to the training data set to generate an arrearage user identification model;

acquiring model input data corresponding to a user to be identified; the model input data corresponding to the user to be identified is determined according to the arrearage record data, the electricity utilization associated information and the credit representation information of the user to be identified;

and inputting the model input data corresponding to the user to be identified into the arrearage user identification model, and predicting to obtain a potential arrearage identification result of the user to be identified.

In one embodiment, the performing machine learning training according to the training data set, and generating an arrearage user recognition model includes:

constructing a model construction factor according to the training data set;

and iteratively performing machine learning training on the weight of each model construction factor according to the training data set to obtain an arrearage user identification model.

In one embodiment, before the obtaining of the model input data corresponding to the user to be recognized, the method further includes:

acquiring historical arrearage user information and model input data corresponding to the historical arrearage user information;

inputting model input data corresponding to the historical arrearage user information into the arrearage user identification model to predict arrearage user information;

verifying the accuracy of the arrearage user identification model according to the difference comparison result of the historical arrearage user information and the predicted arrearage user information;

and after the verification is passed, executing the step of acquiring the model input data corresponding to the user to be identified.

In one embodiment, the electricity consumption related information comprises at least one of electricity consumption data, weather data during electricity consumption, power supply scheme data, holiday data during electricity consumption and power failure and power recovery data during electricity consumption; the credit representation information comprises at least one of electricity utilization credit data, industry popularity information of the industry to which the user belongs and negative information of the enterprise to which the user belongs.

In one embodiment, the statistical analysis processing of the arrearage record data, the electricity consumption correlation information and the credit characterization information of each sample user includes at least one of the following steps:

carrying out statistical analysis on the arrearage recorded data to obtain at least one of arrearage times, arrearage years and months, arrearage amount and arrearage frequency of a sample user;

analyzing and processing the power consumption data, and determining the power consumption in a preset period of a user;

analyzing and processing weather data during power utilization, and determining a contemporaneous temperature difference value during power utilization;

and effective data screening processing is carried out on the power supply scheme data, the holiday data during power utilization and the power failure and recovery data during the power utilization.

In one embodiment, the method further comprises:

determining a potential arrearage user according to the potential arrearage identification result;

according to the credit representation information of the potential arrearage users, corresponding credit levels are divided for the potential arrearage users;

aiming at a low credit level which is less than or equal to a preset level threshold, acquiring a collection urging template which is set corresponding to the low credit level;

and generating collection prompting plan contents aiming at the potential arrearage users at the low credit level according to the collection prompting template.

In one embodiment, the method further comprises:

acquiring historical electricity utilization information of the potential arrearage user;

and carrying out classification statistics on the potential defaulting users according to the historical electricity utilization information and preset statistical dimensions.

A potential owed user identification apparatus, the apparatus comprising:

the data acquisition module is used for acquiring arrearage record data, electricity utilization correlation information and credit representation information of each sample user from the database;

the statistical analysis module is used for performing statistical analysis processing on the arrearage record data, the electricity utilization correlation information and the credit representation information of each sample user to obtain a training data set corresponding to each sample user;

the model training module is used for performing machine learning training according to the training data set to generate an arrearage user identification model;

the data acquisition module is also used for acquiring model input data corresponding to the user to be identified; the model input data corresponding to the user to be identified is determined according to the arrearage record data, the electricity utilization associated information and the credit representation information of the user to be identified;

and the potential arrearage user identification module is used for inputting the model input data corresponding to the user to be identified into the arrearage user identification model and predicting to obtain a potential arrearage identification result of the user to be identified.

A computer device comprising a memory storing a computer program and a processor implementing the steps of the method of identifying a potential owed user according to embodiments of the present application when executing the computer program.

A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method for identifying a potential owed user according to the embodiments of the present application.

According to the method, the device, the computer equipment and the storage medium for identifying the potential arrearage user, the arrearage record data, the electricity utilization correlation information and the credit representation information of each sample user are analyzed and processed to construct the arrearage user identification model, so that the potential arrearage identification result of the user to be identified can be predicted according to the arrearage user identification model. Namely, the potential arrearage user can be predicted through mining analysis of the information, and the prediction of the potential arrearage user can provide a very important reference for subsequent power supply processing, so that the expressed information amount is increased compared with the traditional method.

Drawings

FIG. 1 is a diagram of an exemplary environment in which a method for identifying potential owed users may be implemented;

FIG. 2 is a schematic flow chart diagram of a potential owed user identification method in one embodiment;

FIG. 3 is a schematic flow chart diagram of a potential owed user identification method in one embodiment;

FIG. 4 is a block diagram of a potential owed user identification system in one embodiment;

FIG. 5 is a block diagram of a potential owed user identification device in one embodiment;

FIG. 6 is a block diagram of a potential owed user identification device in one embodiment;

FIG. 7 is a block diagram of a potential owed user identification device in one embodiment;

FIG. 8 is a block diagram of a potential owed user identification device in one embodiment;

FIG. 9 is a diagram illustrating an internal structure of a computer device according to an embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

The method for identifying the potential arrearage user can be applied to the application environment shown in figure 1. Wherein the computer device 102 obtains data from the database 104 over a network. The computer device 102 may be a terminal or a server, and the database 104 may be a database storing data of arrearage records, electricity consumption related information, credit representation information, and the like of the user.

In one embodiment, as shown in fig. 2, a method for identifying a potential owing user is provided, which is illustrated by applying the method to the computer device in fig. 1, and comprises the following steps:

s202, acquiring arrearage record data, electricity utilization related information and credit representation information of each sample user from the database.

The database is corresponding to arrearage record data, electricity utilization correlation information and credit representation information of each sample user. Each sample user is a user sample used for machine learning training as a training data set. The arrearage record data is data representing the user electricity arrearage record. The electricity consumption related information is information related to the electricity consumption of the user. The credit representation information is information that can represent the credit situation of the user.

Specifically, the computer equipment acquires the data and the information from a database corresponding to the arrearage record data, the electricity utilization related information and the credit representation information of each sample user.

In one embodiment, the electricity consumption related information may include at least one of electricity consumption data, weather data during electricity consumption, power supply scheme data, holiday data during electricity consumption, and power failure and restoration data during electricity consumption.

In one embodiment, the credit characterization information may include at least one of electricity usage credit data, industry landscape information of an industry to which the user belongs, negative information of a business the user is incumbent on, and the like.

In one embodiment, the computer device obtains data from a database corresponding to different data types according to the data types.

Specifically, the computer device may obtain the arrearage record data from the marketing system. The computer device may obtain power usage data from the metering automation system. The computer device may obtain weather data on electricity usage from a database of a weather website. The computer device may obtain a power scheme ledger from the marketing system as power scheme data. The power scheme data may include: the power supply contract capacity, the running capacity, the stop and report, the start and stop, the electricity price data and the standing book data of the non-residential users, and the electricity price data and the standing book data of the residential users. The computer device may obtain holiday data when power is used from the internet device. The computer equipment can obtain the power failure and restoration data in the power utilization process from the client omnibearing system. The computer device may obtain the electricity credit data from the china electric power enterprise consortium. The computer equipment can acquire the industry information of the industry to which the user belongs from the industry information monitoring platform of the State Council development research center. The computer equipment can obtain the negative information of the enterprises which the user is willing to work in from the local court and the authoritative newspaper, wherein the negative information comprises dispute, judicial law and the like.

In one embodiment, the system background of the computer device automatically and periodically acquires the arrearage record data, the electricity utilization correlation information and the credit characterization information of each sample user from the database. Data of recent years can be acquired once at the beginning, and then subsequent data can be acquired according to a certain frequency. The data of recent years can be acquired at one time, and the acquisition is not continued subsequently, or the data is acquired at a certain frequency from the beginning. The data acquisition frequency can be preset according to different data types. For example, the frequency of acquiring business landscape information may be once a month, and the frequency of acquiring negative information of a user-assigned business may be once a month.

And S204, carrying out statistical analysis processing on the arrearage recorded data, the electricity utilization correlation information and the credit representation information of each sample user to obtain a training data set corresponding to each sample user.

The statistical analysis processing is to perform corresponding statistical analysis on the acquired data of different types, so as to obtain a training data set required by machine learning training. The training data set is a data set for generating an arrearage user recognition model by performing machine learning training. The training data set corresponding to each sample user is a data set which is obtained after statistical analysis processing is carried out on arrearage recorded data, electricity utilization correlation information and credit representation information of each sample user and is used for carrying out machine learning training to generate an arrearage user identification model.

Specifically, the computer device performs corresponding statistical analysis processing on different types of data in the arrearage recorded data, the electricity consumption correlation information and the credit representation information of each sample user, and then corresponds the obtained analysis result to the corresponding sample user to obtain a training data set corresponding to each sample user.

In one embodiment, different types of data may be targeted, such as: and respectively carrying out corresponding statistical analysis processing on different types of arrearage recorded data, power consumption data, weather data during power consumption, power supply scheme data and the like, and then corresponding the obtained analysis results to corresponding sample users to obtain training data sets corresponding to the sample users.

And S206, performing machine learning training according to the training data set to generate an arrearage user identification model.

The machine learning training is a method for obtaining a target model by training a data set with a learner. The arrearage user identification model is a model which can output an arrearage user identification result by inputting data obtained by performing statistical analysis processing on arrearage record data, electricity utilization related information and credit representation information of a user.

Specifically, the computer device inputs the training data set into a learner, performs iterative machine learning training, and generates an arrearage user identification model.

In one embodiment, the machine learning may be either conventional or deep.

S208, obtaining model input data corresponding to the user to be identified; and the model input data corresponding to the user to be identified is determined according to the arrearage record data, the electricity utilization associated information and the credit representation information of the user to be identified.

The user to be identified is the user needing to identify whether the user is a potential arrearage user through the arrearage user identification model. The model input data is data for inputting the arrearage user identification model. The model input data corresponding to the user to be identified is data of an input owing user identification model corresponding to a user who needs to be identified as a potential owing user by the owing user identification model. The model input data corresponding to the user to be identified is determined according to the arrearage record data, the electricity utilization associated information and the credit representation information of the user to be identified, and means that the model input data corresponding to the user to be identified is obtained by performing statistical analysis on the arrearage record data, the electricity utilization associated information and the credit representation information of the user to be identified.

Specifically, the computer device obtains model input data corresponding to a user to be identified.

In one embodiment, the computer device may first obtain the arrearage record data, the electricity consumption association information and the credit representation information of the user to be identified from the database, and then perform statistical analysis on the obtained data to obtain the model input data corresponding to the user to be identified.

And S210, inputting model input data corresponding to the user to be identified into the arrearage user identification model, and predicting to obtain a potential arrearage identification result of the user to be identified.

Wherein, the identification result of the potential arrearage is the identification result of whether the user is the potential arrearage user. And predicting to obtain a potential arrearage recognition result of the user to be recognized, namely predicting whether the user to be recognized is a potential arrearage user, wherein the potential arrearage recognition result is an output result of the arrearage user recognition model.

Specifically, the computer equipment inputs model input data corresponding to the user to be identified into the arrearage user identification model, and outputs a predicted potential arrearage identification result of the user to be identified through the model.

In one embodiment, the computer device may display the potential arrearage users corresponding to the potential arrearage identification result on the display in a list form, so as to facilitate subsequent management, statistics and analysis of the potential arrearage users.

In the embodiment, the arrearage user identification model is constructed through analyzing and processing the arrearage record data, the electricity utilization correlation information and the credit representation information of each sample user, so that the potential arrearage identification result of the user to be identified can be predicted according to the arrearage user identification model. Namely, the potential arrearage user can be predicted through mining analysis of the information, and the prediction of the potential arrearage user can provide a very important reference for subsequent power supply processing, so that the information amount is increased compared with the traditional method.

In one embodiment, the step of performing machine learning training according to a training data set to generate an arrearage user recognition model specifically includes the following steps: constructing a model construction factor according to the training data set; and (4) iteratively performing machine learning training on the weight of each model construction factor according to the training data set to obtain an arrearage user identification model.

The model construction factor is a constituent element of the target model to be constructed. The weight of the model construction factor is a proportion of the model construction factor in the model, that is, a coefficient of the model construction factor.

Specifically, the computer device constructs model construction factors according to a training data set, and then iteratively performs machine learning training on the weight of each model construction factor according to the training data set, namely the proportion of each model construction factor in the model, to obtain the weight of each model characteristic factor, thereby obtaining the arrearage user identification model.

In one embodiment, the computer device may construct a model construction factor according to a training data set obtained by performing statistical analysis processing on arrearage record data, power consumption association information and credit characterization information of each sample user acquired from a database, and then perform machine learning training iteratively on the weight of the model construction factor according to the training data set to obtain an arrearage user identification model.

In this embodiment, the computer device constructs the model formation factor according to the training data set, and iteratively performs machine learning training on the weight of the model formation factor to obtain the arrearage user identification model, so as to lay down prediction of a potential arrearage identification result of the user to be identified according to the arrearage user identification model, and compared with a traditional method, the information amount is increased.

In one embodiment, before obtaining the model input data corresponding to the user to be recognized, the method further comprises the following steps: acquiring historical arrearage user information and model input data corresponding to the historical arrearage user information; inputting model input data corresponding to the historical arrearage user information into an arrearage user identification model, and predicting arrearage user information; verifying the accuracy of the arrearage user identification model according to the difference comparison result of the historical arrearage user information and the predicted arrearage user information; and when the verification is passed, executing the step of acquiring the model input data corresponding to the user to be identified.

The historical arrearage user information is information of a user who has been recorded with arrearages in the past. The model input data is data to be input into the model, and is a data set obtained by performing statistical analysis on the arrearage record data, the electricity consumption related information and the credit representation information of the user. The model input data corresponding to the historical arrearage user information is a data set obtained by performing statistical analysis on arrearage record data, electricity consumption related information and credit representation information of the historical arrearage user. The information of the predicted arrearage user is the arrearage user predicted by the arrearage user recognition model. The comparison result of the difference between the historical arrearage user information and the predicted arrearage user information is the difference between the historical arrearage user information and the predicted arrearage user information. And verifying the accuracy of the arrearage user identification model, namely verifying whether the arrearage user identification model can reach the expected accuracy degree. And the model can reach the expected accuracy degree after verification is passed.

Specifically, the computer device acquires historical arrearage user information and model input data corresponding to the historical arrearage user information, and inputs the model input data corresponding to the historical arrearage user information into the arrearage user identification model. And then, the obtained model output result is used as the information of the forecast arrearage user, and the difference comparison is carried out on the information of the forecast arrearage user and the historical information of the arrearage user, so as to verify whether the identification model of the arrearage user can achieve the expected accuracy degree. And if the model can reach the expected accuracy degree, executing the step of acquiring the model input data corresponding to the user to be identified.

In one embodiment, the computer device may first obtain the arrearage record data, the electricity consumption correlation information and the credit representation information of the historical arrearage user from the database, and then perform statistical analysis on the data to obtain the model input data corresponding to the historical arrearage user information.

In another embodiment, the computer device may also divide the data set obtained by performing statistical analysis processing on the arrearage record data, the power consumption association information and the credit characterization information of each sample user in S204 into a training data set and a verification data set, and obtain the verification data set as model input data corresponding to the historical arrearage user information.

In one embodiment, in the process of verifying the accuracy of the arrearage user identification model by the computer device, if the verification is passed, that is, the model is verified to reach the expected accuracy, the step of obtaining model input data corresponding to the user to be identified is executed, and the model input data is determined according to the arrearage record data, the electricity utilization related information and the credit representation information of the user to be identified.

In another embodiment, in the process of verifying the accuracy of the arrearage user identification model by the computer device, if the verification fails, that is, if the model cannot reach the expected accuracy degree after the verification, the machine learning training is iteratively performed on the weights of the model composition factors according to the model input data corresponding to the historical arrearage user information, and the weights are adjusted until the verification passes, then the step of acquiring the model input data corresponding to the user to be identified is performed.

In the embodiment, the computer equipment acquires the historical arrearage user information and the model input data corresponding to the historical arrearage user information, and carries out accuracy verification on the arrearage user identification model, so that the accuracy of the model is improved, and potential arrearage users can be identified more accurately.

In one embodiment, the electricity consumption related information comprises at least one of electricity consumption data, weather data during electricity consumption, power supply scheme data, holiday data during electricity consumption and power failure and restoration data during electricity consumption; credit characterizing information including at least one of credit data, industry popularity information of an industry to which the user belongs, and negative information of a business to which the user is tasked.

The electricity consumption data is data representing the electricity consumption of the user. The weather data during power consumption is data representing weather conditions during power consumption of a user. The power supply scheme data is data representing a power supply scheme. The holiday data during power utilization is data representing holiday conditions during power utilization of a user. The power failure and recovery data in the power utilization process are data representing power failure and recovery conditions in the power utilization period of a user. The power consumption credit data is data for representing credit of the enterprise which the user owns. The business prosperity information of the industry to which the user belongs is information representing the business prosperity condition of the industry to which the user belongs. The negative information of the user-assigned enterprise is information representing the negative information condition of the user-assigned enterprise.

In one embodiment, the power usage data may be an electrical energy representation obtained from the metering automation system at 24 points of time per day by the user. The weather data during power utilization can be weather data of the area where the user is located every day obtained from a database of a weather website. The power supply scheme data may be a power supply scheme ledger acquired from a marketing system, and specifically includes: the power supply contract capacity, the running capacity, the stop and report, the start and stop, the electricity price data and the standing book data of the non-residential users, and the electricity price data and the standing book data of the residential users. The holiday data when power is used may be holiday data during which power is obtained from the internet appliance, including saturday, sunday, and legal holiday. The power-off and power-off data in the power utilization process can be the power-off and power-off data obtained by the client omnibearing system. The electricity credit data may be the last three years credit information of the user incumbent enterprise obtained from a database of the chinese power enterprise consortium. The industry information of the industry to which the user belongs can be the industry information of the industry to which the user belongs obtained from a database of a national institute development and research center industry information monitoring platform, and the data obtaining frequency is once a month. The negative information of the user-assigned enterprises can be obtained from local courts and authoritative newspapers, the negative information of the user-assigned enterprises including disputes, judicial laws and the like is obtained, the original score is taken as 100 points, weighting deduction is carried out according to the occurrence frequency, and the data obtaining frequency is once a month.

In the embodiment, data are acquired from multiple database channels, and a defaulting user identification model is laid, so that the information quantity is improved, the model accuracy is also improved, and the accuracy of potential defaulting user identification is improved.

In one embodiment, the statistical analysis processing is performed on the arrearage record data, the electricity utilization correlation information and the credit characterization information of each sample user, and comprises at least one of the following steps: carrying out statistical analysis on the arrearage recorded data to obtain at least one of arrearage times, arrearage years and months, arrearage amount and arrearage frequency of the sample user; analyzing and processing the power consumption data, and determining the power consumption in a preset period of a user; analyzing and processing weather data during power utilization, and determining a contemporaneous temperature difference value during power utilization; and effective data screening processing is carried out on the power supply scheme data, the holiday data during power utilization and the power failure and recovery data during the power utilization.

The defaulting times are the defaulting times of the user in a preset period. The arrearage month is the month in which the user arrearage acts. The arrearage amount is an amount owed by the user. The frequency of defaulting is how often the user defaults. The electricity consumption in the preset period is how much electricity the user uses in the preset period. The contemporaneous temperature difference is the difference of the temperatures at the same time on different dates. The valid data screening process is to remove invalid data (for example, incomplete information or unclear information) and screen out valid data.

Specifically, the computer device carries out statistical analysis on the arrearage recorded data to obtain at least one of the arrearage times, the arrearage years and months, the arrearage amount and the arrearage frequency of the sample user. The computer equipment analyzes and processes the electricity consumption data and determines the electricity consumption (such as time-sharing electricity consumption, daily electricity consumption, monthly electricity consumption and the like) in a preset period of the user. The computer equipment analyzes and processes the weather data during power utilization and determines the contemporaneous temperature difference and/or the human body comfort level index during power utilization. And the computer equipment carries out effective data screening processing on the power supply scheme data, the holiday data during power utilization and the power failure and restoration data during the power utilization.

In the embodiment, corresponding processing is performed according to different data types of the arrearage recorded data, the electricity consumption correlation information and the credit representation information of each sample user, so that the information amount of the data set is improved, and the accuracy of identification of potential arrearage users is improved.

In one embodiment, the method further comprises the steps of: determining potential arrearage users according to the potential arrearage identification result; dividing corresponding credit grades for the potential arrearage users according to the credit representation information of the potential arrearage users; aiming at a low credit level which is less than or equal to a preset level threshold, acquiring a collection urging template which is set corresponding to the low credit level; and generating the collection urging plan content aiming at the potential arrearage user at the low credit level according to the collection urging template.

The potential defaulting user is a user who may be defaulting in the future. The credit rating is a division of the credit level of the user. The preset grade threshold is a preset grade threshold and is used for evaluating whether the credit grade of the user reaches a preset degree. The collection urging template is used for urging collection of the electric charge for the user with low credit level. The collection urging plan contents are plans for urging collection of users with low credit levels.

Specifically, the computer device determines the potential arrearage user according to a potential arrearage identification result output by the arrearage user identification model. And according to the credit representation information of the potential defaulting users, corresponding credit grades are divided for the potential defaulting users. And if the credit level of the potential defaulting user is less than or equal to the preset level threshold, judging the potential defaulting user to be in a low credit level. And acquiring a collection urging template corresponding to the low credit level, and generating collection urging plan contents aiming at the potential arrearage user at the low credit level according to the collection urging template.

In one embodiment, potential defaulting users are ranked according to their credit characterization information. Such as: the potential defaulting users who are frequently defaulted and frequently urged are low in credit level. The credit rating is higher for potential arrearage users who occasionally arrear once.

In this embodiment, the computer device divides the corresponding credit levels for the potential defaulting users, and makes the collection urging plan content for the potential defaulting users at a low credit level, so that the rating of the credit level of the users can be enhanced, and auxiliary support is provided for the collection urging of the electric charges.

In one embodiment, the method further comprises the steps of: determining potential arrearage users according to the potential arrearage identification result; acquiring historical electricity utilization information of potential defaulting users; and carrying out classification statistics on the potential defaulting users according to historical electricity utilization information and preset statistical dimensions.

The historical electricity consumption information is information representing the electricity consumption situation of the user in the past. The preset statistical dimension is a preset dimension for performing classification statistics.

In one embodiment, the computer device may classify and count historical power usage information of potential owed users by regional dimension. For example: the areas where potential defaulting users are primarily distributed are analyzed. The computer equipment can also classify and count the historical electricity utilization information of the potential defaulting users according to the dimension of the electricity utilization type. For example: electricity for life, electricity for production, and the like. The computer equipment can also classify and count the historical electricity utilization information of the potential defaulting users according to the quantity dimension. For example: the number of defaulting users in each period can be compared to verify the effectiveness of the collection urging work. The computer equipment can classify and count the historical electricity utilization information of the potential arrearage users according to dimensions such as arrearage times, arrearage duration and the like.

In this embodiment, the computer device performs classification statistics on potential defaulting users according to preset statistical dimensions, and provides auxiliary support for electric charge collection.

In one embodiment, as shown in FIG. 3, a flow diagram of a method for identifying a potential owed user is shown. After the process is started, firstly, the computer equipment prepares data, namely, the arrearage record data, the electricity utilization related information and the credit representation information of each sample user are obtained from the database. And then, processing the data, namely, performing statistical analysis on the arrearage record data, the electricity utilization correlation information and the credit representation information of each sample user to obtain a training data set corresponding to each sample user. And then, constructing a feature model, namely performing machine learning training according to the training data set to generate an arrearage user identification model. Then, the model is verified, i.e., the accuracy of the owed user identification model is verified. And then, identifying the potential arrearage user, namely inputting model input data corresponding to the user to be identified into the arrearage user identification model, and predicting to obtain a potential arrearage identification result of the user to be identified. And finally, managing the potential defaulting users, namely, making a collection prompting plan content aiming at the potential defaulting users with low credit level, and carrying out classification statistics on the potential defaulting users according to preset statistical dimensions.

In one embodiment, as shown in FIG. 4, there is a block diagram of a potential owed user identification system. Firstly, the data preparation module acquires data from a background service system, namely acquires arrearage record data, electricity utilization related information and credit representation information of each sample user from a database. Then, the feature model building module uses the data preparation module to build a model from the data acquired by the background service system, namely, the arrearage recorded data, the electricity consumption correlation information and the credit representation information of each sample user are subjected to statistical analysis processing to obtain a training data set corresponding to each sample user, and machine learning training is carried out according to the training data set to generate an arrearage user identification model. Then, the model verification module performs accuracy verification on the model constructed by the feature model construction module, namely, the accuracy of the arrearage user identification model is verified. And then, the potential arrearage user identification module inputs the model input data corresponding to the user to be identified into the arrearage user identification model passing the verification, and the potential arrearage identification result of the user to be identified is obtained through prediction. And finally, the potential arrearage user management module manages the potential arrearage users, namely, aiming at the potential arrearage identification result identified by the potential arrearage user identification module, the potential arrearage users at a low credit level formulate the urging plan content, and classifying and counting the potential arrearage users according to the preset counting dimensionality.

It should be understood that, although the steps in the flowchart of fig. 2 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a portion of the steps in fig. 2 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

In one embodiment, as shown in fig. 5, there is provided a potential owing user identification apparatus 500 comprising: a data acquisition module 502, a statistical analysis module 504, a model training module 506, and a potential owing subscriber identification module 508, wherein:

and the data acquisition module 502 is used for acquiring the arrearage record data, the electricity utilization related information and the credit representation information of each sample user from the database.

And the statistical analysis module 504 is configured to perform statistical analysis processing on the arrearage record data, the power consumption association information, and the credit characterization information of each sample user to obtain a training data set corresponding to each sample user.

And the model training module 506 is used for performing machine learning training according to the training data set to generate an arrearage user identification model.

The data acquisition module 502 is further configured to acquire model input data corresponding to the user to be identified; and the model input data corresponding to the user to be identified is determined according to the arrearage record data, the electricity utilization associated information and the credit representation information of the user to be identified.

And the potential arrearage user identification module 508 is used for inputting the model input data corresponding to the user to be identified into the arrearage user identification model and predicting to obtain the potential arrearage identification result of the user to be identified.

In one embodiment, model training module 506 is further configured to construct model construction factors from the training data set; and (4) iteratively performing machine learning training on the weight of the model construction factor according to the training data set to obtain the arrearage user identification model.

In this embodiment, the computer device constructs model construction factors according to the training data set, and iteratively performs machine learning training on the weights of the model construction factors to obtain the arrearage user identification model, so as to lay down the prediction of the potential arrearage identification result of the user to be identified according to the arrearage user identification model, and compared with the conventional method, the information amount is increased.

In one embodiment, as shown in fig. 6, the potential owing subscriber identification means 500 further comprises:

a model verification module 510, configured to obtain historical arrearage user information and model input data corresponding to the historical arrearage user information; inputting model input data corresponding to the historical arrearage user information into an arrearage user identification model, and predicting arrearage user information; verifying the accuracy of the arrearage user identification model according to the difference comparison result of the historical arrearage user information and the predicted arrearage user information; when the verification passes, the data acquisition module 502 is notified to perform the step of acquiring the model input data corresponding to the user to be identified.

In one embodiment, the statistical analysis module 504 is further configured to perform at least one of the following: carrying out statistical analysis on the arrearage recorded data to obtain at least one of arrearage times, arrearage years and months, arrearage amount and arrearage frequency of the sample user; analyzing and processing the power consumption data, and determining the power consumption in a preset period of a user; analyzing and processing weather data during power utilization, and determining a contemporaneous temperature difference value during power utilization; and effective data screening processing is carried out on the power supply scheme data, the holiday data during power utilization and the power failure and recovery data during the power utilization.

In one embodiment, as shown in fig. 7, the potential owing subscriber identification means 500 further comprises:

the potential arrearage user collection module 512 is used for determining potential arrearage users according to the potential arrearage identification result; dividing corresponding credit grades for the potential arrearage users according to the credit representation information of the potential arrearage users; aiming at a low credit level which is less than or equal to a preset level threshold, acquiring a collection urging template which is set corresponding to the low credit level; and generating the collection urging plan content aiming at the potential arrearage user at the low credit level according to the collection urging template.

In one embodiment, as shown in fig. 8, the potential owing subscriber identification means 500 further comprises:

a classification and statistics module 514 for potential defaulting users, configured to determine potential defaulting users according to the identification result of potential defaulting; acquiring historical electricity utilization information of potential defaulting users; and carrying out classification statistics on the potential defaulting users according to historical electricity utilization information and preset statistical dimensions.

For specific limitations of the potential defaulting user identification means, reference may be made to the above limitations of the potential defaulting user identification method, which are not described in detail herein. The various modules in the potential owing subscriber identification means described above may be implemented in whole or in part by software, hardware, and combinations thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 9. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer equipment is used for storing data such as arrearage record data, electricity utilization related information, credit representation information and the like of each sample user. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of identifying a potential owing user.

Those skilled in the art will appreciate that the architecture shown in fig. 9 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program: acquiring arrearage record data, electricity utilization correlation information and credit representation information of each sample user from a database; carrying out statistical analysis processing on arrearage recorded data, electricity utilization correlation information and credit representation information of each sample user to obtain a training data set corresponding to each sample user; performing machine learning training according to the training data set to generate an arrearage user identification model; acquiring model input data corresponding to a user to be identified; the model input data corresponding to the user to be identified is determined according to the arrearage record data, the electricity utilization associated information and the credit representation information of the user to be identified; and inputting model input data corresponding to the user to be identified into the arrearage user identification model, and predicting to obtain a potential arrearage identification result of the user to be identified.

In one embodiment, the processor, when executing the computer program, further performs the steps of: constructing a model construction factor according to the training data set; and (4) iteratively performing machine learning training on the weight of each model construction factor according to the training data set to obtain an arrearage user identification model.

In one embodiment, the processor, when executing the computer program, further performs the steps of: acquiring historical arrearage user information and model input data corresponding to the historical arrearage user information; inputting model input data corresponding to the historical arrearage user information into an arrearage user identification model, and predicting arrearage user information; verifying the accuracy of the arrearage user identification model according to the difference comparison result of the historical arrearage user information and the predicted arrearage user information; and when the verification is passed, executing the step of acquiring the model input data corresponding to the user to be identified.

In one embodiment, the processor, when executing the computer program, further performs at least one of the following: carrying out statistical analysis on the arrearage recorded data to obtain at least one of arrearage times, arrearage years and months, arrearage amount and arrearage frequency of the sample user; analyzing and processing the power consumption data, and determining the power consumption in a preset period of a user; analyzing and processing weather data during power utilization, and determining a contemporaneous temperature difference value during power utilization; and effective data screening processing is carried out on the power supply scheme data, the holiday data during power utilization and the power failure and recovery data during the power utilization.

In one embodiment, the processor, when executing the computer program, further performs the steps of: determining potential arrearage users according to the potential arrearage identification result; dividing corresponding credit grades for the potential arrearage users according to the credit representation information of the potential arrearage users; aiming at a low credit level which is less than or equal to a preset level threshold, acquiring a collection urging template which is set corresponding to the low credit level; and generating the collection urging plan content aiming at the potential arrearage user at the low credit level according to the collection urging template.

In one embodiment, the processor, when executing the computer program, further performs the steps of: determining potential arrearage users according to the potential arrearage identification result; acquiring historical electricity utilization information of potential defaulting users; and carrying out classification statistics on the potential defaulting users according to historical electricity utilization information and preset statistical dimensions.

In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:

acquiring arrearage record data, electricity utilization correlation information and credit representation information of each sample user from a database; carrying out statistical analysis processing on arrearage recorded data, electricity utilization correlation information and credit representation information of each sample user to obtain a training data set corresponding to each sample user; performing machine learning training according to the training data set to generate an arrearage user identification model; acquiring model input data corresponding to a user to be identified; the model input data corresponding to the user to be identified is determined according to the arrearage record data, the electricity utilization associated information and the credit representation information of the user to be identified; and inputting model input data corresponding to the user to be identified into the arrearage user identification model, and predicting to obtain a potential arrearage identification result of the user to be identified.

In one embodiment, the computer program when executed by the processor further performs the steps of: constructing a model construction factor according to the training data set; and (4) iteratively performing machine learning training on the weight of each model construction factor according to the training data set to obtain an arrearage user identification model.

In one embodiment, the computer program when executed by the processor further performs the steps of: acquiring historical arrearage user information and model input data corresponding to the historical arrearage user information; inputting model input data corresponding to the historical arrearage user information into an arrearage user identification model, and predicting arrearage user information; verifying the accuracy of the arrearage user identification model according to the difference comparison result of the historical arrearage user information and the predicted arrearage user information; and when the verification is passed, executing the step of acquiring the model input data corresponding to the user to be identified.

In one embodiment, the computer program when executed by the processor further performs at least one of the following: carrying out statistical analysis on the arrearage recorded data to obtain at least one of arrearage times, arrearage years and months, arrearage amount and arrearage frequency of the sample user; analyzing and processing the power consumption data, and determining the power consumption in a preset period of a user; analyzing and processing weather data during power utilization, and determining a contemporaneous temperature difference value during power utilization; and effective data screening processing is carried out on the power supply scheme data, the holiday data during power utilization and the power failure and recovery data during the power utilization.

In one embodiment, the computer program when executed by the processor further performs the steps of: determining potential arrearage users according to the potential arrearage identification result; dividing corresponding credit grades for the potential arrearage users according to the credit representation information of the potential arrearage users; aiming at a low credit level which is less than or equal to a preset level threshold, acquiring a collection urging template which is set corresponding to the low credit level; and generating the collection urging plan content aiming at the potential arrearage user at the low credit level according to the collection urging template.

In one embodiment, the computer program when executed by the processor further performs the steps of: determining potential arrearage users according to the potential arrearage identification result; acquiring historical electricity utilization information of potential defaulting users; and carrying out classification statistics on the potential defaulting users according to historical electricity utilization information and preset statistical dimensions.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. A method of identifying a potential owed user, the method comprising:

2. The method of claim 1, wherein the machine learning training from the training data set, generating an arrearage user recognition model comprises:

constructing a model construction factor according to the training data set;

3. The method of claim 1, wherein prior to said obtaining model input data corresponding to a user to be identified, the method further comprises:

4. The method according to claim 1, wherein the electricity consumption related information comprises at least one of electricity consumption data, weather data during electricity consumption, power supply scheme data, holiday data during electricity consumption and power failure and power recovery data during electricity consumption; the credit representation information comprises at least one of electricity utilization credit data, industry popularity information of the industry to which the user belongs and negative information of the enterprise to which the user belongs.

5. The method of claim 4, wherein the statistical analysis of the arrearage record data, the electricity usage correlation information and the credit characterization information of each of the sample users comprises at least one of the following steps:

6. The method of claim 1, further comprising:

7. The method of claim 1, further comprising:

8. A potential owed user identification apparatus, comprising:

9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.