Intelligent payment reminding method, device, equipment and medium

Abstract

The invention discloses an intelligent payment reminding method, device and equipment and a medium. The intelligent payment reminding method comprises the steps of obtaining multi-source data of a user; wherein the multi-source data comprises payment time, a payment state, a payment channel, a load fluctuation rate, total arrearage times and a payment urging response condition in a historical time period, and a GDP state and a user credit score of a current area; according to the multi-source data and a pre-constructed arrearage risk secondary evaluation model, determining the arrearage probability of the user in the current month; and determining a current month payment reminding strategy of the user according to the current month arrearage probability and a preset payment reminding strategy comparison table. The intelligent payment reminding device, the equipment and the medium are used for realizing the intelligent payment reminding method.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power collection, and in particular to an intelligent collection method, device, equipment and medium. BACKGROUND

[0002] The traditional electricity fee arrears risk assessment method mainly relies on static data such as user historical payment records, and is difficult to respond to dynamic risk factors such as user behavior changes in real time. The existing technology usually adopts fixed collection rules, lacks personalized strategies, and leads to high user complaint rate and low collection efficiency. In addition, there is a "data island" problem between internal data of the power system and other external data, and multi-dimensional data cannot be effectively integrated, which limits the risk prediction accuracy and makes it difficult to support accurate collection decisions, cannot remind the customer to pay in time, and is easy to cause unexpected power outage, affecting the customer's life.

[0003] In the prior art, there have been related researches trying to build a multi-dimensional evaluation model through user historical payment behavior, but the dynamic power consumption behavior characteristics have not been fully integrated. At the same time, the existing collection strategy optimization relies on manual experience and lacks automatic adjustment capability, resulting in strategy update lag and insufficient adaptability. These problems make it difficult to improve the electricity fee arrears risk prediction accuracy, and there is an urgent need for more efficient data integration and dynamic analysis methods to improve the electricity fee arrears risk prediction accuracy, and then optimize the collection strategy, remind the customer in time, and avoid customer power outage. SUMMARY

[0004] The purpose of the present application is to provide an intelligent collection method, device, equipment and medium to solve the problem of not reminding the customer in time and causing unexpected power outage due to strong static data dependence, single collection strategy and data integration capability of the existing collection method mentioned in the background technology.

[0005] In order to achieve the above purpose, the present application adopts the following technical solutions: In a first aspect, the present application provides an intelligent collection method, comprising: obtaining multi-source data of a user; wherein the multi-source data includes payment time, payment state, payment channel, load fluctuation rate, total arrears times and collection response in a historical time period, and current regional GDP state and user credit score; determining the monthly arrears probability of the user according to the multi-source data and a pre-constructed arrears risk secondary evaluation model; determining the monthly collection strategy of the user according to the monthly arrears probability and a preset collection strategy reference table.

[0006] Optionally, the determination of the monthly arrears probability of the user according to the multi-source data and the pre-constructed arrears risk secondary evaluation model comprises: determine the payment time preference, the payment state preference, the payment channel preference and the load fluctuation rate preference of the user in the historical time period according to the payment time, the payment state, the payment channel and the load fluctuation rate in the historical time period; input the payment time preference, the payment state preference, the payment channel preference and the load fluctuation rate preference in the historical time period, the total number of arrears and the response to the arrears in the historical time period, and the current GDP state of the region and the credit score of the user into the arrears risk secondary evaluation model, and the arrears risk secondary evaluation model outputs the arrears probability of the current month.

[0007] Optionally, the arrears risk secondary evaluation model is constructed by the following method, comprising: constructing the primary influence factors and the factor weight of each factor, wherein the primary influence factors include the payment time preference, the payment state preference, the payment channel preference and the load fluctuation rate preference in the historical time period, the total number of arrears and the response to the arrears in the historical time period, and the current GDP state of the region and the credit score of the user, and the sum of the factor weight of each factor is 1; subdivide the primary influence factors to construct the secondary influence factors; determine the initial score of each factor of the secondary influence factors by expert evaluation method, wherein the initial score of each factor score of each secondary influence factor ranges from 0 to 1; determine the score of the tth month of each factor of the secondary influence factors based on the initial score of each factor of the secondary influence factors; calculate the arrears probability of the tth month of the user based on the factor weight of each factor of the primary influence factors and the score of the tth month of each factor of the secondary influence factors, and construct the arrears risk secondary evaluation model.

[0008] Optionally, the determination of the score of the tth month of each factor of the secondary influence factors based on the initial score of each factor of the secondary influence factors comprises: determine the number of total arrears users in the t-1th month and the total arrears users in the t-1th month of each factor of the secondary influence factors; determine the score adjustment coefficient of each factor of the secondary influence factors in the tth month based on the number of total arrears users in the t-1th month and the total arrears users in the t-1th month of each factor of the secondary influence factors; determine the score adjustment scheme of each factor of the secondary influence factors in the tth month according to the score adjustment coefficient of each factor of the secondary influence factors in the tth month; determining the score of the factor of the secondary influence factor in the tth month based on the initial score of the factor of the secondary influence factor and the score adjustment scheme of the factor of the secondary influence factor in the tth month.

[0009] Optionally, the score adjustment scheme of the factor of the secondary influence factor in the tth month is determined according to the score adjustment coefficient of the factor of the secondary influence factor in the tth month, and the score adjustment scheme of the factor of the secondary influence factor in the tth month comprises: if 33.3%≤ the score adjustment coefficient of the factor of the secondary influence factor in the tth month ≤66.6%, the score of the factor of the secondary influence factor in the tth month is not adjusted; if the score adjustment coefficient of the factor of the secondary influence factor in the tth month <33.3%, the score of the factor of the secondary influence factor in the tth month is reduced; if the score adjustment coefficient of the factor of the secondary influence factor in the tth month >66.6%, the score of the factor of the secondary influence factor in the tth month is increased.

[0010] Optionally, the monthly collection strategy of the user is determined according to the probability of the current month's arrears and a preset collection strategy table, and the method comprises the following steps of: matching the probability of the current month's arrears with the state score interval of the collection strategy table to determine the state score interval corresponding to the probability of the current month's arrears; determining the monthly collection strategy according to the state score interval corresponding to the probability of the current month's arrears.

[0011] Optionally, the method further comprises the following steps of: collecting the monthly collection response result of the user based on the monthly collection strategy, wherein the monthly collection response result comprises a successful payment on time, a successful payment after a time delay, a user complaint and a user non-response; matching the monthly collection response result with a preset score evaluation table to determine a monthly collection response score; determining a monthly collection response total score according to the monthly collection response score and a preset collection response initial score; matching the monthly collection response total score with the score interval of a preset strategy adjustment table to determine whether the collection strategy needs to be adjusted; wherein if the collection strategy needs to be adjusted, the monthly collection strategy is downgraded based on the priority of each collection strategy.

[0012] According to the second aspect of the present application, an intelligent collection device is provided, which comprises: The collection module acquires multi-source data of the user; wherein the multi-source data comprises payment time, payment state, payment channel, load fluctuation rate, total arrears times and fee collection response condition in a historical time period, and current region GDP state and user credit score; The arrears prediction module is configured to determine the current month arrears probability of the user according to the multi-source data and a pre-constructed arrears risk secondary evaluation model. The fee collection strategy decision module is configured to determine the current month fee collection strategy of the user according to the current month arrears probability and a preset fee collection strategy reference table.

[0013] In a third aspect, the present application provides an electronic device comprising a processor and a memory, wherein the processor is configured to execute a computer program stored in the memory to implement the intelligent fee collection method according to any one of the above embodiments.

[0014] In a fourth aspect, the present application provides a computer readable storage medium storing at least one instruction, wherein the at least one instruction is executed by a processor to implement the intelligent fee collection method according to any one of the above embodiments.

[0015] Compared with the prior art, the present application has the following advantages: The intelligent fee collection method provided by the present application solves the problem that the existing fee collection method in the prior art cannot timely remind the customer and causes unexpected power outage due to strong static data dependency, single fee collection strategy and poor data integration capability, and has the beneficial effects that: by integrating user behavior data (payment time, channel, arrears times), economic environment indicators (region GDP) and credit scores, multi-source data cross-validation can be performed, the comprehensiveness and accuracy of arrears risk assessment can be significantly improved, and in combination with a secondary evaluation model, real-time updating of the arrears probability can be realized to avoid the hysteresis problem of the traditional static model; and based on the arrears probability threshold, the fee collection strategy is matched to reduce the invalid fee collection cost, improve the collection efficiency, improve the prediction accuracy, optimize the fee collection strategy, and thus the customer can be timely reminded to avoid unexpected power outage.

[0016] Further, the original data is converted into quantifiable user preference indicators, and in combination with economic indicators and credit scores, data layering processing is realized to reduce the model calculation complexity, enhance the interpretability of the model input features, improve the prediction accuracy, and thus the customer can be timely reminded to avoid unexpected power outage.

[0017] Further, by combining the first factor weight distribution with the secondary factor dynamic scoring, a double-layer analysis of "macro weight + micro scoring" is realized to improve the interpretability of the model, and the arrears probability is adjusted according to the latest secondary factor score every month to make the model have real-time adaptability, and thus the customer can be timely reminded to avoid unexpected power outage.

[0018] Further, dynamically adjust the score based on the actual arrears user data of last month, so that the model continuously fits the latest user behavior characteristics, directly reflects the causal relationship between factors and arrears through the score adjustment coefficient, enhances the logical interpretability of the model, at the same time, through the monthly rolling update of the score, avoids the model failure caused by the sudden change of user behavior, improves the prediction accuracy, and then can timely remind the customer to avoid unexpected power failure.

[0019] Further, based on the statistical three-part principle, the fixed threshold interval is avoided, the score is frequently adjusted due to short-term data fluctuations, the model robustness is improved, the influence of irrelevant interference on the model is reduced, the prediction accuracy is improved, and then the customer can be reminded in time to avoid unexpected power failure.

[0020] Further, the arrears probability is mapped to a preset interval, the subjective bias of manual judgment is eliminated, the matching accuracy of the collection measures and the risk is improved, at the same time, according to the probability interval division strategy, the user complaint caused by excessive collection is avoided, the customer is reminded in a flexible way, and the unexpected power failure is avoided.

[0021] Further, according to the actual response of the user, the collection strategy strength is adjusted in real time, the user loss or legal risk caused by the strategy rigidity is avoided, at the same time, the response score quantifies the sensitivity of the user to the collection measures, a fine user response portrait is constructed, the prediction accuracy is improved, and then the customer can be reminded in time to avoid unexpected power failure.

[0022] The intelligent collection device, the electronic equipment and the computer readable storage medium provided by the application also solve the problem of the prior art that the existing collection method cannot timely remind the customer and causes unexpected power failure due to strong static data dependence, single collection strategy and data integration capability. BRIEF DESCRIPTION OF DRAWINGS

[0023] The drawings accompanying the specification of this application serve to provide a further understanding of the application, the illustrative embodiments of the application and their descriptions serve to explain the application, and do not constitute an improper limitation on the application. In the drawings: Figure 1 The flowchart of the intelligent collection method provided for the embodiments of the application; Figure 2 The structural block diagram of the intelligent collection device provided for the embodiments of the application; Figure 3 The structural block diagram of an electronic equipment provided for the embodiments of the application.

[0024] Among them, 100, electronic equipment;101, memory;102, processor;103, computer program;104, communication bus. DETAILED DESCRIPTION

[0025] The application will be described in detail below with reference to the drawings and in conjunction with embodiments. It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0026] The following detailed description is exemplary and is intended to provide further explanation of the application. Unless otherwise defined, all technical terms used in the present application have the same meanings as generally understood by those skilled in the art. The terms used in the present application are only for the purpose of describing the specific embodiments and are not intended to limit the exemplary embodiments according to the present application.

[0027] Embodiment 1 As Figure 1 shown, in a first aspect of the present application, an intelligent collection method is provided, comprising: S1: obtaining multi-source data of a user; wherein the multi-source data includes payment time, payment state, payment channel, load fluctuation rate, total arrears times and collection response situation in a historical time period, and current regional GDP state and user credit score; S2: determining the monthly arrears probability of the user according to the multi-source data and a pre-constructed arrears risk secondary evaluation model; S3: determining the monthly collection strategy of the user according to the monthly arrears probability and a preset collection strategy reference table.

[0028] It should be noted that the load fluctuation rate can be calculated by the daily load standard deviation and the monthly average load; the historical time period can be the past 12 months; and the data update frequency can be automatically captured on the 1st day of each month.

[0029] Therefore, by integrating user behavior data (payment time, channel, arrears times), economic environment indicators (regional GDP) and credit scores, multi-source data cross-validation can be performed, which significantly improves the comprehensiveness and accuracy of arrears risk assessment. In combination with the secondary evaluation model, the arrears probability is updated in real time to avoid the lag problem of traditional static models. Based on the arrears probability threshold, the collection strategy is matched to reduce the cost of invalid collection, improve the efficiency of collection, and further remind the customer in time to avoid unexpected power outage.

[0030] In step S1: obtaining multi-source data of a user; wherein the multi-source data includes payment time, payment state, payment channel, load fluctuation rate, total arrears times and collection response situation in a historical time period, and current regional GDP state and user credit score.

[0031] Here, the collection module capable of acquiring multi-source data of the user can be configured. Thus, by integrating the user behavior data (payment time, channel, number of arrears), economic environment indicators (regional GDP), and credit scores, multi-source data cross-validation can be performed, significantly improving the comprehensiveness and accuracy of the arrears risk assessment.

[0032] In step S2: according to the multi-source data and the pre-constructed arrears risk secondary evaluation model, the monthly arrears probability of the user is determined.

[0033] Here, the arrears prediction module capable of determining the monthly arrears probability of the user according to the multi-source data and the pre-constructed arrears risk secondary evaluation model can be configured. Thus, in combination with the secondary evaluation model, real-time updating of the arrears probability is realized, avoiding the lag problem of traditional static models.

[0034] In one embodiment, the determination of the monthly arrears probability of the user according to the multi-source data and the pre-constructed arrears risk secondary evaluation model comprises: determining the payment time preference, payment state preference, payment channel preference, and load fluctuation rate preference of the user in the historical time period according to the payment time, payment state, payment channel, and load fluctuation rate in the historical time period; inputting the payment time preference, payment state preference, payment channel preference, and load fluctuation rate preference in the historical time period, the total number of arrears and the response to the arrears collection in the historical time period, and the current regional GDP state and user credit score into the arrears risk secondary evaluation model, and the arrears risk secondary evaluation model outputs the monthly arrears probability.

[0035] Here, the arrears prediction module can be configured to determine the payment time preference, payment state preference, payment channel preference, and load fluctuation rate preference of the user in the historical time period according to the payment time, payment state, payment channel, and load fluctuation rate in the historical time period. It should be noted that the payment time, payment state, payment channel, and load fluctuation rate in the historical time period with the highest proportion in the historical time period of the user are used as the payment time preference, payment state preference, payment channel preference, and load fluctuation rate preference in the historical time period of the user. Thus, the original data is converted into quantifiable user preference indicators, and in combination with economic indicators and credit scores, data layering processing is realized, the model calculation complexity is reduced, the interpretability of the model input features is enhanced, the prediction accuracy is improved, and the customer can be timely reminded to avoid unexpected power outages.

[0036] In one embodiment, the arrears risk secondary evaluation model is constructed by the following method, comprising: constructing a primary influencing factor and a weight value of each factor of the primary influencing factor; wherein the primary influencing factor comprises a payment time preference, a payment state preference, a payment channel preference and a load fluctuation rate preference in a historical time period, a total number of overdue payments and a payment reminder response in the historical time period, and a current region GDP state and a user credit score, and a sum of the weight values of the factors is 1; performing a lower-level subdivision on the primary influencing factor to construct a secondary influencing factor; determining an initial score of each factor of the secondary influencing factor by using an expert evaluation method; wherein an initial score range of each factor score of each secondary influencing factor is between 0 and 1; determining a score of each factor of the secondary influencing factor in the tth month based on the initial score of each factor of the secondary influencing factor; calculating a current-month overdue probability of the user in the tth month based on the weight value of each factor of the primary influencing factor and the score of each factor of the secondary influencing factor in the tth month to construct the overdue risk secondary evaluation model.

[0037] It should be noted that in the expert evaluation method, at least 3 field experts independently score and the mean value is taken; the primary influencing factor and the weight value of each factor thereof can be determined and the secondary influencing factor can be constructed by using a pre-set user overdue risk evaluation table, and the user overdue risk evaluation table is shown in the following table:

[0038] Therefore, by combining the primary factor weight value distribution and the secondary factor dynamic scoring, a double-layer analysis of “macroscopic weight value + microscopic scoring” is realized, the model interpretability is improved, meanwhile, the overdue probability is adjusted according to the latest secondary factor score every month, so that the model has real-time adaptability, the prediction accuracy is improved, and the customer can be reminded in time to avoid unexpected power outage.

[0039] In one embodiment, the formula used for calculating the current-month overdue probability of the user in the tth month based on the weight value of each factor of the primary influencing factor and the score of each factor of the secondary influencing factor in the tth month is as follows:

[0040] wherein the total risk score is the current-month overdue probability of the user in the tth month, the weight i is the weight value of the factor i of the primary influencing factor, the score i is the score of each factor of the secondary influencing factor corresponding to the factor i of the primary influencing factor in the tth month, and n is the total number of the primary influencing factors.

[0041] It should be noted that according to the multi-source data of the user, firstly, the score of the secondary factor corresponding to the user is multiplied by the weight of each first-level influencing factor to obtain the weighted score of each factor, and the weighted scores of each first-level influencing factor are accumulated to obtain the probability of the user's arrears in the month.

[0042] In one embodiment, determining the score of each factor of the secondary influencing factor in the tth month based on the initial score of each factor of the secondary influencing factor comprises: determining the number of times each factor of the secondary influencing factor appears in the total arrears users in the t-1th month and the total arrears users in the t-1th month; determining the score adjustment coefficient of each factor of the secondary influencing factor in the tth month based on the number of times each factor of the secondary influencing factor appears in the total arrears users in the t-1th month and the total arrears users in the t-1th month; determining the score adjustment scheme of each factor of the secondary influencing factor in the tth month according to the score adjustment coefficient of each factor of the secondary influencing factor in the tth month; determining the score of each factor of the secondary influencing factor in the tth month based on the initial score of each factor of the secondary influencing factor and the score adjustment scheme of each factor of the secondary influencing factor in the tth month.

[0043] Thus, the score is dynamically adjusted based on the actual arrears user data of the last month, so that the model continuously fits the latest user behavior characteristics, the causal relationship between the factor and the arrears is directly reflected through the score adjustment coefficient, the logical interpretability of the model is enhanced, at the same time, the score is updated monthly, avoiding the model failure caused by the sudden change of user behavior, and thus the customer can be reminded in time to avoid unexpected power failure.

[0044] In one embodiment, determining the score adjustment scheme of each factor of the secondary influencing factor in the tth month according to the score adjustment coefficient of each factor of the secondary influencing factor in the tth month comprises: if 33.3%≤ the score adjustment coefficient of each factor of the secondary influencing factor in the tth month ≤66.6%, the score of each factor of the secondary influencing factor in the tth month is not adjusted; if the score adjustment coefficient of each factor of the secondary influencing factor in the tth month <33.3%, the score of each factor of the secondary influencing factor in the tth month is reduced; if the score adjustment coefficient of each factor of the secondary influencing factor in the tth month >66.6%, the score of each factor of the secondary influencing factor in the tth month is increased.

[0045] Therefore, based on the statistical ternary principle, a fixed threshold range is used to avoid frequent adjustments to the score due to short-term data fluctuations, thereby improving the robustness of the model, reducing the impact of irrelevant interference on the model, improving prediction accuracy, and enabling timely reminders to customers to avoid unexpected power outages.

[0046] In one embodiment, the formula used to reduce the scores of each factor in the secondary influencing factors in month t is:

[0047] in, Factors that are secondary influencing factors i The rating for month t. To retain coefficients for historical ratings Factors that are secondary influencing factors i The score for month t-1 Factors that are secondary influencing factors i The number of times it appears among the total number of users in arrears in month t-1. For users with total arrears in month t-1, and The adjustment range is ±5%.

[0048] In one embodiment, the formula used to improve the scores of each factor in the secondary influencing factors in month t is:

[0049] in, Factors that are secondary influencing factors i The rating for month t. To retain coefficients for historical ratings Factors that are secondary influencing factors i The score for month t-1 Factors that are secondary influencing factors i The number of times it appears among the total number of users in arrears in month t-1. For users with total arrears in month t-1, and The adjustment range is ±5%.

[0050] In step S3: Based on the monthly arrears probability and the preset collection strategy comparison table, the user's monthly collection strategy is determined.

[0051] It should be noted that a collection strategy decision module can be configured to determine the user's monthly collection strategy based on the monthly arrears probability and a preset collection strategy comparison table. This allows for matching collection strategies based on the arrears probability threshold, reducing ineffective collection costs and improving collection efficiency.

[0052] In one embodiment, the determination of the current-month collection strategy of the user according to the current-month arrearage probability and a preset collection strategy table comprises: matching the current-month arrearage probability with state score intervals of the collection strategy table to determine a state score interval corresponding to the current-month arrearage probability; determining the current-month collection strategy according to the state score interval corresponding to the current-month arrearage probability.

[0053] It should be noted that the collection strategy table is shown in the following table:

[0054] Thus, the arrearage probability is mapped to a preset interval to eliminate subjective bias of manual judgment and improve matching accuracy of collection measures and risks. Meanwhile, the strategy intensity is divided according to the probability interval to avoid user complaints caused by excessive collection, optimize the collection strategy, remind the client in a flexible way, and avoid unexpected power outage.

[0055] In one embodiment, it further comprises: collecting the user according to the current-month collection strategy to determine a current-month collection response result; the collection response result comprises on-time payment success, delayed payment success, user complaint, and user non-response; matching the current-month collection response result with a preset score evaluation table to determine a current-month collection response score; determining a current-month collection response total score according to the current-month collection response score and a preset collection response initial score; matching the current-month collection response total score with a score interval of a preset strategy adjustment table to determine whether the collection strategy needs to be adjusted; wherein, if the collection strategy needs to be adjusted, the current-month collection strategy is downgraded based on the priority of each collection strategy.

[0056] It should be noted that the preset score evaluation table is shown in the following table:

[0057] The preset strategy adjustment table is shown in the following table:

[0058] Thus, the collection strategy intensity is adjusted in real time according to the actual response of the user to avoid user loss or legal risks caused by rigid strategy, meanwhile, the sensitivity of the user to the collection measures is quantified through the response score to build a fine user response image, optimize the collection strategy, and thus the client can be reminded in time to avoid unexpected power outage.

[0059] Thus, the intelligent collection method provided by the application solves the problem that the existing collection method in the prior art cannot timely remind the customer and causes unexpected power outage due to strong static data dependency, single collection strategy and data integration capability, and achieves beneficial effects: through integration of user behavior data (payment time, channel, number of arrears), economic environment indicators (regional GDP) and credit scores, multi-source data cross verification can be performed, the comprehensiveness and accuracy of the arrears risk assessment are significantly improved, at the same time, in combination with a secondary evaluation model, real-time updating of the arrears probability is realized, the hysteresis problem of the traditional static model is avoided, and based on the arrears probability threshold, the collection strategy is matched, the invalid collection cost is reduced, the remittance efficiency is improved, the prediction accuracy is improved, the collection strategy is optimized, and thus the customer can be timely reminded and unexpected power outage is avoided.

[0060] Further, the original data is converted into quantifiable user preference indicators, and then economic indicators and credit scores are combined to realize data hierarchical processing, reduce model calculation complexity, enhance the interpretability of model input features, improve prediction accuracy, and thus the customer can be timely reminded and unexpected power outage is avoided.

[0061] Further, through combination of the first factor weight distribution and the second factor dynamic scoring, double-layer analysis of "macro weight + micro score" is realized, the model interpretability is improved, and at the same time, the arrears probability is adjusted according to the latest second factor score every month, so that the model has real-time adaptability, and thus the customer can be timely reminded and unexpected power outage is avoided.

[0062] Further, the score is dynamically adjusted based on the actual arrears user data of the last month, so that the model continuously adapts to the latest user behavior characteristics, the scoring adjustment coefficient directly reflects the causal relationship between the factors and the arrears, the model logic interpretability is enhanced, at the same time, the score is updated monthly, the model failure caused by sudden change of user behavior is avoided, the prediction accuracy is improved, and thus the customer can be timely reminded and unexpected power outage is avoided.

[0063] Further, based on the statistical three-part principle, the threshold interval is fixed, frequent score adjustment caused by short-term data fluctuations is avoided, the model robustness is improved, the influence of irrelevant interference on the model is reduced, the prediction accuracy is improved, and thus the customer can be timely reminded and unexpected power outage is avoided.

[0064] Further, the arrears probability is mapped to a preset interval, subjective bias of artificial judgment is eliminated, the matching accuracy of the collection measures and the risk is improved, at the same time, the strategy strength is divided according to the probability interval, user complaints caused by excessive collection are avoided, the customer is reminded in a flexible way, and unexpected power outage is avoided.

[0065] Furthermore, the intensity of collection strategies can be adjusted in real time based on actual user responses to avoid user churn or legal risks caused by rigid strategies. At the same time, by quantifying user sensitivity to collection measures through response scoring, a refined user response profile can be built to improve prediction accuracy, thereby enabling timely reminders to customers and avoiding unexpected power outages.

[0066] Example 2 like Figure 2 As shown, based on the same inventive concept as the above embodiments, the present invention also provides an intelligent collection and payment reminder device, comprising: The data acquisition module obtains multi-source data of users; wherein, the multi-source data includes payment time, payment status, payment channel, load fluctuation rate, total number of overdue payments and payment reminder response status within a historical time period, as well as the current GDP status of the region and the user's credit score; The overdue payment prediction module is used to determine the probability of a user's overdue payment in the current month based on the multi-source data and the pre-built secondary overdue payment risk assessment model; The collection strategy decision module is used to determine the user's monthly collection strategy based on the monthly arrears probability and a preset collection strategy comparison table.

[0067] Example 3 like Figure 3 As shown, the present invention also provides an electronic device 100 for implementing the intelligent collection method as provided in any of the above embodiments; The electronic device 100 includes a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and executable on at least one processor 102, and at least one communication bus 104.

[0068] The memory 101 can be used to store the computer program 103. The processor 102 implements the intelligent collection method steps provided in any of the above embodiments by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101.

[0069] The memory 101 may primarily include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created based on the use of the electronic device 100 (such as audio data), etc. In addition, the memory 101 may include non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other non-volatile solid-state storage device.

[0070] The at least one processor 102 can be a central processing unit (CPU), and can also be other general-purpose processors, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, etc. The processor 102 can be a microprocessor or the processor 102 can also be any conventional processor, etc. The processor 102 is a control center of the electronic device 100, and is connected to various parts of the entire electronic device 100 through various interfaces and lines.

[0071] The memory 101 in the electronic device 100 stores a plurality of instructions to implement an intelligent collection method, and the processor 102 can execute the plurality of instructions to implement: Obtaining multi-source data of a user; wherein the multi-source data includes payment time, payment state, payment channel, load fluctuation rate, total number of arrears and collection response situation in a historical time period, and current regional GDP state and user credit score; Determining a current month arrears probability of the user according to the multi-source data and a pre-constructed arrears risk secondary evaluation model; Determining a current month collection strategy of the user according to the current month arrears probability and a preset collection strategy reference table.

[0072] Embodiment 4 In a fourth aspect, the present application provides a computer readable storage medium, the computer readable storage medium stores at least one instruction, and the at least one instruction is executed by a processor to implement the intelligent collection method provided in any one of the above embodiments.

[0073] It should be noted that if the modules / units integrated in the electronic device 100 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, and read-only memory (ROM).

[0074] The intelligent collection device, electronic device, and computer-readable storage medium provided by this invention also solve the problems of existing collection methods mentioned in the background section, which are unable to remind customers in a timely manner and cause unexpected power outages due to their strong dependence on static data, single collection strategies, and poor data integration capabilities.

[0075] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0076] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0077] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the flow Figure 1 one or more flows and / or blocks Figure 1 one or more blocks or multiple blocks.

[0078] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the flow Figure 1 one or more flows and / or blocks Figure 1 one or more blocks or multiple blocks.

[0079] In the description of the present specification, the description of the terms "one embodiment", "an example", "a specific example" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0080] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application rather than limiting them, although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that: the specific embodiments of the present application can be modified or replaced by the same, without departing from the spirit and scope of the present application, any modification or equivalent replacement, which should be covered in the protection scope of the claims of the present application.

Claims

1. A smart dunning method, characterized in that, The method comprises the following steps: obtaining multi-source data of a user; wherein the multi-source data comprises payment time, payment state, payment channel, load fluctuation rate, total number of arrears and fee collection response situation in a historical time period, and current regional GDP state and user credit score; determining a current month arrears probability of the user according to the multi-source data and a pre-constructed arrears risk secondary evaluation model; determining a current month fee collection strategy of the user according to the current month arrears probability and a preset fee collection strategy table.

2. The intelligent dunning method of claim 1, wherein, The step of determining the current month arrears probability of the user according to the multi-source data and the pre-constructed arrears risk secondary evaluation model comprises the following steps: determining payment time preference, payment state preference, payment channel preference and load fluctuation rate preference of the user in the historical time period according to the payment time, payment state, payment channel and load fluctuation rate in the historical time period; inputting the payment time preference, payment state preference, payment channel preference and load fluctuation rate preference in the historical time period, the total number of arrears and fee collection response situation in the historical time period, and the current regional GDP state and user credit score into the arrears risk secondary evaluation model, and the arrears risk secondary evaluation model outputs the current month arrears probability.

3. The intelligent dunning method of claim 2, wherein, The arrears risk secondary evaluation model is constructed by the following method, comprising the following steps: constructing primary influencing factors and factor weights of each factor; wherein the primary influencing factors comprise payment time preference, payment state preference, payment channel preference and load fluctuation rate preference in a historical time period, total number of arrears and fee collection response situation in the historical time period, and current regional GDP state and user credit score, and the sum of the factor weights of each factor is 1; performing secondary subdivision on the primary influencing factors to construct secondary influencing factors; determining initial scores of each factor of the secondary influencing factors by expert evaluation method; wherein the initial score range of each factor score of each secondary influencing factor is between 0 and 1; determining the score of each factor of the secondary influencing factors in the tth month based on the initial scores of each factor of the secondary influencing factors; calculating the current month arrears probability of the user in the tth month based on the factor weights of each factor of the primary influencing factors and the score of each factor of the secondary influencing factors in the tth month, and constructing the arrears risk secondary evaluation model.

4. The intelligent dunning method of claim 3, wherein, The step of determining the score of each factor of the secondary influencing factors in the tth month based on the initial scores of each factor of the secondary influencing factors comprises the following steps: determining the number of total arrears users in which each factor of the secondary influencing factors appears in the t-1th month and the total arrears users in the t-1th month; determining a score adjustment coefficient of each factor of the secondary influencing factors in the tth month based on the number of total arrears users in which each factor of the secondary influencing factors appears in the t-1th month and the total arrears users in the t-1th month; determining a score adjustment scheme of each factor of the secondary influencing factors in the tth month according to the score adjustment coefficient of each factor of the secondary influencing factors in the tth month; Determine the score of the tth month of each factor of the secondary influence factor based on the initial score of each factor of the secondary influence factor and the score adjustment scheme of the tth month of each factor of the secondary influence factor.

5. The intelligent dunning method of claim 4, wherein, Determine the score adjustment scheme of the tth month of each factor of the secondary influence factor according to the score adjustment coefficient of the tth month of each factor of the secondary influence factor, comprising: If 33.3%≤ the score adjustment coefficient of the tth month of each factor of the secondary influence factor ≤ 66.6%, do not adjust the score of the tth month of each factor of the secondary influence factor; If the score adjustment coefficient of the tth month of each factor of the secondary influence factor < 33.3%, reduce the score of the tth month of each factor of the secondary influence factor; If the score adjustment coefficient of the tth month of each factor of the secondary influence factor > 66.6%, increase the score of the tth month of each factor of the secondary influence factor.

6. The intelligent dunning method of claim 1, wherein, Determine the current month's collection strategy of the user according to the current month's arrears probability and a preset collection strategy table, comprising: Match the current month's arrears probability with the state score interval of the collection strategy table to determine the state score interval corresponding to the current month's arrears probability; Determine the current month's collection strategy according to the state score interval corresponding to the current month's arrears probability.

7. The intelligent dunning method of claim 1, wherein, Further comprising: Collect the current month's collection response result based on the current month's collection strategy; The collection response result comprises on-time payment success, delayed payment success, user complaint and user non-response; Match the current month's collection response result with a preset score evaluation table to determine the current month's collection response score; Determine the current month's collection response total score according to the current month's collection response score and a preset collection response initial score; Match the current month's collection response total score with the score interval of a preset strategy adjustment table to determine whether the collection strategy needs to be adjusted; wherein, If the collection strategy needs to be adjusted, down-regulate the current month's collection strategy based on the priority of each collection strategy.

8. A smart dunning device of claim 1, wherein, Comprise: The collection module acquires the multi-source data of the user; wherein, the multi-source data comprises the payment time, payment state, payment channel, load fluctuation rate, total arrears times and collection response situation in the historical time period, and the current regional GDP state and user credit score; The arrears prediction module is configured to determine the current month's arrears probability of the user according to the multi-source data and a pre-constructed arrears risk secondary evaluation model; The collection strategy decision module is configured to determine the current month's collection strategy of the user according to the current month's arrears probability and a preset collection strategy table.

9. An electronic device, comprising: The computer readable storage medium stores at least one instruction, and the at least one instruction is executed by the processor to implement the intelligent collection method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores at least one instruction, and the at least one instruction is executed by the processor to implement the intelligent collection method according to any one of claims 1 to 7.

Images