A heat supply arrears user risk early warning method and system based on user portrait

Abstract

The present application relates to the technical field of heat supply user management, and relates to a heat supply arrear user risk early warning method and system based on user portrait. The present application sets a heat consumption threshold value through historical heat consumption data of a heat supply user account in a recent historical period, determines whether to send a payment reminder in combination with the current remaining heat consumption of the heat supply user account, continuously monitors the real-time remaining heat consumption and the remaining heat consumption change curve in the triggering period after the platform sends the payment reminder, determines whether to trigger user heat risk assessment, evaluates the payment time behavior risk score and the payment amount behavior risk score based on the heat consumption balance and the payment fee at each historical payment time, accumulates the scores to determine the total arrear risk score, matches the heat supply user arrear risk level, and issues early warning measures and adjusts the current heat consumption credit limit, so as to realize differentiated and dynamic management of the heat supply user arrear risk, and improve the flexibility of heat supply service.

Description

Technical Field

[0001] This invention relates to the field of heating user management technology, and specifically to a risk warning method and system for heating user arrears based on user profiles. Background Technology

[0002] As the service scale of the heating industry continues to expand and the number of heating users increases daily, heating billing models are gradually transforming towards intelligence and precision. Heating arrears have become a significant pain point restricting the stable operation of heating companies. Different users have significantly different heating habits, payment cycles, and preferred payment amounts. If the risk of user arrears cannot be accurately identified and warned in advance, it will not only affect the efficiency of heating companies' cash flow but may also lead to problems such as a decline in user experience due to heating shutdowns caused by arrears.

[0003] Existing technologies have many shortcomings in the early warning of heating arrears risks, specifically in the following aspects: 1. Existing technologies mostly use fixed values ​​when setting heat consumption thresholds, without dynamically adjusting them based on users' recent historical heat consumption data and heat consumption patterns at different times. This results in a disconnect between the heat consumption threshold setting and the user's actual heat consumption, leading to unreasonable timing of payment reminders. Either the reminders are sent too early, causing unnecessary disturbance to users, or they are sent too late, making it impossible to intervene in the risk of arrears in a timely manner.

[0004] 2. When triggering a user's heating risk assessment, existing technologies often only use the remaining heating amount being lower than a set threshold as the sole criterion. They do not construct a curve of the remaining heating amount after sending a payment reminder to track the dynamics of heating usage. This makes it impossible to accurately capture the dynamic risk characteristics of the user's heating usage and easily miss the best time for risk intervention.

[0005] 3. Existing technologies for assessing the risk of heating users' arrears often only focus on the number of arrears, without combining the user's historical heat consumption balance and payment amount for multi-dimensional quantitative scoring. This results in a one-sided risk assessment dimension, which can lead to significant deviations in the judgment of users' arrears risk, making it impossible to accurately classify the user's arrears risk level and resulting in insufficient targeting of early warning measures. Summary of the Invention

[0006] The purpose of this invention is to provide a method and system for risk warning of heating users with arrears based on user profiles, so as to achieve accurate and timely warning of the risk of heating users with arrears, and at the same time realize the dynamic adjustment of the user's heat consumption credit limit.

[0007] The technical solution adopted by the present invention to solve its technical problem is as follows: On the one hand, the present invention provides a risk warning method for users with heating arrears based on user profiles, including: extracting historical heat consumption data of heating user accounts in recent historical periods, dynamically setting heat consumption thresholds based on the current time point, and determining whether to send a payment reminder based on the current remaining heat consumption of the heating user account.

[0008] After the platform sends a payment reminder, it continuously monitors the real-time remaining heat consumption and the remaining heat consumption change curve during the triggered period.

[0009] If the slope of the real-time remaining heat consumption and the remaining heat consumption change curve meets the risk assessment triggering conditions, then the user's heat consumption risk assessment will be triggered.

[0010] Extract the heat consumption balance and payment amount from each historical payment in the heating user's account. Based on the heat consumption balance and payment amount, assess the risk score of payment time behavior and the risk score of payment amount behavior, and sum them up to determine the total arrears risk score.

[0011] The total score for overdue payment risk is matched with the score range corresponding to each set overdue payment risk level to obtain the overdue payment risk level of the heating user and issue corresponding early warning measures. At the same time, the current heat consumption credit limit is adjusted according to the overdue payment risk level of the heating user.

[0012] On the other hand, the present invention provides a risk warning system for users with overdue heating bills based on user profiles, including a payment reminder judgment module, a heat consumption continuous monitoring module, a risk assessment trigger module, an overdue payment risk scoring module, and a warning measure processing module.

[0013] The connections between the modules are as follows: the payment reminder judgment module communicates with the heat consumption continuous monitoring module; the risk assessment trigger module communicates with both the heat consumption continuous monitoring module and the overdue payment risk scoring module; and the overdue payment risk scoring module communicates with the early warning measures processing module.

[0014] The payment reminder judgment module extracts the historical heat consumption data of the heating user account in the recent historical period, dynamically sets the heat consumption threshold based on the current time, and determines whether to send a payment reminder based on the current remaining heat consumption of the heating user account.

[0015] The heat consumption monitoring module continuously monitors the real-time remaining heat consumption and the remaining heat consumption change curve during the triggered period after the platform sends a payment reminder.

[0016] The risk assessment trigger module will trigger a user heat risk assessment if the slope of the real-time remaining heat consumption and the remaining heat consumption change curve meets the risk assessment trigger conditions.

[0017] The arrears risk scoring module extracts the heat consumption balance and payment amount from the heating user's account for each historical payment. Based on the heat consumption balance and payment amount, it assesses the payment time behavior risk score and the payment amount behavior risk score, and adds them up to determine the total arrears risk score.

[0018] The early warning measures processing module matches the total score of the arrears risk with the score range corresponding to each set arrears risk level to obtain the arrears risk level of the heating user and issues corresponding early warning measures. At the same time, it adjusts the current heat consumption credit limit according to the arrears risk level of the heating user.

[0019] Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention extracts the historical heat consumption data of the heating user account in the recent historical period, dynamically sets the heat consumption threshold based on the current time point, and determines whether to send a payment reminder based on the current remaining heat consumption of the heating user account. It accurately matches the actual heat consumption pattern of the user in different unit time periods, so that the timing of the payment reminder is more in line with the user's heat consumption rhythm, laying the foundation for early intervention of the risk of arrears.

[0020] (2) This invention monitors the real-time remaining heat consumption and the remaining heat consumption change curve during the trigger period. If the slope of the real-time remaining heat consumption and the remaining heat consumption change curve meets the risk assessment triggering conditions, the user's heat consumption risk assessment is triggered. This allows for timely capture of the dynamic risk characteristics of the user's heat consumption, ensuring that the risk assessment is only triggered when the user's heat consumption risk actually increases, thereby improving the timeliness and accuracy of risk identification.

[0021] (3) This invention extracts the heat consumption balance and payment amount of each historical payment from the heating user's account, evaluates the payment time behavior risk score and payment amount behavior risk score, and sums them to determine the total arrears risk score. This solves the problem of risk judgment deviation caused by single-dimensional assessment, realizes the accurate quantification of user arrears risk, improves the accuracy of arrears risk level classification, and provides reference data for the later targeted early warning measures.

[0022] (4) This invention matches the arrears risk level of heating users according to the total arrears risk score, and adjusts the current heat consumption credit limit according to the arrears risk level of heating users, so as to realize differentiated and dynamic management of the heat consumption credit of heating users, establish a linkage management mechanism between the arrears risk level and the heat consumption credit limit, ensure the financial security of heating enterprises, and improve the flexibility of heating services. Attached Figure Description

[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram showing the connection of the method steps of the present invention.

[0025] Figure 2 This is a schematic diagram illustrating the step of determining whether to send a payment reminder in this invention.

[0026] Figure 3 This is a schematic diagram of the payment time behavior risk scoring and assessment steps in this invention.

[0027] Figure 4 This is a schematic diagram of the system module connections of the present invention. Detailed Implementation

[0028] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the invention. Furthermore, it should be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale.

[0029] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the invention or its application or use. Techniques, methods, and apparatus known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and apparatus should be considered part of the specification.

[0030] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0031] Please see Figure 1 As shown, the present invention provides a risk warning method for users with overdue heating fees based on user profiles, including: S1, extracting historical heat consumption data of heating user accounts in recent historical periods, dynamically setting a heat consumption threshold based on the current time point, and determining whether to send a payment reminder based on the current remaining heat consumption of the heating user account.

[0032] Considering the significant differences in heating behavior among different users, such as higher heating consumption for larger households and lower heating consumption for working users during the day, relying solely on a general fixed threshold cannot match individual habits. However, by extracting the user's recent historical heating data, the user's daily heating patterns can be accurately reflected, providing data support for calculating heating characteristics per unit time period and avoiding threshold settings that are out of touch with the user's actual heating scenario.

[0033] In one embodiment of the present invention, for example, the recent historical period is set to 15 days at the beginning of the heating season to quickly adapt to users' current habits; and the recent historical period is set to 30 days in the middle of the heating season to ensure data stability.

[0034] In another embodiment, the recent historical period can be flexibly adjusted by the implementer according to the heating cycle.

[0035] Based on this, such as Figure 2 As shown, the content of dynamically setting the heat consumption threshold is as follows: First, extract the historical heat consumption data within the recent historical period from the heating user account database, and obtain the historical heat consumption of each historical day in each unit time period based on the historical heat consumption data.

[0036] Secondly, the mean difference analysis of the historical heat consumption of each historical day in each unit time period and its adjacent unit time period is performed to obtain the rate of change of heat consumption in each unit time period.

[0037] The method for obtaining the rate of change of heat consumption in each unit time period is as follows: the historical heat consumption of each historical day in each unit time period is analyzed by difference with the historical heat consumption of the previous adjacent unit time period to obtain the rate of change of heat consumption of each historical day in each unit time period, and the average value is used to calculate the rate of change of heat consumption in each unit time period.

[0038] Then, determine the unit time period of the current time point. If the next unit time period corresponding to the unit time period of the current time point is a non-sleep period, then obtain the estimated heat consumption of the unit time period of the current time point based on the rate of change of heat consumption of the unit time period of the current time point, and use it as the heat consumption threshold.

[0039] It should be noted that the estimated heat consumption for the current time period is the product of the rate of change of heat consumption for the corresponding time period and one hour.

[0040] Finally, if the next time period corresponding to the current time point is a sleep period, then the rate of change of heat consumption corresponding to all sleep periods is filtered to determine the estimated total heat consumption for all sleep periods, which is then used as the heat consumption threshold.

[0041] It should be noted that the estimated total heat consumption for all sleep periods is the sum of the estimated heat consumption for each sleep period. Using the estimated total heat consumption as a heat consumption threshold is to accurately match the actual heat consumption characteristics of users during sleep periods, ensure that the payment reminder is timely, avoid the problem of heating being interrupted during sleep periods due to insufficient heat consumption margin, and improve the user experience.

[0042] In this invention, the day is divided into 24 time periods based on hours, while also taking into account the sleep patterns of most heating users, for example, sleep periods are... Non-sleep periods are The implementer can also adjust the schedule according to the sleeping and non-sleeping times of the heating users.

[0043] Specifically, the process of determining whether to send a payment reminder is as follows: obtain the current remaining heat consumption of the heating user account from the heating user management platform, compare the current remaining heat consumption with the set heat consumption threshold, and if the current remaining heat consumption is less than the set heat consumption threshold, it is determined that a payment reminder needs to be sent; otherwise, it is determined that no payment reminder needs to be sent.

[0044] This invention extracts historical heat consumption data from heating user accounts within recent historical periods, dynamically sets heat consumption thresholds based on the current time point, and determines whether to send a payment reminder based on the current remaining heat consumption of the heating user account. This accurately matches the actual heat consumption patterns of users in different time periods, making the timing of payment reminders more in line with the user's heat consumption rhythm, thus laying the foundation for early intervention against the risk of arrears.

[0045] S2. After the platform sends a payment reminder, continuously monitor the real-time remaining heat consumption and the remaining heat consumption change curve during the trigger period.

[0046] Considering that there are two key scenarios after users receive payment reminders: the first is timely payment and continued use of heat, and the second is delayed payment and continued consumption of remaining heat; simply sending reminders cannot distinguish between the two scenarios, let alone determine whether the user has a real risk of arrears. If the user delays payment, the heat will be consumed faster and the heating will be shut off sooner.

[0047] Therefore, continuous monitoring allows for real-time tracking of heat consumption after user reminders. Real-time remaining heat consumption directly reflects whether the current reserve can support subsequent heating needs and whether the user has paid on time. The remaining heat consumption curve reflects whether the rate of heat consumption is abnormal; accelerated consumption indicates higher risk. Combining these two metrics provides dynamic data for subsequent risk assessment, preventing missed opportunities for optimal intervention.

[0048] Based on this, the triggering period is constructed as follows: First, the real-time remaining heat consumption of the heating user account at multiple consecutive time points after the reminder is obtained, and a curve of the change in remaining heat consumption is constructed with the time point as the horizontal axis and the real-time remaining heat consumption as the vertical axis.

[0049] Next, the heat consumption threshold is multiplied by the preset heat consumption trigger ratio threshold to obtain the heat consumption trigger threshold. This threshold is then substituted into the remaining heat consumption change curve to output the time point corresponding to the heat consumption trigger threshold.

[0050] Finally, the current time point and the time point corresponding to the heat trigger threshold are used to form the trigger period.

[0051] It should be noted that the preset heat consumption trigger threshold is calibrated based on historical arrears data from heating companies. For example, if historical data shows that when remaining heat consumption drops to 0.5 times the heat consumption threshold, the probability of user arrears increases sharply from 10% to 60%, then the preset heat consumption trigger threshold can be set to 0.5 to avoid premature risk assessment due to an excessively high value or delayed assessment due to an excessively low value. Implementers can dynamically adjust this threshold based on annual arrears statistics, but must ensure that the value is between 0 and 1.

[0052] S3. If the slope of the real-time remaining heat consumption and the remaining heat consumption change curve meets the risk assessment triggering conditions, then the user's heat consumption risk assessment will be triggered.

[0053] Considering that after a user receives a payment reminder, if the remaining heat consumption is less than the heat consumption threshold at all points within the triggered period, it only indicates insufficient reserves. However, it cannot distinguish whether the user temporarily forgot to pay and is using heat to maintain basic consumption, or whether the user is deliberately delaying payment while consuming excessive heat. The risk of arrears differs significantly between the two scenarios. If the assessment is triggered solely by the remaining heat consumption being less than the heat consumption threshold, low-risk scenarios may be misjudged as high-risk, or high-risk scenarios may not be given sufficient attention.

[0054] Based on this, the risk assessment triggering conditions are as follows: a) The real-time remaining heat consumption at all time points during the triggering period is less than the heat consumption threshold.

[0055] b. The slope of the remaining heat consumption change curve during the triggering period is greater than the slope of the remaining heat consumption change curve in the previous unit period corresponding to the triggering period.

[0056] This invention monitors the real-time remaining heat consumption and the remaining heat consumption change curve during the trigger period. If the slope of the real-time remaining heat consumption and the remaining heat consumption change curve meets the risk assessment triggering conditions, a user heat consumption risk assessment is triggered. This allows for timely capture of the dynamic risk characteristics of user heat consumption, ensuring that risk assessment is only triggered when the user heat consumption risk actually increases, thus improving the timeliness and accuracy of risk identification.

[0057] S4. Extract the heat consumption balance and payment amount from the heating user's account for each historical payment. Based on the heat consumption balance and payment amount, assess the payment time behavior risk score and payment amount behavior risk score, and sum them up to determine the total arrears risk score.

[0058] Considering that the differences in arrears risk among different users stem from variations in their historical heating usage and payment habits—for example, some users habitually pay in advance when they have ample remaining heat, while others habitually delay payment when their remaining heat is extremely low—relying solely on the number of arrears cannot differentiate between them. For instance, even with the same arrears record, a user who pays when their remaining heat is 10% has a lower risk than a user who pays when their remaining heat is 0%.

[0059] Based on this, such as Figure 3 As shown, the risk assessment method for payment time behavior is as follows: First, based on the heat consumption balance corresponding to each historical payment, filter the number of historical payments within each heat consumption balance range after the payment reminder.

[0060] The heat usage balance intervals are categorized as follows: the heat usage balance interval from the heat usage threshold to zero; the heat usage balance interval from zero to the heat usage credit limit; and the heat usage balance interval exceeding the heat usage credit limit. Statistical analysis of historical payment frequency within each heat usage balance interval can provide a preliminary indication of which balance interval a user tends to pay in.

[0061] In this invention, the heat credit limit is less than 0.

[0062] The second step is to divide each heat consumption balance interval into several equally spaced sub-intervals, and based on the historical payment pairs of each heat consumption balance interval, filter the historical payment times corresponding to each sub-interval within each heat consumption balance interval.

[0063] The third step is to obtain the percentage of historical payments for each sub-interval within each heat consumption balance interval based on the number of historical payments within each heat consumption balance interval, and then, in combination with the set risk score weights for each sub-interval, to obtain the risk score correction ratio for each heat consumption balance interval.

[0064] The fourth step is to integrate the risk score correction ratios of each heat consumption balance interval, normalize the integration results, and obtain the payment time behavior risk score.

[0065] In a specific embodiment of the present invention, the normalization process can be performed using min-max standardization.

[0066] It should be noted that the risk score correction ratios for each heat consumption balance interval are weighted and summed for fusion. The lower the heat consumption balance corresponding to each heat consumption balance interval, the higher the weight of that interval. In one example of this invention, the heat consumption balance interval greater than the heat consumption credit limit has the lowest heat consumption balance, so its corresponding weight is set to 0.5; the weight of the heat consumption balance interval from zero to the heat consumption credit limit is set to 0.33; and the weight of the heat consumption balance interval from the heat consumption threshold to zero is set to 0.17. In other embodiments, the implementer can set the corresponding weights as needed.

[0067] Considering the fundamental differences in the delay risk of different sub-intervals, the smaller the balance of a sub-interval, the higher the delay risk. If all sub-intervals are given the same weight, the behavior of high-risk sub-intervals will not be given priority consideration. Through weighted analysis, the proportion of historical payment times of high-risk sub-intervals has a greater impact on the score. The risk score correction ratio provides a quantitative coefficient of risk level for the subsequent total score, avoiding the underestimation of risk due to equal weights.

[0068] Based on this, the steps for setting the risk score weights for each sub-interval are as follows: First, sort each sub-interval in descending order of heat consumption balance to obtain the sorting number of each sub-interval.

[0069] Secondly, the sorting numbers of each sub-interval are summed to obtain the total sorting number. The ratio of the sorting number of each sub-interval to the total sorting number is analyzed, and the ratio is used as the risk score weight of each sub-interval.

[0070] Specifically, the risk assessment method for the payment amount behavior is as follows: First, based on the payment amount corresponding to each historical payment, the unit price of the heat consumption is determined to determine the heat consumption for recharge, and this is added to the heat consumption balance to obtain the remaining heat consumption.

[0071] The second step is to compare the remaining heat consumption corresponding to each historical payment with its corresponding heat consumption threshold, count the number of historical payments that did not exceed the heat consumption threshold, and determine the percentage of historical payments that did not exceed the heat consumption threshold.

[0072] The third step is to analyze the percentage of historical payments with negative residual heat based on the residual heat corresponding to each payment, and combine this with the percentage of historical payments that did not exceed the heat consumption threshold to assess the risk score of payment behavior.

[0073] It should be noted that since the remaining heat consumption is negative, it means that the user has had their heating stopped due to insufficient funds, which is considered an overdue payment. Multiplying this percentage by the percentage of historical payments that did not exceed the heat consumption threshold, and using the product as the risk score for payment behavior, can make the risk of overdue payments have a greater impact on the score. This avoids ignoring the high risk of overdue payments by only looking at those that did not exceed the threshold, and ensures that the score can comprehensively cover the amount dimension.

[0074] This invention extracts the heat consumption balance and payment amount from each historical payment in the heating user's account, assesses the risk score of payment time behavior and the risk score of payment amount behavior, and sums them to determine the total arrears risk score. This solves the problem of risk judgment bias caused by single-dimensional assessment, realizes the accurate quantification of user arrears risk, improves the accuracy of arrears risk level classification, and provides reference data for the subsequent targeted early warning measures.

[0075] S5. Match the total arrears risk score with the score range corresponding to each set arrears risk level to obtain the arrears risk level of the heating user and issue corresponding early warning measures. At the same time, adjust the current heat consumption credit limit according to the arrears risk level of the heating user.

[0076] It should be noted that the aforementioned levels of delinquency risk include low, medium, and high delinquency risk. Differentiated measures are implemented for each level. For example, users with low delinquency risk only receive SMS reminders, users with medium delinquency risk receive payment reminders via SMS and voice call, and users with high delinquency risk are contacted by phone and visited in person to verify their payment intentions. This differentiated approach ensures that users with high delinquency risk receive strong intervention to prevent their debts from worsening, while ensuring that users with low delinquency risk are not subjected to unnecessary disturbances, thus improving their experience. Simultaneously, it optimizes the efficiency of enterprise resource allocation and avoids cost waste.

[0077] Specifically, the process for adjusting the current heat consumption credit limit is as follows: S51, compare the heat user's arrears risk level with the corresponding account's current arrears risk level. If the heat user's arrears risk level is the same as the account's current arrears risk level, then there is no need to adjust the current heat consumption credit limit.

[0078] S52. Conversely, when the risk level of a heating user's arrears is greater than the current risk level of the account, the deviation between the total arrears risk score and the maximum value of the corresponding score range of the current arrears risk level of the account is obtained, and the credit limit correction value is obtained by combining it with the current heat consumption credit limit.

[0079] The process of analyzing the deviation value is as follows: obtain the absolute difference between the total score of the arrears risk and the maximum value of the score range corresponding to the current arrears risk level of the account, and use the ratio of the absolute difference to the maximum value of the score range corresponding to the current arrears risk level of the account as the deviation value.

[0080] The credit limit correction value is the product of the current heat credit limit and the corresponding deviation value.

[0081] S53. The difference between the current heat consumption credit limit and the credit limit adjustment value is used to obtain the adjusted heat consumption credit limit.

[0082] S54. When the heating user's arrears risk level is lower than the account's current arrears risk level, the deviation between the minimum value of the scoring range corresponding to the account's current arrears risk level and the total arrears risk score is obtained, and the credit limit correction value is obtained by combining it with the current heat consumption credit limit.

[0083] S55. The current heat consumption credit limit is added to the credit limit adjustment value to obtain the adjusted heat consumption credit limit.

[0084] This invention matches the arrears risk level of heating users based on the total arrears risk score, and adjusts the current heat consumption credit limit according to the arrears risk level. This achieves differentiated and dynamic management of heating users' heat consumption credit, establishes a linkage management mechanism between arrears risk level and heat consumption credit limit, ensures the financial security of heating companies, and improves the flexibility of heating services.

[0085] On the other hand, such as Figure 4 As shown, the present invention provides a risk early warning system for users with overdue heating fees based on user profiles, including a payment reminder judgment module, a heat consumption continuous monitoring module, a risk assessment trigger module, an overdue payment risk scoring module, and an early warning measure processing module.

[0086] The connections between the modules are as follows: the payment reminder judgment module communicates with the heat consumption continuous monitoring module; the risk assessment trigger module communicates with both the heat consumption continuous monitoring module and the overdue payment risk scoring module; and the overdue payment risk scoring module communicates with the early warning measures processing module.

[0087] The payment reminder judgment module extracts the historical heat consumption data of the heating user account in the recent historical period, dynamically sets the heat consumption threshold based on the current time, and determines whether to send a payment reminder based on the current remaining heat consumption of the heating user account.

[0088] The heat consumption monitoring module continuously monitors the real-time remaining heat consumption and the remaining heat consumption change curve during the triggered period after the platform sends a payment reminder.

[0089] The risk assessment trigger module will trigger a user heat risk assessment if the slope of the real-time remaining heat consumption and the remaining heat consumption change curve meets the risk assessment trigger conditions.

[0090] The arrears risk scoring module extracts the heat consumption balance and payment amount from the heating user's account for each historical payment. Based on the heat consumption balance and payment amount, it assesses the payment time behavior risk score and the payment amount behavior risk score, and adds them up to determine the total arrears risk score.

[0091] The early warning measures processing module matches the total score of the arrears risk with the score range corresponding to each set arrears risk level to obtain the arrears risk level of the heating user and issues corresponding early warning measures. At the same time, it adjusts the current heat consumption credit limit according to the arrears risk level of the heating user.

[0092] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0093] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0094] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0095] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0096] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A heat supply arrears user risk early warning method based on user portraits, characterized in that, include: Extract historical heat consumption data of heating user accounts in recent historical periods, dynamically set heat consumption thresholds based on the current time point, and determine whether to send a payment reminder based on the current remaining heat consumption of the heating user account. After the platform sends a payment reminder, it continuously monitors the real-time remaining heat consumption and the remaining heat consumption change curve during the triggered period. The triggered period is composed of the following: Obtain the real-time remaining heat consumption of heating user accounts at multiple consecutive time points after the reminder, and construct a curve of the change in remaining heat consumption; The heat consumption threshold is calculated by multiplying the heat consumption threshold with the preset heat consumption trigger ratio threshold. The heat consumption trigger threshold is then substituted into the remaining heat consumption change curve to output the time point corresponding to the heat consumption trigger threshold. The current time point and the time point corresponding to the heat consumption trigger threshold constitute the trigger period. If the slope of the real-time remaining heat consumption and the remaining heat consumption change curve meets the risk assessment triggering conditions, then the user's heat consumption risk assessment will be triggered. Extract the heat consumption balance and payment amount from the heating user's account for each historical payment. Based on the heat consumption balance and payment amount, assess the payment time behavior risk score and payment amount behavior risk score, and sum them up to determine the total arrears risk score. The method for assessing the risk score of payment time behavior is as follows: Based on the remaining heat consumption corresponding to each historical payment, filter the number of historical payments within each heat consumption balance range after the payment reminder; Each heat consumption balance interval is divided into several equally spaced sub-intervals. Based on the historical payment for each heat consumption balance interval within each heat consumption balance interval, the number of historical payments corresponding to each sub-interval within each heat consumption balance interval is filtered. Based on the number of historical payments within each heat consumption balance interval, the percentage of historical payments for each sub-interval within each heat consumption balance interval is obtained. Combined with the set risk score weights for each sub-interval, the risk score correction ratio for each heat consumption balance interval is obtained by weighting. The risk score correction ratios of each heat consumption balance range are integrated, and the integration results are normalized to obtain the payment time behavior risk score. The method for assessing the risk score of payment amount behavior is as follows: Based on the payment amount corresponding to each historical payment, the unit price of the heat consumption is determined to determine the heat consumption for recharge, and this is added to the heat consumption balance to obtain the remaining heat consumption. Compare the remaining heat consumption corresponding to each historical payment with its corresponding heat consumption threshold, count the number of historical payments that did not exceed the heat consumption threshold, and determine the percentage of historical payments that did not exceed the heat consumption threshold. Based on the remaining heat consumption corresponding to each historical payment, the percentage of historical payments with negative remaining heat consumption is analyzed. Combined with the percentage of historical payments that did not exceed the heat consumption threshold, the risk score of payment amount behavior is assessed. The total score of the arrears risk is matched with the score range corresponding to each set arrears risk level to obtain the arrears risk level of the heating user and issue corresponding early warning measures. At the same time, the current heat consumption credit limit is adjusted according to the arrears risk level of the heating user.

2. The user portrait-based risk early warning method for heating arrears users according to claim 1, characterized in that: The details of the dynamically set heat threshold are as follows: Extract historical heat consumption data for recent historical periods from the heating user account database, and obtain historical heat consumption data for each historical day and each time period based on the historical heat consumption data; The average difference analysis of the historical heat consumption of each historical day in each unit time period and its adjacent unit time period is performed to obtain the rate of change of heat consumption in each unit time period. Determine the unit time period in which the current time point is located. If the next unit time period corresponding to the unit time period in which the current time point is located is a non-sleep period, then obtain the estimated heat consumption of the unit time period in which the current time point is located based on the rate of change of heat consumption in the unit time period in which the current time point is located, and use it as the heat consumption threshold. If the next time period corresponding to the current time point is a sleep period, then filter the rate of change of heat consumption for all sleep periods, determine the estimated total heat consumption for all sleep periods, and use it as the heat consumption threshold.

3. The method for risk warning of users with heating arrears based on user profiles according to claim 2, characterized in that: The process for determining whether to send a payment reminder is as follows: The system retrieves the current remaining heat consumption of each heating user's account from the heating user management platform and compares it with a set heat consumption threshold. If the current remaining heat consumption is less than the set heat consumption threshold, a payment reminder is sent; otherwise, no payment reminder is sent.

4. The method for risk warning of users with heating arrears based on user profiles according to claim 1, characterized in that: The risk assessment is triggered under the following conditions: a) The real-time remaining heat consumption at all time points within the trigger period is less than the heat consumption threshold; b. The slope of the remaining heat consumption change curve during the triggering period is greater than the slope of the remaining heat consumption change curve in the previous unit period corresponding to the triggering period.

5. The method for risk warning of users with heating arrears based on user profiles according to claim 1, characterized in that: The steps for setting the risk score weights for each sub-interval are as follows: Sort each sub-interval in descending order of heat consumption balance to obtain the sorting number of each sub-interval; The sorting numbers of each sub-interval are summed to obtain the total sorting number. The ratio of the sorting number of each sub-interval to the total sorting number is analyzed, and the ratio is used as the risk score weight of each sub-interval.

6. The method for risk warning of users with heating arrears based on user profiles according to claim 1, characterized in that: The process for adjusting the current heat consumption credit limit is as follows: Compare the heating user's arrears risk level with the corresponding account's current arrears risk level. If the heating user's arrears risk level is the same as the account's current arrears risk level, then there is no need to adjust the current heat consumption credit limit. Conversely, when the risk level of a heating user's arrears is greater than the current risk level of the account, the deviation between the total arrears risk score and the maximum value of the corresponding score range of the current arrears risk level of the account is obtained, and the credit limit correction value is obtained by combining it with the current heat consumption credit limit. The adjusted heat credit limit is obtained by subtracting the current heat credit limit from the credit limit adjustment value. When a heating user's arrears risk level is lower than the current arrears risk level of the account, the deviation between the minimum value of the scoring range corresponding to the current arrears risk level of the account and the total arrears risk score is obtained, and the credit limit correction value is obtained by combining it with the current heat consumption credit limit. The adjusted heat consumption credit limit is obtained by adding the current heat consumption credit limit to the credit limit adjustment value.

7. A user profile-based risk warning system for users with overdue heating payments, implemented using the user profile-based risk warning method for users with overdue heating payments as described in any one of claims 1-6, characterized in that, include: The payment reminder judgment module extracts the historical heat consumption data of the heating user account in the recent historical period, dynamically sets the heat consumption threshold in combination with the current time point, and determines whether to send a payment reminder based on the current remaining heat consumption of the heating user account. The heat consumption monitoring module continuously monitors the real-time remaining heat consumption and the remaining heat consumption change curve during the triggered period after the platform sends a payment reminder; The risk assessment trigger module will trigger a user heat risk assessment if the slope of the real-time remaining heat consumption and the remaining heat consumption change curve meets the risk assessment trigger conditions. The arrears risk scoring module extracts the heat consumption balance and payment amount from the heating user's account at each historical payment time. Based on the heat consumption balance and payment amount, it assesses the payment time behavior risk score and the payment amount behavior risk score, and adds them up to determine the total arrears risk score. The early warning measures processing module matches the total score of the arrears risk with the score range corresponding to each set arrears risk level to obtain the arrears risk level of the heating user and issues corresponding early warning measures. At the same time, it adjusts the current heat consumption credit limit according to the arrears risk level of the heating user.

Images