Intelligent repayment strategy generation system and method driven by multi-source information analysis

Abstract

The present invention relates to the field of financial technology. The present invention discloses an intelligent repayment strategy generation system and method driven by multi-source information analysis, including determining a repayment ability value based on real-time financial information and historical repayment information, determining a repayment willingness value based on repayment willingness information, and then determining an initial repayment strategy based on the repayment ability value, the repayment willingness value and a preset standard repayment strategy, determining user burden information based on real-time financial information and the initial repayment strategy, calculating a repayment burden value based on the user burden information, and finally optimizing the initial repayment strategy based on an intelligent optimization algorithm and the repayment burden value. In this way, the present invention can dynamically adjust the repayment strategy by comprehensively analyzing real-time financial information, historical repayment information and repayment willingness information in combination with an intelligent optimization algorithm. The optimal repayment strategy finally generated is dynamic and flexible, which improves user experience and enhances the service capabilities and user trust of lending institutions.

Description

Technical Field

[0001] The present invention relates to the field of financial technology, and more specifically, to an intelligent repayment strategy generation system and method driven by multi-source information analysis. Background Art

[0002] Existing repayment strategies often lack personalization and flexibility, and fail to fully consider the user's actual financial situation and repayment ability, resulting in inaccurate strategy design. For example, some strategies may be set based on the user's credit score or historical repayment record, but ignore the user's current income status, spending pressure and sudden financial changes. This "one-size-fits-all" approach can easily cause users to encounter difficulties in the repayment process, which may result in excessive monthly payment pressure or default risk.

[0003] For example, the Chinese patent application with publication number CN111210332A provides a method, device and electronic device for generating a post-loan management strategy. The patent extracts target users by screening user loan data, and generates a repayment probability by combining user characteristics with basic data input into a repayment model; a post-loan management strategy for the target user is formulated based on the repayment probability. The Chinese patent application with publication number CN106952156A provides a method and device for generating an installment repayment plan. The patent obtains repayment elements of preset groups and their input quantities, determines unknown quantities using an association strategy, and formulates an installment repayment plan strategy by combining the input quantities and the determined unknown quantities.

[0004] Existing technologies usually formulate repayment strategies based on user account information, but these strategies lack dynamic adjustment mechanisms and are difficult to reflect changes in users' financial status in real time. They are unable to automatically optimize repayment plans based on users' latest financial status, which not only affects users' repayment experience but may also undermine their trust in lending institutions.

[0005] In view of this, the present invention proposes an intelligent repayment strategy generation system and method driven by multi-source information analysis to solve the above problems. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides an intelligent repayment strategy generation system and method driven by multi-source information analysis.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] First, it provides an intelligent repayment strategy generation method driven by multi-source information analysis, including:

[0009] Obtain the target user's real-time financial information, historical repayment information, and repayment willingness information, determine the repayment ability value based on the real-time financial information and historical repayment information, and determine the repayment willingness value based on the repayment willingness information;

[0010] Determine an initial repayment strategy based on the repayment ability value, the repayment willingness value, and the preset standard repayment strategy, determine user burden information based on real-time financial information and the initial repayment strategy, and calculate the repayment burden value based on the user burden information;

[0011] The initial repayment strategy is optimized according to the intelligent optimization algorithm and the repayment burden value to obtain the optimal repayment strategy.

[0012] Furthermore, the method of obtaining repayment willingness information includes:

[0013] Obtain user communication data and user behavior data, perform semantic analysis on the user communication data to obtain the repayment keyword frequency, use each control in the user behavior data as a node, and the relationship between controls as an edge, build a behavior logic graph, and use the repayment keyword frequency and behavior logic graph as repayment intention information.

[0014] Furthermore, the real-time financial information includes income amount, expenditure amount, debt amount and credit limit value, and the historical repayment information includes historical repayment amount, historical overdue number and historical repayment frequency. The method for determining the repayment capacity value based on the real-time financial information and historical repayment information includes:

[0015] Calculate the difference between the income amount and the expenditure amount to get the disposable amount, take the ratio of the disposable amount to the income amount as the disposable income ratio, calculate the sum of the credit limit value and the historical repayment amount as the repayment potential value, take the ratio of the debt amount to the repayment potential value as the debt pressure ratio, take the difference between the historical repayment frequency and the historical overdue times as the historical fulfillment rate value, take the weighted sum of the disposable income ratio, debt pressure ratio and historical fulfillment rate value to get the repayment ability value.

[0016] Furthermore, the method for determining the repayment willingness value according to the repayment willingness information includes:

[0017] The positive and negative values ​​are calculated according to the repayment keyword frequency, and the positive and negative values ​​are weightedly summed to obtain the keyword comprehensive value. The sum of the edge weights in the behavior logic graph is calculated according to the graph algorithm, and the keyword comprehensive value and the edge weight sum are cascaded and integrated to obtain the repayment willingness value.

[0018] Furthermore, the method for determining the initial repayment strategy based on the repayment ability value, the repayment willingness value and the preset standard repayment strategy includes:

[0019] Input the repayment ability value, repayment willingness value and preset standard repayment strategy into the pre-built strategy adjustment model to obtain the initial repayment strategy;

[0020] The construction method of the strategy adjustment model includes:

[0021] Acquire a sample data set, wherein the sample data set includes a historical repayment ability value, a historical repayment willingness value, a historical standard repayment strategy, and a historical initial repayment strategy;

[0022] Divide the sample data set into a sample training set and a sample test set, and build a regression network;

[0023] The historical repayment ability value, historical repayment willingness value and historical standard repayment strategy in the sample training set are used as the input data of the regression network, and the historical initial repayment strategy in the sample training set is used as the output data of the regression network. The regression network is trained to obtain an initial regression network for predicting the initial repayment strategy.

[0024] The initial regression network is tested using the sample test set, and the initial regression network with an output smaller than the preset error value is used as the strategy adjustment model.

[0025] Furthermore, the method for determining user burden information based on real-time financial information and initial repayment strategy includes:

[0026] Determine the repayment amount and interest amount for each period in the initial repayment strategy, take the ratio of each repayment amount to the disposable amount as the repayment burden ratio, take the ratio of each interest amount to the disposable amount as the interest burden ratio, take the ratio of each repayment amount to the income amount as the monthly payment-to-income ratio, and determine the repayment burden ratio, interest burden ratio and monthly payment-to-income ratio as user burden information.

[0027] Furthermore, the method for calculating the repayment burden value according to the user burden information includes:

[0028] RBV= ;

[0029] In the formula, RBV is the repayment burden value, IBR is the interest burden ratio, REB is the repayment burden ratio, and MPR is the monthly payment-to-income ratio. is the inverse cotangent function, is the hyperbolic cosine function, is a logarithmic function with base e.

[0030] Furthermore, the intelligent optimization algorithm is a particle swarm optimization algorithm. The initial repayment strategy is optimized according to the particle swarm optimization algorithm and the repayment burden value. The method for obtaining the optimal repayment strategy includes:

[0031] S301: Define the particle's position vector as ,in, Representative Initial repayment strategies form the initial population, and the position and velocity of each particle are randomly initialized. is the total number of initial repayment strategies;

[0032] S302: Define a fitness function and constraints. The fitness function includes:

[0033] ;

[0034] In the formula, is the fitness function, is the repayment burden value;

[0035] The constraints include:

[0036] ;

[0037] In the formula, is the repayment period in the initial repayment strategy, is the maximum standard value of the repayment period, is the repayment amount per period in the initial repayment strategy, The minimum standard value for the repayment amount per period;

[0038] S303: updating the particle speed according to the particle's current speed, the particle's optimal position and the global optimal position;

[0039] S304: Update the particle position according to the new velocity:

[0040] = ;

[0041] In the formula, For particles exist The updated position at any time. For particles exist The location at the moment, For particles exist The speed of the moment;

[0042] S305: Compare the current fitness value of each particle with its historical best fitness value. If the current fitness value is better, update the optimal position of the particle. , compare the current fitness values ​​corresponding to all particles, and select the optimal position of the particle corresponding to the best current fitness value As the global optimal position ;

[0043] S306: Repeat S303-S305 above until the preset number of iterations is reached, and the final global optimal position Represents the optimal repayment strategy.

[0044] Furthermore, the method of updating the particle velocity includes:

[0045] ;

[0046] In the formula, For particles exist The speed of time, For particles exist The speed of time, is the inertia weight, is the particle learning factor, is the social learning factor, , Both A random number between is the current global optimal position of the entire particle swarm, For particles The current optimal position of the particle.

[0047] In a second aspect, a multi-source information analysis-driven intelligent repayment strategy generation system is provided, which is used to implement the multi-source information analysis-driven intelligent repayment strategy generation method described above, including:

[0048] Data analysis module: used to obtain the target user's real-time financial information, historical repayment information and repayment willingness information, determine the repayment ability value based on the real-time financial information and historical repayment information, and determine the repayment willingness value based on the repayment willingness information;

[0049] Strategy generation module: used to determine the initial repayment strategy based on the repayment ability value, the repayment willingness value and the preset standard repayment strategy, determine the user burden information based on the real-time financial information and the initial repayment strategy, and calculate the repayment burden value based on the user burden information;

[0050] Strategy optimization module: used to optimize the initial repayment strategy based on the intelligent optimization algorithm and the repayment burden value to obtain the optimal repayment strategy.

[0051] In a third aspect, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned intelligent repayment strategy generation method driven by multi-source information analysis when executing the computer program.

[0052] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed, the intelligent repayment strategy generation method driven by the multi-source information analysis is implemented.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] The present invention first determines the repayment capacity value based on real-time financial information and historical repayment information, determines the repayment willingness value based on the repayment willingness information, then determines the initial repayment strategy based on the repayment capacity value, the repayment willingness value and the preset standard repayment strategy, determines the user burden information based on the real-time financial information and the initial repayment strategy, calculates the repayment burden value based on the user burden information, and finally optimizes the initial repayment strategy according to the intelligent optimization algorithm and the repayment burden value to obtain the optimal repayment strategy. In this way, the present invention can dynamically adjust the repayment strategy by comprehensively analyzing the real-time financial information, historical repayment information and repayment willingness information in combination with the intelligent optimization algorithm, thereby solving the problem of lack of dynamic adjustment and personalized design in the prior art. The optimal repayment strategy finally generated is dynamic and flexible, which improves the user experience and enhances the service capability and user trust of the lending institution. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 A schematic diagram of the process of the intelligent repayment strategy generation method driven by multi-source information analysis in the present invention;

[0056] Figure 2 It is a structural diagram of the intelligent repayment strategy generation system driven by multi-source information analysis in the present invention;

[0057] Figure 3 A schematic diagram of the process of obtaining repayment willingness information in the present invention;

[0058] Figure 4 It is a schematic diagram of a process of determining a repayment willingness value according to repayment willingness information in the present invention. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] In the technical solution of the present invention, the user information (including but not limited to user personal information, user image information, user device information, such as location information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with the relevant laws, regulations and standards of the relevant countries and regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0061] Example 1

[0062] See also Figure 1 As shown, this embodiment discloses a method for generating an intelligent repayment strategy driven by multi-source information analysis, including:

[0063] S10: obtaining the target user's real-time financial information, historical repayment information, and repayment willingness information, determining the repayment ability value based on the real-time financial information and historical repayment information, and determining the repayment willingness value based on the repayment willingness information;

[0064] In this embodiment, real-time financial information refers to data that can dynamically reflect the user's current economic status. Real-time financial information includes but is not limited to income amount, expenditure amount, debt amount and credit limit value. It can be understood that the income information, expenditure information, etc. mentioned above can be the income and expenditure of the current month, or the average income or average expenditure of the previous few months. This embodiment does not limit this.

[0065] It should be added that historical repayment information refers to data that reflects the user's previous repayment behavior and repayment records. Historical repayment information includes but is not limited to historical repayment amounts, historical overdue times, and historical repayment frequencies. The historical repayment amount mentioned above can be the average repayment amount of the previous few years. Similarly, the historical overdue times can be the average overdue times of the previous few years. This embodiment does not go into too much detail on this.

[0066] In an optional embodiment, the repayment intention information refers to data reflecting the user's subjective repayment intention and repayment attitude, such as Figure 3 As shown, the methods for obtaining repayment willingness information include:

[0067] Obtain user communication data and user behavior data, perform semantic analysis on the user communication data to obtain the repayment keyword frequency, use each control in the user behavior data as a node, and the relationship between controls as an edge, build a behavior logic graph, and use the repayment keyword frequency and behavior logic graph as repayment intention information.

[0068] It should be noted that user communication data refers to voice data or text data about repayment between bank customer service and users, and repayment keywords refer to keywords expressing repayment commitment, keywords expressing repayment intention, keywords expressing repayment pressure, etc. For example, repayment keywords can be "repay on time", "how to repay", "tight funds", and "what will happen if overdue". The repayment keywords can be pre-classified to obtain positive repayment keywords and negative repayment keywords, so that the user's subjective repayment attitude and willingness can be evaluated according to the frequency of repayment keywords.

[0069] In the above, controls refer to interactive components in software systems or applications, which are used for user input or operation. Exemplarily, controls can be number boxes, drop-down menus, sliders, submit buttons, etc. Number boxes are used to input numerical data, such as repayment amounts or installments. Drop-down menus are used to select repayment methods (such as installments, full repayment). Sliders are used by users to adjust repayment amounts or repayment time ranges. Submit buttons are used to confirm operations, such as confirming repayments or applying for installments. The relationship between controls refers to the logical association and operation sequence between different controls during the user's interaction process. These relationships can reflect the user's operation path, intention, and behavior pattern, such as sequential relationships, dependency relationships, and conditional relationships. Taking sequential relationships as an example, sequential relationships are used to describe the process of users operating controls in chronological order. Exemplarily, the user first selects the repayment method in the drop-down menu, and then adjusts the slider to set the repayment time range.

[0070] It can be understood that the behavior logic diagram refers to a visual model that represents the user interaction behavior logic in the form of a graph structure. By taking the controls operated by the user as nodes and the logical associations and operation sequences between the controls as edges, it intuitively reflects the user's operation path and behavior pattern. For example, after the user logs in to the system, he clicks the menu button to enter the "Repayment Center". In the "Repayment Center" interface, the user selects the "Installment Repayment" option in the drop-down menu. The system displays detailed information about the installment repayment, including the number of periods, interest rate and monthly payment amount. The user clicks the slider to adjust the installment time range and views the monthly payment amount corresponding to different periods. The user clicks the Submit button to confirm the installment repayment plan. Then the above nodes are "Repayment Center", "Installment Repayment", "Slider" and "Submit Button" respectively, and the edges can be the sequential relationship between "Repayment Center" and "Installment Repayment", and the sequential relationship between "Installment Repayment" and "Slider", etc.

[0071] Methods for determining repayment capacity values ​​based on real-time financial information and historical repayment information include:

[0072] Calculate the difference between the income amount and the expenditure amount to get the disposable amount, take the ratio of the disposable amount to the income amount as the disposable income ratio, calculate the sum of the credit limit value and the historical repayment amount as the repayment potential value, take the ratio of the debt amount to the repayment potential value as the debt pressure ratio, take the difference between the historical repayment frequency and the historical overdue times as the historical fulfillment rate value, take the weighted sum of the disposable income ratio, debt pressure ratio and historical fulfillment rate value to get the repayment ability value.

[0073] Methods for obtaining the repayment capacity value by weighted summing the disposable income ratio, debt pressure ratio and historical performance rate include:

[0074] RAV=u1DIR-u2DPR+u3HCR;

[0075] In the formula, RAV is the repayment capacity value, DIR is the disposable income ratio, DPR is the debt pressure ratio, and HCR is the historical performance rate value. , , are all weight factors greater than 0.

[0076] In this embodiment, by acquiring real-time financial information and historical repayment information, combined with repayment willingness information, the user's current economic status and repayment habits can be comprehensively and dynamically reflected. The disposable income ratio and debt pressure ratio are calculated based on real-time financial information, and combined with the historical fulfillment rate value. The repayment ability value finally obtained by weighted summation can accurately quantify the user's comprehensive repayment ability, and can make accurate decision support when providing users with personalized repayment strategy recommendations in the future.

[0077] like Figure 4 As shown, the method for determining the repayment willingness value according to the repayment willingness information includes:

[0078] The positive and negative values ​​are calculated according to the repayment keyword frequency, and the positive and negative values ​​are weightedly summed to obtain the keyword comprehensive value. The sum of the edge weights in the behavior logic graph is calculated according to the graph algorithm, and the keyword comprehensive value and the edge weight sum are cascaded and integrated to obtain the repayment willingness value.

[0079] It should be noted that the above-mentioned positive value can be the frequency value of the positive repayment attitude in the keyword, such as "repay on time" and "how to repay", and the corresponding negative value can be the frequency value of the negative repayment attitude in the keyword, such as "tight funds" and "what will happen if overdue". The above-mentioned graph algorithm can be the shortest path algorithm of the weighted graph or the centrality analysis algorithm of the weighted graph. For example, taking the centrality analysis algorithm of the weighted graph as an example, the PageRank algorithm or the betweenness centrality algorithm is used to calculate the importance of the weight of each edge in the behavior logic graph to the overall behavior path. The sum of the edge weights reflects the concentration and smoothness of the operations in the user behavior logic graph, and indirectly reveals the user's behavioral preference for repayment. It can be understood that the technicians in this field have set the corresponding weight for each edge in advance based on expert experience. In this way, by calculating the sum of the edge weights in the behavior logic graph, the user's operating habits and behavioral preferences on the repayment behavior path can be directly quantified.

[0080] It should be added that the cascade fusion mentioned above refers to taking the keyword comprehensive value and the sum of the edge weights in the behavior logic graph as two independent feature dimensions, and combining them in series (cascade) in order to form a multi-dimensional feature vector, rather than a simple weighted sum. The keyword comprehensive value and the sum of the edge weights will not overlap each other or lose their independent characteristics when cascaded, ensuring that each feature can reflect its unique role.

[0081] S20: determining an initial repayment strategy based on the repayment ability value, the repayment willingness value, and the preset standard repayment strategy, determining user burden information based on the real-time financial information and the initial repayment strategy, and calculating the repayment burden value based on the user burden information;

[0082] In this embodiment, the standard repayment strategy refers to a basic repayment plan or scheme provided to users based on industry norms or rules preset by lending institutions. The standard repayment strategy may include repayment method, repayment period, repayment frequency, repayment amount, interest rate standard, etc.

[0083] Methods for determining the initial repayment strategy based on the repayment ability value, the repayment willingness value, and the preset standard repayment strategy include:

[0084] The repayment ability value, repayment willingness value, and preset standard repayment strategy are input into the pre-built strategy adjustment model to obtain the initial repayment strategy.

[0085] The construction method of the strategy adjustment model includes:

[0086] Acquire a sample data set, wherein the sample data set includes a historical repayment ability value, a historical repayment willingness value, a historical standard repayment strategy, and a historical initial repayment strategy;

[0087] Divide the sample data set into a sample training set and a sample test set, and build a regression network;

[0088] The historical repayment ability value, historical repayment willingness value and historical standard repayment strategy in the sample training set are used as the input data of the regression network, and the historical initial repayment strategy in the sample training set is used as the output data of the regression network. The regression network is trained to obtain an initial regression network for predicting the initial repayment strategy.

[0089] The initial regression network is tested using a sample test set, and the initial regression network with an output smaller than a preset error value is used as a strategy adjustment model. The initial regression network is a deep neural network model.

[0090] It should be noted that the initial repayment strategy in this embodiment is obtained by adjusting the standard repayment strategy. The repayment ability value represents the user's objective repayment ability based on his or her real-time financial information and historical repayment information, and the repayment willingness value represents the user's subjective intention and attitude towards repayment. Therefore, when the repayment ability value and the repayment willingness value are both larger, the standard repayment strategy should be positively adjusted. Exemplarily, the standard repayment strategy includes repayment method, repayment period, repayment frequency, repayment amount and interest rate standard. When the repayment ability value or the repayment willingness value is larger, a more flexible repayment method can be provided, such as adding an early repayment option or allowing a higher repayment amount. When the user has a higher repayment ability and repayment willingness, it can be recommended to shorten the repayment period to reduce the overall interest cost. When the user has a strong repayment willingness, the repayment frequency can be increased (such as from monthly to weekly repayment) to help the user quickly repay the loan.

[0091] Methods for determining user burden information based on real-time financial information and initial repayment strategies include:

[0092] Determine the repayment amount and interest amount for each period in the initial repayment strategy, take the ratio of each repayment amount to the disposable amount as the repayment burden ratio, take the ratio of each interest amount to the disposable amount as the interest burden ratio, take the ratio of each repayment amount to the income amount as the monthly payment-to-income ratio, and determine the repayment burden ratio, interest burden ratio and monthly payment-to-income ratio as user burden information.

[0093] It is understandable that since the initial repayment strategy is obtained by adjusting the standard repayment strategy, the initial repayment strategy also includes the repayment method, repayment period, repayment frequency, repayment amount and interest rate standard, etc., then the user's repayment amount and interest amount for each period can be determined based on the initial repayment strategy.

[0094] The method for calculating the repayment burden value according to the user burden information includes:

[0095] RBV= ;

[0096] In the formula, RBV is the repayment burden value, IBR is the interest burden ratio, REB is the repayment burden ratio, and MPR is the monthly payment-to-income ratio. is the inverse cotangent function, is the hyperbolic cosine function, is a logarithmic function with base e.

[0097] Among them, taking the interest burden ratio and the repayment burden ratio as an example, the larger the interest burden ratio or the repayment burden ratio, the greater the user's repayment pressure or financial burden. From the above content, it can be seen that the repayment burden value is positively correlated with the user's repayment pressure.

[0098] This embodiment first obtains real-time financial information to ensure that the user burden information can reflect the user's current economic situation in real time, and then combines the initial repayment strategy to calculate the user's core indicators such as the repayment burden ratio, interest burden ratio and monthly payment-to-income ratio. The repayment burden value calculated in this way can quantify the user's overall financial pressure and be linked to the need for further adjustment of the initial repayment strategy. This sequential combination ensures that the user burden information does not just remain at the surface description of financial data, but is combined with personalized strategy adjustments to dynamically reflect the user's real repayment pressure.

[0099] S30: Optimize the initial repayment strategy according to the intelligent optimization algorithm and the repayment burden value to obtain the optimal repayment strategy;

[0100] The intelligent optimization algorithm may be a particle swarm optimization algorithm. The initial repayment strategy is optimized according to the particle swarm optimization algorithm and the repayment burden value. Methods for obtaining the optimal repayment strategy include:

[0101] S301: Define the particle's position vector as ,in, Representative Initial repayment strategies form the initial population, and the position and velocity of each particle are randomly initialized. is the total number of initial repayment strategies.

[0102] S302: Define a fitness function and constraints. The fitness function includes:

[0103] ;

[0104] In the formula, is the fitness function, is the repayment burden value;

[0105] The constraints include:

[0106] ;

[0107] In the formula, is the repayment period in the initial repayment strategy, is the maximum standard value of the repayment period, is the repayment amount per period in the initial repayment strategy, The minimum standard value for each repayment amount.

[0108] Among them, this embodiment uses the fitness function to measure whether the interval time of the current particle is appropriate. The smaller the fitness value, the closer the initial repayment strategy is to the optimal. This ensures that the solution in the optimization process not only meets the objective function, but also meets the basic constraints of the actual repayment strategy design, improves the actual availability of the optimization solution, and avoids the generation of unreasonable repayment strategies.

[0109] S303: Update the particle speed according to the particle's current speed, the particle's optimal position and the global optimal position.

[0110] S304: Update the particle position according to the new velocity:

[0111] = ;

[0112] In the formula, For particles exist The updated position at any time. For particles exist The location at the moment, For particles exist The speed of time.

[0113] S305: Compare the current fitness value of each particle with its historical best fitness value. If the current fitness value is better, update the optimal position of the particle. , compare the current fitness values ​​corresponding to all particles, and select the optimal position of the particle corresponding to the best current fitness value As the global optimal position .

[0114] S306: Repeat S303-S305 above until the preset number of iterations is reached, and the final global optimal position Represents the optimal repayment strategy.

[0115] In this embodiment, the initial repayment strategy provides a specific reference starting point for the optimization process. The optimization process dynamically adjusts the strategy variables (such as repayment amount and repayment period) to reduce the repayment pressure of users while meeting their personalized needs. The optimal repayment strategy finally generated can take into account the actual situation and repayment willingness of users, further improving the adaptability and effectiveness of the strategy.

[0116] Methods for updating particle velocity include:

[0117] ;

[0118] In the formula, For particles exist The speed of time, For particles exist The speed of time, is the inertia weight, is the particle learning factor, is the social learning factor, , Both A random number between is the current global optimal position of the entire particle swarm, For particles The current optimal position of the particle.

[0119] This embodiment first determines the repayment capacity value based on real-time financial information and historical repayment information, determines the repayment willingness value based on the repayment willingness information, and then determines the initial repayment strategy based on the repayment capacity value, the repayment willingness value and the preset standard repayment strategy, determines the user burden information based on the real-time financial information and the initial repayment strategy, calculates the repayment burden value based on the user burden information, and finally optimizes the initial repayment strategy according to the intelligent optimization algorithm and the repayment burden value to obtain the optimal repayment strategy. In this way, this embodiment can dynamically adjust the repayment strategy by comprehensively analyzing the real-time financial information, historical repayment information and repayment willingness information in combination with the intelligent optimization algorithm, thereby solving the problem of lack of dynamic adjustment and personalized design in the prior art. The optimal repayment strategy finally generated is dynamic and flexible, which improves the user experience and enhances the service capabilities and user trust of the lending institution.

[0120] Example 2

[0121] See also Figure 2 As shown, based on the same inventive concept, this embodiment discloses an intelligent repayment strategy generation system driven by multi-source information analysis. For details not provided in this embodiment, please refer to the description of the relevant parts in Embodiment 1. The system includes:

[0122] Data analysis module: used to obtain the target user's real-time financial information, historical repayment information and repayment willingness information, determine the repayment ability value based on the real-time financial information and historical repayment information, and determine the repayment willingness value based on the repayment willingness information;

[0123] In this embodiment, real-time financial information refers to data that can dynamically reflect the user's current economic status. Real-time financial information includes but is not limited to income amount, expenditure amount, debt amount and credit limit value. It can be understood that the income information, expenditure information, etc. mentioned above can be the income and expenditure of the current month, or the average income or average expenditure of the previous few months. This embodiment does not limit this.

[0124] It should be added that historical repayment information refers to data that reflects the user's previous repayment behavior and repayment records. Historical repayment information includes but is not limited to historical repayment amounts, historical overdue times, and historical repayment frequencies. The historical repayment amount mentioned above can be the average repayment amount of the previous few years. Similarly, the historical overdue times can be the average overdue times of the previous few years. This embodiment does not go into too much detail on this.

[0125] In an optional embodiment, the repayment intention information refers to data reflecting the user's subjective repayment intention and repayment attitude. The method for obtaining the repayment intention information includes:

[0126] Obtain user communication data and user behavior data, perform semantic analysis on the user communication data to obtain the repayment keyword frequency, use each control in the user behavior data as a node, and the relationship between controls as an edge, build a behavior logic graph, and use the repayment keyword frequency and behavior logic graph as repayment intention information.

[0127] Methods for determining repayment capacity values ​​based on real-time financial information and historical repayment information include:

[0128] Calculate the difference between the income amount and the expenditure amount to get the disposable amount, take the ratio of the disposable amount to the income amount as the disposable income ratio, calculate the sum of the credit limit value and the historical repayment amount as the repayment potential value, take the ratio of the debt amount to the repayment potential value as the debt pressure ratio, take the difference between the historical repayment frequency and the historical overdue times as the historical fulfillment rate value, take the weighted sum of the disposable income ratio, debt pressure ratio and historical fulfillment rate value to get the repayment ability value.

[0129] Methods for obtaining the repayment capacity value by weighted summing the disposable income ratio, debt pressure ratio and historical performance rate include:

[0130] RAV=u1DIR-u2DPR+u3HCR;

[0131] In the formula, RAV is the repayment capacity value, DIR is the disposable income ratio, DPR is the debt pressure ratio, and HCR is the historical performance rate value. , , are all weight factors greater than 0.

[0132] Methods for determining the repayment willingness value based on the repayment willingness information include:

[0133] The positive and negative values ​​are calculated according to the repayment keyword frequency, and the positive and negative values ​​are weightedly summed to obtain the keyword comprehensive value. The sum of the edge weights in the behavior logic graph is calculated according to the graph algorithm, and the keyword comprehensive value and the edge weight sum are cascaded and integrated to obtain the repayment willingness value.

[0134] Strategy generation module: used to determine the initial repayment strategy based on the repayment ability value, the repayment willingness value and the preset standard repayment strategy, determine the user burden information based on the real-time financial information and the initial repayment strategy, and calculate the repayment burden value based on the user burden information;

[0135] In this embodiment, the standard repayment strategy refers to a basic repayment plan or scheme provided to users based on industry norms or rules preset by lending institutions. The standard repayment strategy may include repayment method, repayment period, repayment frequency, repayment amount, interest rate standard, etc.

[0136] Methods for determining the initial repayment strategy based on the repayment ability value, the repayment willingness value, and the preset standard repayment strategy include:

[0137] The repayment ability value, repayment willingness value, and preset standard repayment strategy are input into the pre-built strategy adjustment model to obtain the initial repayment strategy.

[0138] The construction method of the strategy adjustment model includes:

[0139] Acquire a sample data set, wherein the sample data set includes a historical repayment ability value, a historical repayment willingness value, a historical standard repayment strategy, and a historical initial repayment strategy;

[0140] Divide the sample data set into a sample training set and a sample test set, and build a regression network;

[0141] The historical repayment ability value, historical repayment willingness value and historical standard repayment strategy in the sample training set are used as the input data of the regression network, and the historical initial repayment strategy in the sample training set is used as the output data of the regression network. The regression network is trained to obtain an initial regression network for predicting the initial repayment strategy.

[0142] The initial regression network is tested using a sample test set, and the initial regression network with an output smaller than a preset error value is used as a strategy adjustment model. The initial regression network is a deep neural network model.

[0143] Methods for determining user burden information based on real-time financial information and initial repayment strategies include:

[0144] Determine the repayment amount and interest amount for each period in the initial repayment strategy, take the ratio of each repayment amount to the disposable amount as the repayment burden ratio, take the ratio of each interest amount to the disposable amount as the interest burden ratio, take the ratio of each repayment amount to the income amount as the monthly payment-to-income ratio, and determine the repayment burden ratio, interest burden ratio and monthly payment-to-income ratio as user burden information.

[0145] The method for calculating the repayment burden value according to the user burden information includes:

[0146] RBV= ;

[0147] In the formula, RBV is the repayment burden value, IBR is the interest burden ratio, REB is the repayment burden ratio, and MPR is the monthly payment-to-income ratio. is the inverse cotangent function, is the hyperbolic cosine function, is a logarithmic function with base e.

[0148] Strategy optimization module: used to optimize the initial repayment strategy based on the intelligent optimization algorithm and the repayment burden value to obtain the optimal repayment strategy.

[0149] Example 3

[0150] This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the intelligent repayment strategy generation method driven by multi-source information analysis provided by the above-mentioned methods is implemented.

[0151] Since the electronic device introduced in this embodiment is an electronic device used to implement the intelligent repayment strategy generation method driven by multi-source information analysis in the embodiment of the present application, based on the intelligent repayment strategy generation method driven by multi-source information analysis introduced in the embodiment of the present application, technical personnel in this field can understand the specific implementation method of the electronic device of this embodiment and its various variations, so how the electronic device implements the method in the embodiment of the present application is no longer introduced in detail. As long as technical personnel in this field implement the electronic device used by the intelligent repayment strategy generation method driven by multi-source information analysis in the embodiment of the present application, it belongs to the scope of protection of this application.

[0152] Example 4

[0153] The present embodiment discloses a computer-readable storage medium having a computer program stored thereon. When the computer program is executed, the intelligent repayment strategy generation method driven by multi-source information analysis is implemented.

[0154] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters, weights and thresholds in the formula are set by technicians in this field according to actual conditions.

[0155] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired network or a wireless network. The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media sets. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD) or a semiconductor medium. The semiconductor medium may be a solid-state hard disk.

[0156] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0157] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0158] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only one, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0159] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0160] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0161] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

[0162] 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 in the protection scope of the present invention.

Claims

1. An intelligent repayment strategy generation method driven by multi-source information analysis, characterized in that: include: Obtain the target user's real-time financial information, historical repayment information, and repayment willingness information, determine the repayment ability value based on the real-time financial information and historical repayment information, and determine the repayment willingness value based on the repayment willingness information; Determine an initial repayment strategy based on the repayment ability value, the repayment willingness value, and the preset standard repayment strategy, determine user burden information based on real-time financial information and the initial repayment strategy, and calculate the repayment burden value based on the user burden information; Optimize the initial repayment strategy based on the intelligent optimization algorithm and the repayment burden value to obtain the optimal repayment strategy; The method for obtaining the repayment willingness information includes: Obtain user communication data and user behavior data, perform semantic analysis on the user communication data to obtain the repayment keyword frequency, use each control in the user behavior data as a node, and the relationship between controls as an edge, build a behavior logic graph, and use the repayment keyword frequency and behavior logic graph as repayment willingness information; The method for determining the repayment willingness value according to the repayment willingness information includes: Calculate the positive and negative values ​​according to the repayment keyword frequency, perform weighted summation of the positive and negative values ​​to obtain the keyword comprehensive value, calculate the sum of edge weights in the behavior logic graph according to the graph algorithm, cascade the keyword comprehensive value and the sum of edge weights to obtain the repayment willingness value; The method for determining the initial repayment strategy based on the repayment ability value, the repayment willingness value and the preset standard repayment strategy includes: Input the repayment ability value, repayment willingness value and preset standard repayment strategy into the pre-built strategy adjustment model to obtain the initial repayment strategy; The method for determining user burden information based on real-time financial information and initial repayment strategy includes: Determine the repayment amount and interest amount for each period in the initial repayment strategy, take the ratio of each repayment amount to the disposable amount as the repayment burden ratio, take the ratio of each interest amount to the disposable amount as the interest burden ratio, take the ratio of each repayment amount to the income amount as the monthly payment-to-income ratio, and determine the repayment burden ratio, interest burden ratio and monthly payment-to-income ratio as user burden information.

2. The intelligent repayment strategy generation method driven by multi-source information analysis according to claim 1 is characterized in that: The real-time financial information includes income amount, expenditure amount, debt amount and credit limit value, and the historical repayment information includes historical repayment amount, historical overdue number and historical repayment frequency. The method for determining the repayment capacity value based on the real-time financial information and historical repayment information includes: Calculate the difference between the income amount and the expenditure amount to get the disposable amount, take the ratio of the disposable amount to the income amount as the disposable income ratio, calculate the sum of the credit limit value and the historical repayment amount as the repayment potential value, take the ratio of the debt amount to the repayment potential value as the debt pressure ratio, take the difference between the historical repayment frequency and the historical overdue times as the historical fulfillment rate value, take the weighted sum of the disposable income ratio, debt pressure ratio and historical fulfillment rate value to get the repayment ability value.

3. The intelligent repayment strategy generation method driven by multi-source information analysis according to claim 2 is characterized in that: The construction method of the policy adjustment model includes: Acquire a sample data set, wherein the sample data set includes a historical repayment ability value, a historical repayment willingness value, a historical standard repayment strategy, and a historical initial repayment strategy; Divide the sample data set into a sample training set and a sample test set, and build a regression network; The historical repayment ability value, historical repayment willingness value and historical standard repayment strategy in the sample training set are used as the input data of the regression network, and the historical initial repayment strategy in the sample training set is used as the output data of the regression network. The regression network is trained to obtain an initial regression network for predicting the initial repayment strategy. The initial regression network is tested using the sample test set, and the initial regression network with an output smaller than the preset error value is used as the strategy adjustment model.

4. The intelligent repayment strategy generation method driven by multi-source information analysis according to claim 3 is characterized in that: The method for calculating the repayment burden value according to the user burden information includes: RBV= ; In the formula, RBV is the repayment burden value, IBR is the interest burden ratio, REB is the repayment burden ratio, and MPR is the monthly payment-to-income ratio. is the inverse cotangent function, is the hyperbolic cosine function, is a logarithmic function with base e.

5. The intelligent repayment strategy generation method driven by multi-source information analysis according to claim 4 is characterized in that: The intelligent optimization algorithm is a particle swarm optimization algorithm. The initial repayment strategy is optimized according to the particle swarm optimization algorithm and the repayment burden value. The method for obtaining the optimal repayment strategy includes: S301: Define the particle's position vector as ,in, Representative Initial repayment strategies form the initial population, and the position and velocity of each particle are randomly initialized. is the total number of initial repayment strategies; S302: Define a fitness function and constraints. The fitness function includes: ； In the formula, is the fitness function, is the repayment burden value; The constraints include: ； In the formula, is the repayment period in the initial repayment strategy, is the maximum standard value of the repayment period, is the repayment amount per period in the initial repayment strategy, The minimum standard value for the repayment amount per period; S303: updating the particle speed according to the particle's current speed, the particle's optimal position and the global optimal position; S304: Update the particle position according to the new velocity: ； In the formula, For particles exist The updated position at any time. For particles exist The location at the moment, For particles exist The speed of the moment; S305: Compare the current fitness value of each particle with its historical best fitness value. If the current fitness value is better, update the optimal position of the particle. , compare the current fitness values ​​corresponding to all particles, and select the optimal position of the particle corresponding to the best current fitness value As the global optimal position ; S306: Repeat S303-S305 above until the preset number of iterations is reached, and the final global optimal position Represents the optimal repayment strategy.

6. The intelligent repayment strategy generation method driven by multi-source information analysis according to claim 5 is characterized in that: The method for updating the particle speed includes: ； In the formula, For particles exist The speed of time, For particles exist The speed of time, is the inertia weight, is the particle learning factor, is the social learning factor, , Both A random number between is the current global optimal position of the entire particle swarm, For particles The current optimal position of the particle.

7. An intelligent repayment strategy generation system driven by multi-source information analysis, which is used to implement the intelligent repayment strategy generation method driven by multi-source information analysis as described in any one of claims 1 to 6, characterized in that: include: Data analysis module: used to obtain the target user's real-time financial information, historical repayment information and repayment willingness information, determine the repayment ability value based on the real-time financial information and historical repayment information, and determine the repayment willingness value based on the repayment willingness information; Strategy generation module: used to determine the initial repayment strategy based on the repayment ability value, the repayment willingness value and the preset standard repayment strategy, determine the user burden information based on the real-time financial information and the initial repayment strategy, and calculate the repayment burden value based on the user burden information; Strategy optimization module: used to optimize the initial repayment strategy based on the intelligent optimization algorithm and the repayment burden value to obtain the optimal repayment strategy.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the intelligent repayment strategy generation method driven by multi-source information analysis as described in any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed, the intelligent repayment strategy generation method driven by multi-source information analysis as described in any one of claims 1 to 6 is implemented.

Images