Rental contract generation method and device, equipment and storage medium

Abstract

The invention provides a lease contract generation method and device, equipment and a storage medium, and can be applied to the technical field of data processing. The method comprises the steps that in response to a contract generation request of a target interval, a sample contract of a sample interval is acquired, and the sample interval is a year-on-year interval of the target interval; on the basis of the lease data, the lessee features and the subject matter features of the sample contract, obtaining a continuation prediction value of the sample contract; based on the continuation prediction value, the tenant credit, the contract amount of the sample contract and the contract fulfillment rate, carrying out value scoring on the sample contract, and obtaining a target sample contract based on a scoring result; and based on at least one of the scoring result, the tenant features and the subject matter features, correcting contract elements in the target sample contract to obtain a lease contract.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a lease contract generation method, device, equipment, medium and program product. BACKGROUND

[0002] In the current lease business scenario, with the increase of the types of lease target items and the diversification of tenant needs, the prediction demand of lease enterprises for future renewal contract related financial data is increasingly urgent. Accurate prediction results can help enterprises plan cash flow in advance, optimize asset allocation, and reduce operational risks caused by fluctuations in renewal business. However, existing technical means have significant limitations in meeting this demand, mainly facing two core problems:

[0003] 1. Low efficiency of sample contract data processing. In the lease business, predicting future renewal data needs to be based on a large number of historical and in-transit sample contracts. The number of such sample contracts usually reaches hundreds to thousands, and each contract needs to extract lease end date, renewal period, payment plan and other multi-dimensional information to determine whether it meets the renewal simulation conditions. Currently, the industry still relies on manual screening of sample contracts and sorting of key information, which not only consumes time and effort, but also easily leads to sample omission or information deviation due to human operation errors, seriously affecting the accuracy of subsequent prediction data.

[0004] 2. Rigidity of element generation rules, lack of scene adaptability. When predicting assets, liabilities and fees of renewal contracts, core elements of simulated renewal contracts need to be generated based on discount rates, rent increases, lease period calculation rules, etc. Existing technologies only support single fixed rules to generate simulated elements, for example, only providing a rigid mode to select a discount rate, such as "fixed year and month" or "original contract", which cannot dynamically match sample contract values, thus failing to adapt to the diversified scenarios of lease business, resulting in a disconnection between elements and actual renewal needs, and making it difficult for prediction results to meet the business flexibility demand. SUMMARY

[0005] In view of the above problems, the present application provides a lease contract generation method, device, equipment, medium and program product.

[0006] According to a first aspect of the present application, a lease contract generation method is provided, comprising: in response to a contract generation request of a target interval, obtaining a sample contract of a sample interval, the sample interval being a same period interval of the target interval; obtaining a renewal prediction value of the sample contract based on lease data, tenant characteristics and target characteristics of the sample contract; based on the renewal prediction value, tenant credit, contract amount and contract performance rate of the sample contract, performing value scoring on the sample contract, and obtaining a target sample contract based on the scoring result; based on at least one of the scoring result, tenant characteristics and target characteristics, correcting contract elements in the target sample contract to obtain a lease contract.

[0007] According to an embodiment of the present application, the renewal prediction value of the sample contract is obtained, comprising: forming a graph structure data of the sample contract based on the lease data, the tenant features and the subject features, and the association relationship between the plurality of sample contracts; obtaining the renewal prediction value based on the graph structure data by using a graph neural network model, the graph neural network model being trained based on the graph structure data of the historical contract.

[0008] According to an embodiment of the present application, the graph structure data of the sample contract is formed, comprising: obtaining a node based on each sample contract; converting the lease data, the tenant features and the subject features into feature vectors based on a preset quantification rule to obtain node features; constructing a node edge based on the association relationship between the plurality of sample contracts; and forming the graph structure data of the sample contract based on the node features and the node edge.

[0009] According to an embodiment of the present application, the target sample contract is obtained, comprising: assigning corresponding weights to the renewal prediction value, the tenant credit, the contract amount and the contract performance rate based on the lease business target, and calculating a value score of the sample contract, the lease business target comprising at least one of risk control, income improvement, renewal rate improvement and business balance; and outputting the sample contract with the highest value score to obtain the target sample contract.

[0010] According to an embodiment of the present application, the quantification rule is formulated, comprising: formulating a lease data quantification rule based on the lease data, the lease data comprising at least one of an institution code, a contract category, a historical renewal record, a contract amount, a lease period and a contract performance rate; formulating a tenant feature quantification rule based on the tenant features, the tenant features comprising at least one of a tenant code, a tenant industry, a cooperation period, a tenant credit and a historical renewal preference; and formulating a subject feature quantification rule based on the subject features, the subject features comprising at least one of a subject code, a subject type, an aging degree, a historical maintenance cost, a market price and a market liquidity.

[0011] According to an embodiment of the present application, the contract elements in the target sample contract are revised, comprising: revising the lease period based on the score result, the cooperation period and the aging degree; revising the rent amount based on the score result, the tenant credit and the market liquidity; revising the discount rate based on the score result, the tenant credit and the historical maintenance cost; and revising the payment plan based on the cooperation period, the contract amount and the tenant industry.

[0012] According to an embodiment of the present application, after obtaining the lease contract, the lease liability prediction data is obtained based on the rent amount, the lease period and the discount rate of the lease contract; and the lease interest prediction data and the lease depreciation prediction data are obtained based on the renewal prediction value and the lease liability prediction data.

[0013] According to an embodiment of the present application, the graph neural network model is trained in the following manner: based on the lease data, the tenant features and the subject features, and the association relationship between the plurality of historical contracts, a graph structure data of the historical contracts is formed; based on the renewal record of the historical contracts, a renewal label is labeled for each historical contract; and based on the historical contract graph structure data and the renewal label, the graph neural network model is trained.

[0014] A second aspect of the present application provides a lease contract generation apparatus, comprising: a sample module configured to, in response to a contract generation request of a target interval, acquire a sample contract of a sample interval, the sample interval being a same period interval of the target interval; a prediction module configured to acquire a renewal prediction value of the sample contract based on lease data, tenant features and subject features of the sample contract; a scoring module configured to score a value of the sample contract based on the renewal prediction value, tenant credit, contract amount and contract performance rate of the sample contract, and obtain a target sample contract based on a scoring result; and a correction module configured to correct a contract element in the target sample contract based on at least one of the scoring result, the tenant features and the subject features, to obtain the lease contract.

[0015] A third aspect of the present application provides an electronic device, comprising: one or more processors; a memory configured to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors perform the above method.

[0016] A fourth aspect of the present application further provides a computer-readable storage medium having stored executable instructions, which, when executed by a processor, cause the processor to perform the above method.

[0017] A fifth aspect of the present application further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the above method.

[0018] According to the lease contract generation method, apparatus, device, medium and program product provided by the present application, by taking the same period interval of the target interval as the sample interval, the redundant link of constructing a data pool from zero is avoided, and then the renewal prediction value is acquired based on the lease data, tenant features and subject features of the sample contract, and the target sample contract is screened out in combination with the multi-dimensional index of tenant credit, contract amount and performance rate, which greatly shortens the whole process time from data acquisition to target sample contract determination, at least partially solves the low efficiency problem of manually processing sample contracts one by one; in addition, based on the target sample contract, the contract elements are dynamically corrected based on the scoring result, tenant features and subject features, which can adapt to diversified scenarios of lease business, so as to obtain a lease contract that meets the actual business needs. BRIEF DESCRIPTION OF DRAWINGS

[0019] The above and other objects, features and advantages of the present application will become more apparent from the following description when taken in conjunction with the accompanying drawings, in which:

[0020] Figure 1 An application scenario diagram of a lease contract generation method, device, equipment, medium and program product according to embodiments of the present application is schematically shown;

[0021] Figure 2 A flowchart of a lease contract generation method according to embodiments of the present application is schematically shown;

[0022] Figure 3 A flowchart of obtaining a renewal prediction value of a sample contract according to embodiments of the present application is schematically shown;

[0023] Figure 4 A flowchart of obtaining a target sample contract according to embodiments of the present application is schematically shown;

[0024] Figure 5 A flowchart of correcting a contract element in a target sample contract according to embodiments of the present application is schematically shown;

[0025] Figure 6 A diagram of graph structure data of a sample contract according to embodiments of the present application is schematically shown;

[0026] Figure 7 A structural block diagram of a lease contract generation device according to embodiments of the present application is schematically shown; and

[0027] Figure 8 A block diagram of an electronic device suitable for implementing a lease contract generation method according to embodiments of the present application is schematically shown. DETAILED DESCRIPTION

[0028] Hereinafter, embodiments of the present application will be described with reference to the accompanying drawings. It is to be understood, however, that the description is merely exemplary of the present application, and is not intended to limit the scope of the present application. In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the present application. However, it will be apparent to one skilled in the art that one or more embodiments of the present application can be practiced without these specific details. In other instances, well-known structures and techniques have not been shown in detail in order not to obscure aspects of the present application.

[0029] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the present application. As used herein, the term "includes" and / or "including" means encompassing but not limited to.

[0030] All terms used herein, including technical and scientific terms, have the meanings commonly understood by one of ordinary skill in the art unless otherwise defined. It should be noted that the use of any terms herein should be interpreted as consistent with the context of the specification, and should not be interpreted in an idealized or overly formal manner.

[0031] In the case of using expressions similar to "at least one of A, B, and C, etc.", in general, it should be interpreted as having the meaning commonly understood by one of ordinary skill in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, a system having B alone, a system having C alone, a system having both A and B, a system having both A and C, a system having both B and C, and / or a system having A, B, and C, etc.).

[0032] It should be noted that in the embodiments of the present application, some software, components, models, etc. may be mentioned, which should be considered as exemplary, and the purpose is only to illustrate the feasibility of the implementation of the technical solutions of the present application, but it does not mean that the applicant has or will necessarily use the scheme.

[0033] In the technical solutions of the present application, the user information (including but not limited to user personal information, user image information, user equipment information, such as location information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved are information and data authorized by the user or authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of related data comply with relevant laws, regulations and standards, necessary measures are taken, do not violate public order and good customs, and provide corresponding operation portal for user to choose authorization or refusal.

[0034] In the scenario of using personal information for automated decision-making, the method, device and system provided by the embodiments of the present application all provide corresponding operation portal for the user to choose to agree or refuse the automated decision-making result; if the user chooses to refuse, the expert decision-making process is entered. The expression "automated decision-making" here refers to the activity of automatically analyzing, evaluating the behavior habits, interests and hobbies, or economic, health, credit status, etc. of an individual through a computer program, and making decisions. The expression "expert decision-making" here refers to the activity of making decisions by personnel who are engaged in a certain field of work, have special experience, knowledge and skills, and reach a certain professional level.

[0035] An embodiment of this application provides a method for generating a lease contract, comprising: responding to a contract generation request for a target range, obtaining a sample contract for a sample range, wherein the sample range is a year-on-year range of the target range; obtaining a renewal prediction value for the sample contract based on the lease data, tenant characteristics, and property characteristics of the sample contract; assigning a value score to the sample contract based on the renewal prediction value, as well as tenant credit, contract amount, and contract performance rate of the sample contract, and obtaining a target sample contract based on the score result; and modifying the contract elements in the target sample contract based on at least one of the score result, tenant characteristics, and property characteristics to obtain a lease contract.

[0036] Figure 1 The illustration shows an application scenario diagram of the lease contract generation method according to an embodiment of this application.

[0037] like Figure 1 As shown, application scenario 100 according to this embodiment may include terminal devices 101, 102, and 103, network 104, and server 105. Network 104 is used as a medium to provide a communication link between terminal devices 101, 102, and 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links or fiber optic cables, etc.

[0038] Users can use terminal devices 101, 102, and 103 to interact with server 105 via network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, and 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social media platform software, etc. (for example only).

[0039] Terminal devices 101, 102, and 103 can be various electronic devices with displays and web browsing capabilities, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0040] Server 105 can be a server that provides various services, such as a backend management server that supports websites browsed by users using terminal devices 101, 102, and 103 (for example only). The backend management server can analyze and process data such as received user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal devices.

[0041] It should be noted that the lease contract generation method provided in the embodiments of the present application can be generally executed by the server 105. Correspondingly, the lease contract generation apparatus provided in the embodiments of the present application can be generally arranged in the server 105. The lease contract generation method provided in the embodiments of the present application can also be executed by a server or a server cluster different from the server 105 and capable of communicating with the terminal devices 101, 102, 103 and / or the server 105. Correspondingly, the lease contract generation apparatus provided in the embodiments of the present application can also be arranged in a server or a server cluster different from the server 105 and capable of communicating with the terminal devices 101, 102, 103 and / or the server 105.

[0042] It should be understood that Figure 1 The number of terminal devices, networks and servers in the above-mentioned scenario is only illustrative. Any number of terminal devices, networks and servers can be provided according to the implementation needs.

[0043] The lease contract generation method of the disclosed embodiments will be described in detail below based on the scenario described above. Figure 1 Figures 2-6 The lease contract generation method of the disclosed embodiments will be described in detail below based on the scenario described above.

[0044] Figure 2 A flowchart of the lease contract generation method according to the embodiments of the present application is schematically shown.

[0045] As shown in Figure 2 The lease contract generation method of this embodiment includes operations S210-S240.

[0046] At operation S210, in response to a contract generation request of a target interval, a sample contract of a sample interval is obtained, and the sample interval is a same period interval of the target interval.

[0047] Firstly, the core parameters of the target interval contract generation request are analyzed, and the key limiting information of the target interval is extracted from the request, including: clearly defining the specific time range of the target interval, for example, “2025-01-01-2025-03-31”, i.e. the first quarter of 2025; determining the business type of the target interval contract, for example, “office equipment rental”, “production machinery rental”, etc.

[0048] Then, the same period interval of the sample interval and the business matching rule are locked, including: taking the same period interval of the target interval as the time reference of the sample interval, ensuring the consistency of the time period length, for example, the target interval is the first quarter of 2025, and the sample interval is the first quarter of 2024; requiring the sample contract of the sample interval to be completely consistent with the business type of the target interval contract, for example, the business type of the target interval contract is “office equipment rental”, and the sample contract needs to filter the office equipment rental contract in the first quarter of 2024 to avoid the adaptability problem caused by cross-business samples.​

[0049] Then, based on the above rules, a sample contract screening operation is performed in the contract management database, including: calling contracts with contract signing dates or effective dates within the sample interval; filtering out contracts with business types matching the target interval; excluding invalid contracts (such as contracts that are not effective, have been terminated, or have major disputes in clauses), and only keeping valid contracts that have been normally executed or have been fulfilled to the expiration date, to ensure that the sample data has complete lease period records (including key information such as fulfillment and whether to renew).

[0050] Finally, field verification is performed on the screened sample contracts to ensure that the core information is not missing, including: checking whether each sample contract contains information required for subsequent processes, such as lease data, tenant characteristics, subject characteristics, and fulfillment records, and contracts missing key fields are excluded.

[0051] In operation S220, based on the lease data, tenant characteristics, and subject characteristics of the sample contract, a renewal prediction value of the sample contract is obtained.

[0052] According to an embodiment of the present application, for each sample contract, lease data, tenant characteristics, and subject characteristics are extracted and quantitatively processed to form node features; then each sample contract is taken as a node of a graph, and node edges are constructed according to the association between different sample contracts to form sample contract graph structure data; finally, the graph structure data is input into a trained graph neural network model (such as graph sampling and aggregation, graph attention network, etc.), which aggregates node features and associated node features, propagates global association information, calculates and outputs a renewal prediction value of each sample contract. The renewal prediction value can comprehensively reflect the influence of single contract features and contract association on renewal decision.

[0053] In operation S230, based on the renewal prediction value, tenant credit, contract amount, and contract fulfillment rate of the sample contract, a value score of the sample contract is calculated, and a target sample contract is obtained based on the score result.

[0054] According to an embodiment of the present application, the renewal prediction value, tenant credit, contract amount, and contract fulfillment rate are standardized to unify the score scale; based on the lease business target, such as risk control, income improvement, renewal rate improvement, and business balance, the above four dimensions are given corresponding weight values, and the value score of each sample contract is calculated by weighting; finally, the sample contracts are sorted in descending order of value score, and the sample contract with the highest value score is selected as the target sample contract.

[0055] In operation S240, based on at least one of the score result, tenant characteristics, and subject characteristics, contract elements in the target sample contract are corrected to obtain a lease contract.

[0056] According to an embodiment of the present application, the contract elements include a lease period, a rent amount, a discount rate, a payment plan, a contract element correction rule is formulated according to the scoring result, tenant characteristics, and target characteristics, for example, for a target sample contract with a high value score, the lease contract can be extended by 10%-20% of the lease period; for a tenant with good credit, a rent discount of 3%-5% can be enjoyed; for a lease contract with a high aging degree of the target, the lease period can be shortened by 15%-20%; for a tenant with a high compliance rate, the payment period can be relaxed; and finally, the lease contract is formed by integrating the correction results of each element. It should be noted that the above correction rules are only examples, and specific correction rules can be formulated according to actual conditions.

[0057] By taking the same period of the target interval as the sample interval, the redundant link of constructing the data pool from zero is avoided, and the renewal prediction value is obtained based on the lease data, tenant characteristics, and target characteristics of the sample contract, and the target sample contract is selected in combination with the tenant credit, contract amount, and compliance rate multi-dimensional index, which greatly shortens the whole process time from data acquisition to target sample contract determination, avoids the low efficiency problem of manually processing sample contracts one by one; in addition, based on the target sample contract, the contract elements are dynamically corrected based on the scoring result, tenant characteristics, and target characteristics, which can adapt to the diversified scenarios of lease business, so as to obtain a lease contract that meets the actual business needs.

[0058] According to an embodiment of the present application, after obtaining the lease contract, the lease liability prediction data is obtained based on the rent amount, lease period, and discount rate of the lease contract. Specifically, the fixed rent amount of each period in the lease contract and the total number of lease periods are determined, and the interest rate for discount, i.e., the discount rate, is determined; on this basis, the initial measurement amount of the lease liability is calculated by multiplying the future rent amount of each period by the present value coefficient of the corresponding period, i.e., the sum of the present values of the future rent payable; then, the lease period is predicted period by period, and finally, the complete lease liability prediction data including the initial amount of the lease liability, the interest expense of each period, the principal repayment amount of each period, and the balance at the end of each period is formed.

[0059] For example, a certain enterprise leases an office equipment, and the contract elements of the lease contract signed are as follows: an equal amount of rent of 100,000 yuan is paid at the end of each year (a total of 3 periods, non-prepaid); the total number of lease periods is 3 periods; the discount rate is 5%. Based on the above parameters, the calculation process of the lease liability prediction data is as follows:

[0060] First, the initial measurement amount of the lease liability (the total present value of future rent) is calculated. Since the rent is paid at the end of each year, the present value is calculated using the ordinary annuity present value factor, which can be obtained from the ordinary annuity present value factor table. The ordinary annuity present value factor for a 3-year term and a discount rate of 5% is 2.7232. The initial measurement amount of the lease liability is calculated as follows: initial measurement amount of the lease liability = annual rent amount x ordinary annuity present value factor = 100,000 yuan x 2.7232 = 273,200 yuan (i.e., the initial balance of the lease liability to be recognized on the start date of the lease term).

[0061] Next, the interest expense, principal repayment amount, and balance at the end of each period of the lease liability are predicted. The interest expense for the first period is calculated as follows: interest expense for the first period = initial balance of the lease liability (initial measurement amount) x discount rate = 273,200 yuan x 5% = 13,660 yuan. The principal repayment amount for the first period is calculated as follows: principal repayment amount for the first period = annual rent amount - interest expense for the first period = 100,000 yuan - 13,660 yuan = 86,340 yuan. The balance of the lease liability at the end of the first period is calculated as follows: balance of the lease liability at the end of the first period = initial balance - principal repayment amount for the first period = 273,200 yuan - 86,340 yuan = 186,860 yuan. The interest expense for the second period is calculated as follows: interest expense for the second period = balance of the lease liability at the end of the first period x discount rate = 186,860 yuan x 5% ≈ 9,340 yuan. The principal repayment amount for the second period is calculated as follows: principal repayment amount for the second period = annual rent amount - interest expense for the second period = 100,000 yuan - 9,340 yuan = 90,660 yuan. The balance of the lease liability at the end of the second period is calculated as follows: balance of the lease liability at the end of the second period = balance of the lease liability at the end of the first period - principal repayment amount for the second period = 186,860 yuan - 90,660 yuan = 96,200 yuan. The interest expense for the third period is calculated as follows: interest expense for the third period = balance of the lease liability at the end of the second period x discount rate = 96,200 yuan x 5% ≈ 4,810 yuan. The principal repayment amount for the third period is calculated as follows: principal repayment amount for the third period = annual rent amount - interest expense for the third period = 100,000 yuan - 4,810 yuan = 95,190 yuan. The balance of the lease liability at the end of the third period is calculated as follows: balance of the lease liability at the end of the third period = balance of the lease liability at the end of the second period - principal repayment amount for the third period = 96,200 yuan - 95,190 yuan = 1,010 yuan. The predicted data of the lease liability is shown in Table 1.

[0062] Table 1: Predicted data of the lease liability

[0063]

[0064] According to an embodiment of the present application, based on the renewal prediction value and the lease liability prediction data, the lease interest prediction data is obtained. If the renewal prediction value is less than the threshold value, i.e. the probability of non-renewal is high, the renewal period does not need to be considered. According to the above example, the lease interest prediction data directly uses the "current period interest expense" in Table 1, and the lease interest prediction data is the sum of the interest expenses of the three periods, i.e. 277,000 yuan. If the renewal prediction value is greater than or equal to the threshold value, i.e. the probability of renewal is high, the initial amount of the lease liability in the renewal period is first calculated. Assuming that the renewal period is 2 years, the annual rent is 105,000 yuan, and the discount rate is 5%, the initial amount of the lease liability in the renewal period is calculated using the ordinary annuity present value coefficient, i.e. the initial amount of the lease liability in the renewal period = annual rent x annuity present value coefficient = 10.5 x 1.8594 ≈ 19.52 million yuan. According to the above example, the lease liability balance at the end of the third period is 0 (the liability has been settled), so the initial lease liability balance at the beginning of the first year of the renewal period is equal to the initial amount of the lease liability in the renewal period, i.e. 19.52 million yuan. The lease interest prediction data in the renewal period is calculated according to the logic of "initial balance x discount rate = current period interest", and the interest of the fourth period is 0.98 million yuan and the interest of the fifth period is 0.5 million yuan. Therefore, the lease interest prediction data is the sum of the interest of the first to fifth periods, i.e. 4.25 million yuan.

[0065] According to an embodiment of the present application, based on the renewal prediction value and the lease liability prediction data, the lease depreciation prediction data is obtained. If the renewal prediction value is less than the threshold value, i.e. the probability of non-renewal is high, the renewal period does not need to be considered. According to the above example, the use right asset is the initial amount of the lease liability, which is 27.23 million yuan, and the depreciation period is 3 years, so the depreciation of each period is 27.23 ÷ 3 ≈ 9.08 million yuan / year (9.08 million yuan is deducted as depreciation every year in 3 years). If the renewal prediction value is greater than or equal to the threshold value, i.e. the probability of renewal is high, according to the above example, the use right asset is the initial amount of the lease liability plus the initial amount of the lease liability in the renewal period, i.e. 27.23 + 19.52 = 46.75 million yuan, and the depreciation period is 5 years, so the depreciation of each period is 46.75 ÷ 5 ≈ 9.35 million yuan / year (9.35 million yuan is deducted as depreciation every year in 5 years).

[0066] Based on the generation of the lease contract, by inputting the rent amount, lease period and discount rate in the contract, the present value of future lease payments and the change of principal and interest in each period can be accurately quantified to form structured lease liability prediction data. Further, by combining the renewal prediction value with the above lease liability prediction data, the lease interest in different scenarios can be dynamically calculated according to the possibility of renewal, and the depreciation period of the use right asset can be determined based on the probability of renewal to generate lease depreciation prediction data synchronously, thereby forming a complete financial prediction chain covering lease liability, lease interest and lease depreciation, and providing accurate and implementable data support for subsequent financial accounting, cost control and cash flow planning of the enterprise's leasing business.

[0067] Figure 3A flowchart of acquiring a renewal prediction value of a sample contract according to an embodiment of the present application is shown.

[0068] As shown in the figure, acquiring a renewal prediction value of a sample contract of the embodiment includes operation S310 to operation S320. Figure 3

[0069] In operation S310, based on the lease data, tenant features and subject features, and the correlation between multiple sample contracts, a graph structure data of the sample contracts is formed.

[0070] First, a node is obtained based on each sample contract. Specifically, each sample contract is taken as an independent node in the graph, and a unique identifier (such as contract code) is assigned to each node to ensure that the node corresponds to the sample contract one by one, for example, “contract 001” corresponds to node 1, “contract 002” corresponds to node 2, forming an initial “node set”.

[0071] Then, based on a preset quantification rule, the lease data, tenant features and subject features are converted into feature vectors to obtain node features. Specifically, according to the preset quantification rule, the numerical conversion is performed on the three types of information of each sample contract, for example, “lease period 2 years” in the lease data is quantified as “24”, “cooperation period 3 years” in the tenant features is quantified as “3”, and “moderate aging” in the subject features is quantified as “3”; then all the quantified numerical values of each sample contract are arranged in a fixed order to form the feature vector of the node, i.e. the node feature, for example, the feature vector of node 1 (contract 001) is [24, 3, 3].

[0072] According to an embodiment of the present application, based on the lease data, a lease data quantification rule is formulated, and the lease data includes at least one of agency code, contract category, historical renewal record, contract amount, lease period, and contract performance rate. Among them, the agency code is assigned a unique integer value in the order of system registration; the contract category is mapped to a fixed number according to the lease nature, for example, operating lease = 1, financial lease = 2, sale-leaseback = 3, and other special lease = 4; the historical renewal record counts the number of renewals of the same tenant by the agency in the past three years, and 0 is recorded if there is no relevant record; the contract amount is quantified in multiple intervals, for example, less than 50,000 = 1, 50,000-100,000 = 2, 100,000-200,000 = 3, 200,000-500,000 = 4, and more than 500,000 = 5; the lease period is uniformly converted into months; and the contract performance rate is calculated according to the actual value.

[0073] ​According to an embodiment of the present application, based on tenant characteristics, tenant characteristic quantification rules are formulated, the tenant characteristics including at least one of tenant code, tenant industry, cooperation time, tenant credit, and historical renewal preference. Among them, the tenant code is assigned a unique continuous integer value in the order of first cooperation registration of the tenant; the tenant industry is mapped to a fixed number, for example, manufacturing industry = 1, service industry = 2, retail industry = 3, financial industry = 4, and other industries = 5; the cooperation time is uniformly converted into years; the tenant credit is mapped to a specific value, for example, AAA level = 5, AA level = 4, A level = 3, BBB level = 2, and lower than BBB level = 1; and the historical renewal preference is graded after calculating the proportion of "the number of contracts actively proposed for renewal" to "the total number of expired contracts" in the last 3 years, for example, greater than or equal to 80% = 4, 50%-80% = 3, 20%-50% = 2, less than 20% = 1, and 0 for no expired contract.

[0074] According to an embodiment of the present application, based on target characteristics, target characteristic quantification rules are formulated, the target characteristics including at least one of target code, target type, aging degree, historical maintenance cost, market price, and market liquidity. Among them, the target code is assigned a unique continuous integer value in the order of first warehouse registration; the target type is mapped to a fixed number according to the purpose, for example, office equipment = 1, production equipment = 2, transportation tool = 3, office furniture = 4, and other types = 5; the aging degree is quantified in multiple intervals, for example, brand new = 1, slightly aged = 2, moderately aged = 3, severely aged = 4, and scrap edge = 5; the historical maintenance cost is divided into multiple intervals according to the cumulative cost in the last year, for example, less than 1000 yuan = 1 and more than 20000 yuan = 5; the market price is divided into multiple grades according to the current brand new price, for example, less than 50,000 yuan = 1 and more than 500,000 yuan = 5; and the market liquidity is divided into multiple grades according to the monthly average transaction volume of the same type in the last 3 months, for example, more than 50 units = 5 and less than 5 units = 1.

[0075] By formulating corresponding quantification rules for lease data, tenant characteristics, and target characteristics respectively, the unified conversion of the core information of the lease business from non-standardized description to structured numerical value is realized, thereby converting into quantified features recognizable by the model, and finally building a full-dimensional data standardization system covering lease transactions, tenant attributes, and target state. This provides standardized, consistent, and high-quality basic data support for the subsequent node feature construction of sample contract graph structure data and the input of graph neural network model training, effectively avoiding the model training deviation caused by chaotic data format, and improving the accuracy and reliability of subsequent renewal prediction and other core tasks.

[0076] Then, node edges are constructed based on the correlation between multiple sample contracts. Specifically, the generation rule of the node edge is set according to the actual correlation between the sample contracts. If the tenants, the types of the subject matter, and the industries of the tenants of two contracts are the same, an edge is established between the corresponding two nodes. For example, node 1 (contract 001, tenant code T01) and node 3 (contract 003, tenant code T01) are contracts of the same tenant, and an edge is established between them to form a node edge set containing the two node codes and the correlation type, such as "node 1-node 3: same tenant".

[0077] Finally, the graph structure data of the sample contract is formed based on the node features and the node edges, such as Figure 6 as shown, for subsequent use by a graph neural network model. The above process realizes the conversion of the leasing business data from unstructured information to structured graph data, which not only retains the core attribute information of a single contract, but also reflects the correlation logic between multiple contracts, providing high-quality and strongly correlated data support for subsequent core tasks such as renewal prediction based on graph neural networks, effectively improving the adaptability and analysis accuracy of the model to the leasing business scenario.

[0078] In operation S320, based on the graph structure data, a renewal prediction value is obtained using a graph neural network model.

[0079] According to an embodiment of the present application, the graph neural network model is trained in the following manner: based on the lease data, tenant features and target features, and the association relationship between a plurality of historical contracts, a graph structure data of the historical contracts is formed; based on the renewal record of the historical contracts, a renewal label is marked for each historical contract; and based on the historical contract graph structure data and the renewal label, the graph neural network model is trained. Specifically, the full data of the historical contracts is collected, including the lease data, tenant features, target features, and the association relationship between the contracts, the above-mentioned unstructured information is converted into feature vectors through a pre-set quantification rule, and at the same time, each historical contract is assigned an independent node, and the node edges are constructed according to the association relationship, forming the historical contract graph structure data containing nodes, feature vectors, and node edges; based on the actual performance of the historical contracts, a label is marked, for example, if the contract is successfully renewed after expiration, the corresponding node is marked as 1 (representing "renewal"), if it is not renewed, the corresponding node is marked as 0 (representing "non-renewal"), and for atypical cases such as early termination, invalid samples are marked and removed, ensuring that the label is consistent with the real business scenario; then, the historical contracts are divided into a training set and a validation set; finally, the graph attention network is used as the core model for training, and the attention mechanism is used to aggregate the features of each node and the features of the neighborhood nodes, for example, a certain contract node focuses on the features of the high-renewal-rate nodes of the same tenant, and finally a stable convergent model is obtained. The renewal label provides a clear training target for the graph neural network model, ensuring that the model learns the renewal rules that fit the actual business, and the graph neural network can aggregate the features of the nodes and the features of the neighborhood nodes to mine deep association rules that cannot be reflected by a single contract data, for example, a certain contract can learn the features of high-renewal-rate contracts of the same tenant, and find that "contracts between high-credit tenants and low-aging targets have a higher renewal rate", and finally the trained model has higher renewal prediction accuracy and business adaptability, providing reliable technical support for subsequent new contract renewal prediction.

[0080] According to an embodiment of the present application, based on the graph structure data of the sample contract, the above-mentioned graph neural network model is used to obtain a renewal prediction value. Specifically, the graph structure data of the sample contract is input into the trained graph neural network model, the model aggregates the features of the nodes and the features of the associated nodes, propagates the global association information, calculates and outputs the renewal prediction value of each sample contract, and the renewal prediction value can comprehensively reflect the influence of single contract features and contract association on the renewal decision. By using the graph neural network model verified by historical data, the learned renewal rules are efficiently reused, and deep reasoning is performed based on the features of the sample contract and the features of the neighborhood nodes. Compared with single data prediction, it is more accurate, and finally provides a quantitative basis for the renewal decision of the sample contract that is consistent with the actual business and has high reliability.

[0081] Figure 4 A flowchart for obtaining a target sample contract according to an embodiment of the present application is schematically shown.

[0082] As shown in the embodiment, the target sample contract is obtained by operation S410~operation S420. Figure 4

[0083] In operation S410, based on the leasing business target, the renewal prediction value, the tenant credit, the contract amount, and the contract performance rate are given corresponding weights, and the value score of the sample contract is calculated.

[0084] According to an embodiment of the present application, the leasing business target includes at least one of risk control, income improvement, renewal rate improvement, and business balance. Among them, the risk control target is to reduce the risk of tenant default, target damage and cash flow disruption; the income improvement target is to realize the two-way improvement of leasing business income growth and cost saving; the renewal rate improvement target is to enhance tenant stickiness to improve the proportion of renewal after the contract expires; and the business balance target is to ensure the stable development of the overall business and avoid the impact of local risks on the overall business.

[0085] According to an embodiment of the present application, based on the above leasing business target, the renewal prediction value, the tenant credit, the contract amount, and the contract performance rate are given corresponding weights, and the weight is strongly bound with the core demand of the business target. For example, when the risk control is the business target, the tenant credit can directly reflect the probability of tenant default, and the high credit tenant can significantly reduce the risk of overdue rent and target damage, which is the core index of risk control and has the highest weight; the contract performance rate reflects the historical performance of the tenant, and the tenant with high performance rate has lower subsequent default risk, which is the key basis for risk prediction and has the second highest weight; the renewal can reduce the potential risk of target vacancy period, such as maintenance cost and asset depreciation caused by idling, but has lower priority than direct risk indicators; the amount is not directly related to the risk, and is only used as a reference, so it has the lowest weight; therefore, the weight assignment is as follows: renewal prediction value 20%, tenant credit 35%, contract amount 15%, and contract performance rate 30%. It can be understood that the above weight values are only examples, and can be adjusted according to actual conditions.

[0086] According to an embodiment of the present application, the sample contract with the given weight is given a value score. Specifically, the renewal prediction value, the tenant credit, the contract amount, and the contract performance rate are standardized, for example, converted into the same order of magnitude value of 0-100 points, and then combined with the index weight under different leasing business targets to calculate the value score of the sample contract by weighted summation.

[0087] In operation S420, the sample contract with the highest value score is output to obtain the target sample contract.

[0088] ​According to the embodiment of the present application, the value score result obtained in operation S410 is arranged in descending order, and the sample contract ranked first after ordering is the target sample contract. By first giving the renewal prediction value, tenant credit, contract amount, and contract performance rate of the leasing business target different weights that fit the target demand, the dispersed business indicators are strongly bound to the core target, and then the value score of the sample contract is calculated based on the weights, realizing the conversion of multi-dimensional indicators to a unified quantitative standard, avoiding the one-sidedness of single indicator judgment; then the sample contract with the highest value score is output to determine the target sample contract, ensuring that the selected contract can accurately match the high-quality attributes under the current business target, filtering out the target sample contract with the most business value, providing a data-driven accurate basis for subsequent adjustment of contract elements to generate a leasing contract, and improving the efficiency of leasing business resource allocation and the achievement rate of the core target.

[0089] Figure 5 A flowchart of modifying contract elements in the target sample contract according to the embodiment of the present application is shown schematically.

[0090] As shown in Figure 5 , the contract elements in the target sample contract of this embodiment include operations S510-S540.

[0091] In operation S510, the lease period is modified based on the score result, the cooperation period, and the aging degree.

[0092] According to the embodiments of the present application, first, the sample lease period of the target sample contract is taken as the benchmark, and then the combination of the score result, the cooperation period, and the aging degree is combined to adjust the lease period through the mode of "sample lease period ± correction month", to ensure that the corrected lease period matches the contract value and balances the tenant stability and the life of the subject. First, the three influencing factors are defined by filing to define the correction direction and amplitude, for example, for high-value (score result 80-100 points), medium-value (score result 60-79 points), and low-value (score result less than 60 points) target sample contracts, the lease period is respectively corrected by ±3~6 months, ±1~3 months, and -3~6 months; for long-term cooperation (cooperation period more than 5 years), medium-term cooperation (cooperation period 1-5 years), and short-term cooperation (less than 1 year), the lease period is respectively corrected by +2~4 months, ±1~2 months, and -2~4 months; for light aging, moderate aging, and severe aging of the subject, the lease period is respectively corrected by +2~3 months, ±1 month, and -2~3 months. Then, according to the principle of factor superposition and reasonable value, the corrected lease period is calculated, for example, if it is a high-value, long-term cooperation, and light aging combination, the target sample contract lease period of 24 months can be corrected to 33 months, locking the long-term income of high-quality tenants and low-loss subjects. After correction, it is necessary to ensure that the lease period is not less than the industry minimum lease and not more than the reasonable remaining life of the subject, and finally a reasonable lease period that fits multiple dimensions and adapts to business goals is obtained.

[0093] In operation S520, the rent amount is corrected based on the score result, tenant credit, and market liquidity.

[0094] According to the embodiments of the present application, first, the rent amount of the target sample contract is taken as the benchmark, and then the combination of the score result, the tenant credit, and the market liquidity is combined to superimpose and correct through the mode of "target sample contract rent amount × (1 + each factor correction ratio)", to ensure that the corrected rent fits the market situation and matches the contract value and tenant attributes. First, the three influencing factors are defined by filing to define the correction ratio, for example, for high-value, medium-value, and low-value target sample contracts, the rent amount is respectively corrected by -5%~-3%, 0%~±2%, and +3%~+5%; for tenants with credit ratings of AAA, A / AA, and BBB and below, the rent amount is respectively corrected by -4%~-2%, 0%~±1%, and 2%~+4%; for high-liquidity, medium-liquidity, and low-liquidity subjects, the rent amount is respectively corrected by +3%~+5%, 0%~±2%, and -5%~-3%.

[0095] According to the principle of proportional superposition and control amplitude, the corrected rent amount is calculated, for example, if it is a high-quality combination of high value, AAA credit, high liquidity, the target sample contract rent amount of 5000 yuan / month can be corrected to 5000x[1+(-4%-3%+4%)]=4850 yuan / month, that is, the high-quality tenants are locked by credit and score concessions, and the market heat is matched by moderately raising the price through liquidity. At the same time, an upper limit of the correction amplitude is set to avoid the rent deviating from the reasonable range of the market; after correction, the rent level of the same type and same condition target is verified to ensure that the result reflects differentiation and conforms to market recognition, and finally a reasonable rent amount that takes into account the yield, risk and market adaptability is obtained, improving the scientificity and executability of the lease clause.

[0096] In operation S530, the discount rate is corrected based on the score result, tenant credit and historical maintenance cost.

[0097] According to the embodiments of the present application, the base discount rate is first taken as a reference, and then combined with the combination of the score result, tenant credit and historical maintenance cost, the adjustment is superimposed by the method of "base discount rate + each factor correction point", to ensure that the corrected discount rate matches the business cost of funds and covers risks and costs, wherein the base discount rate is usually set by reference to enterprise cost of funds, industry average discount rate and market interest rate. The three influencing factors are first defined by grading correction points, for example, for high value, medium value and low value target sample contracts, the discount rate is corrected by-0.75%~-0.5%, ±0.25%, +0.5%~+0.75% respectively; for tenants with credit ratings of AAA, A / AA, BBB and below, the discount rate is corrected by-0.75%~-0.5%, ±0.25%, +0.5%~+0.75% respectively; for target objects with historical maintenance costs less than 1000 yuan, 1000-20000 yuan and more than 20000 yuan, the discount rate is corrected by-0.5%~-0.25%, ±0.1%, +0.25%~+0.5% respectively. The corrected discount rate is obtained by superimposing the correction points of the three factors and the base discount rate, for example, if it is a high-quality combination of high value, AAA credit and low historical maintenance cost, the base discount rate of 5% can be corrected to 3.5%. At the same time, an upper limit of the total correction amplitude is set to prevent the discount rate from deviating from the reasonable range of the industry.

[0098] In operation S540, the payment plan is corrected based on the cooperation period, contract amount and tenant industry.

[0099] According to the embodiments of the present application, the payment period and the down payment ratio are adjusted in combination with the cooperation period, the contract amount, and the combination of the tenant industry based on the payment plan of the target sample contract, so as to ensure that the revised payment plan controls the enterprise risk and is consistent with the actual payment capacity of the tenant. The three factors are first classified and specific adjustment schemes are set, for example, for the tenants with short-term cooperation, medium-term cooperation, and long-term cooperation, the payment period is shortened, maintained, and lengthened respectively, and the down payment ratio is increased by 5% to 10%, unchanged, and decreased by 5% to 8% respectively; for the contracts with low amount (less than 100,000 yuan), medium amount (100,000 to 500,000 yuan), and high amount (more than 500,000 yuan), the down payment ratio is decreased by 3% to 5%, unchanged, and increased by 8% to 12% respectively; for the industries with high cash flow but large fluctuations, such as retail industry, the down payment ratio can be decreased by 3%, for the industries with long and stable cash flow, such as manufacturing industry, the down payment ratio is unchanged, and for the industries with stable and sufficient cash flow, such as financial industry, the down payment ratio can be decreased by 5% to 8%. The revised payment plan is obtained by superimposing the revision requirements of the three factors, for example, for the combination of long-term cooperation, low amount, and financial industry, the payment period can be extended from monthly payment to quarterly payment, and the down payment ratio can be revised from 10% to 0%, so as to maximize the stability and cash flow characteristics of the high-quality tenants.

[0100] Based on the multi-dimensional revision of the contract elements, each adjustment of the contract elements has clear business data support, avoiding subjective decision bias. The revised contract terms are consistent with the actual situation of the tenants and can effectively control the enterprise risk. Finally, the high-quality contract terms that take into account the interests of both parties, are controllable in risk and can be implemented, so as to improve the stability of the lease business and the core target achievement rate.

[0101] Based on the above lease contract generation method, the present application also provides a lease contract generation device. The following will be described in detail. Figure 7 The device is described in detail.

[0102] Figure 7 The structure block diagram of the lease contract generation device according to the embodiments of the present application is schematically shown.

[0103] As shown in Figure 7 , the lease contract generation device 700 of this embodiment includes a sample module 710, a prediction module 720, a scoring module 730, and a revision module 740.

[0104] The sample module 710 is configured to obtain a sample contract of a sample interval in response to a contract generation request of a target interval, the sample interval being a same period interval of the target interval. In an embodiment, the sample module 710 can be configured to perform the operation S210 described above, which will not be described herein again.

[0105] The prediction module 720 is configured to obtain a renewal prediction value of the sample contract based on the lease data of the sample contract, the tenant feature and the subject feature. In an embodiment, the prediction module 720 can be configured to perform the operation S220 described above, and details are not repeated here.

[0106] The scoring module 730 is configured to score the sample contract based on the renewal prediction value, the tenant credit, the contract amount of the sample contract and the contract performance rate, and obtain a target sample contract based on the scoring result. In an embodiment, the scoring module 730 can be configured to perform the operation S230 described above, and details are not repeated here.

[0107] The correction module 740 is configured to correct a contract element in the target sample contract based on at least one of the scoring result, the tenant feature and the subject feature, to obtain the lease contract. In an embodiment, the correction module 740 can be configured to perform the operation S240 described above, and details are not repeated here.

[0108] According to an embodiment of the present application, any of the sample module 710, the prediction module 720, the scoring module 730 and the correction module 740 can be combined in one module, or any of them can be split into multiple modules. Alternatively, at least part of the function of one or more of these modules can be combined with at least part of the function of other modules, and implemented in one module. According to an embodiment of the present application, at least one of the sample module 710, the prediction module 720, the scoring module 730 and the correction module 740 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on chip, a system on board, a system on package, an application specific integrated circuit (ASIC), or any other reasonable way of hardware or firmware that can be integrated or packaged, or any one of software, hardware and firmware or any appropriate combination of several of them. Alternatively, at least one of the sample module 710, the prediction module 720, the scoring module 730 and the correction module 740 can be at least partially implemented as a computer program module which can perform corresponding functions when running.

[0109] Figure 8 A block diagram of an electronic device suitable for implementing the lease contract generation method according to an embodiment of the present application is schematically shown.

[0110] As Figure 8As shown, the electronic device 800 according to the embodiments of the present application includes a processor 801, which can perform various appropriate actions and processes according to a program stored in a read only memory (ROM) 802 or a program loaded into a random access memory (RAM) 803 from a storage section 808. The processor 801 can include, for example, a general purpose microprocessor (e.g., a CPU), an instruction set processor, and / or a related chipset, and / or a special purpose microprocessor (e.g., an application specific integrated circuit (ASIC)), and so on. The processor 801 can also include an on-board memory for cache use. The processor 801 can include a single processing unit or multiple processing units to perform the various actions of the method processes according to the embodiments of the present application.

[0111] In the RAM 803, various programs and data required for the operation of the electronic device 800 are stored. The processor 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. The processor 801 performs various operations of the method processes according to the embodiments of the present application by executing the programs in the ROM 802 and / or the RAM 803. Note that the programs can also be stored in one or more memories other than the ROM 802 and the RAM 803. The processor 801 can also perform various operations of the method processes according to the embodiments of the present application by executing the programs stored in the one or more memories.

[0112] According to the embodiments of the present application, the electronic device 800 can further include an input / output (I / O) interface 805, which is also connected to the bus 804. The electronic device 800 can further include one or more of the following components connected to the I / O interface 805: an input section 806 including a keyboard, a mouse, etc.; an output section 807 including a display such as a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 808 including a hard disk, etc.; and a communication section 809 including a network interface card such as a LAN card, a modem, etc. The communication section 809 performs communication processing via a network such as the Internet. A drive 810 is also connected to the I / O interface 805 as necessary. A removable recording medium 811 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is attached to the drive 810 as necessary, so that a computer program read therefrom is installed into the storage section 808 as necessary.

[0113] The present application also provides a computer readable storage medium, which can be included in the device / apparatus / system described in the above embodiments, or can exist separately without being assembled into the device / apparatus / system. The above computer readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of the present application.

[0114] According to an embodiment of the present application, the computer readable storage medium can be a non-transitory computer readable storage medium, for example, can include but not limited to: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In this application, a computer readable storage medium can be any tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. For example, according to an embodiment of the present application, the computer readable storage medium can include the ROM 802 and / or the RAM 803 described above and / or one or more memory other than the ROM 802 and the RAM 803.

[0115] Embodiments of the present application also include a computer program product, which includes a computer program containing program codes for executing the methods shown in the flowcharts. When the computer program product is run in a computer system, the program codes are used to make the computer system implement the item recommendation method provided by the embodiments of the present application.

[0116] The above functions defined in the system / device of the embodiments of the present application are performed when the computer program is executed by the processor 801. According to an embodiment of the present application, the system, device, module, unit, etc. described above can be implemented by computer program modules.

[0117] In one embodiment, the computer program can rely on tangible storage media such as optical storage media, magnetic storage media, etc. In another embodiment, the computer program can also be transmitted, distributed, and downloaded in the form of signals on network media. The computer program containing program codes can be transmitted by any appropriate network media, including but not limited to: wireless, wired, etc., or any suitable combination of the foregoing.

[0118] In such an embodiment, the computer program can be downloaded and installed from the network by the communication part 809, and / or installed from the detachable medium 811. When the computer program is executed by the processor 801, the above functions defined in the system of the embodiments of the present application are performed. According to an embodiment of the present application, the system, device, apparatus, module, unit, etc. described above can be implemented by computer program modules.

[0119] According to embodiments of the present application, program code for implementing the computer programs provided by embodiments of the present application can be written in any combination of one or more programming languages, and can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages include, but are not limited to, Java, C++, python, "C", or the like. Program code can execute entirely on a user's computing device, partly on the user's device, as a stand-alone software package, partly on a remote computing device, or entirely on the remote computing device or server. In the latter scenario, the remote computing device can be connected to the user's computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computing device, such as through the Internet using an Internet Service Provider.

[0120] The computer program instructions can also be loaded onto a computer or other programmable information processing apparatus to cause a series of operations to be performed on the computer or other programmable information processing apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable information processing apparatus implement the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0121] Those skilled in the art will understand that features of the various embodiments and / or claims of the present application can be combined or / and integrated with one another, even though such combinations or integrations are not expressly disclosed in the present application. In particular, features of the various embodiments and / or claims of the present application can be combined and / or integrated with one another in any manner, without departing from the spirit and teachings of the present application. All such combinations and / or integrations are within the scope of the present application.

[0122] The embodiments of the application have been described above. However, these embodiments are merely meant to be illustrative, and not meant to limit the scope of the application. Although each of the embodiments has been described above separately, this does not mean that the measures in the individual embodiments cannot be used advantageously in combination. The scope of the application is defined by the claims appended hereto and their equivalents. Various alternatives and modifications can be made to the embodiments of the application without departing from the scope of the application, and it is intended that all such alternatives and modifications be included within the scope of the application.

Claims

1. A method for generating a lease contract, characterized in that, include: In response to a contract generation request for a target range, a sample contract for a sample range is obtained, wherein the sample range is the same as the target range. Based on the lease data, tenant characteristics, and property characteristics of the sample contracts, the renewal prediction value of the sample contracts is obtained; Based on the renewal forecast, tenant credit, contract amount and performance rate of the sample contract, the sample contract is valued and a target sample contract is obtained based on the scoring results. Based on at least one of the scoring results, the tenant characteristics, and the subject matter characteristics, the contract elements in the target sample contract are modified to obtain a lease contract.

2. The lease contract generation method according to claim 1, characterized in that, Obtaining the renewal forecast value of the sample contracts includes: Based on the lease data, the tenant characteristics, and the characteristics of the subject matter, as well as the relationships between the multiple sample contracts, a graph structure data of the sample contracts is formed; Based on the graph structure data, the renewal prediction value is obtained using a graph neural network model, which is trained based on the graph structure data of historical contracts.

3. The lease contract generation method according to claim 2, characterized in that, The graph structure data that forms the sample contract includes: Nodes are obtained based on each of the sample contracts; Based on preset quantization rules, the rental data, tenant characteristics, and target object characteristics are converted into feature vectors to obtain node features; Node edges are constructed based on the relationships between multiple sample contracts; The graph structure data of the sample contract is formed based on the node features and the node edges.

4. The lease contract generation method according to claim 1, characterized in that, The target sample contract includes: Based on the leasing business objectives, the renewal forecast value, the tenant credit, the contract amount, and the contract performance rate are assigned corresponding weights, and the value score of the sample contracts is calculated. The leasing business objectives include at least one of risk control, revenue improvement, renewal rate improvement, and business balance. Output the sample contract with the highest value score to obtain the target sample contract.

5. The lease contract generation method according to claim 3, characterized in that, The formulation of the quantification rules includes: Based on the lease data, lease data quantification rules are formulated, wherein the lease data includes at least one of the following: institution code, contract type, historical renewal records, contract amount, lease term, and contract performance rate. Based on the tenant characteristics, tenant characteristic quantification rules are formulated. The tenant characteristics include at least one of the following: tenant code, tenant industry, years of cooperation, tenant credit, and historical renewal preferences. Based on the characteristics of the target asset, quantitative rules for the characteristics of the target asset are formulated. The characteristics of the target asset include at least one of the following: target asset code, target asset type, degree of aging, historical maintenance cost, market price, and market liquidity.

6. The lease contract generation method according to claim 5, characterized in that, The contract elements in the target sample contract are modified, including: Based on the scoring results, the duration of cooperation, and the degree of aging, the lease term is adjusted. The rent amount is adjusted based on the rating results, the tenant credit, and the market liquidity. The discount rate is adjusted based on the scoring results, the tenant credit, and the historical maintenance costs. The payment plan is revised based on the cooperation period, the contract amount, and the tenant's industry.

7. The lease contract generation method according to claim 6, characterized in that, After obtaining the lease agreement, it also includes: Based on the rental amount, lease term, and discount rate of the lease contract, lease liability forecast data is obtained; Based on the renewal forecast and the lease liability forecast, lease interest forecast and lease depreciation forecast are obtained.

8. The lease contract generation method according to claim 2, characterized in that, The graph neural network model is trained using the following method: Based on the lease data, the tenant characteristics, and the characteristics of the subject matter, as well as the relationships between multiple historical contracts, a graph structure data of the historical contracts is formed; Based on the renewal records of the historical contracts, each of the historical contracts is labeled with a renewal tag; The graph neural network model is trained based on the historical contract graph structure data and the renewal labels.

9. A lease contract generation device, characterized in that, include: The sample module is used to obtain sample contracts for the sample range in response to a contract generation request for the target range, wherein the sample range is the same as the target range. The prediction module is used to obtain the renewal prediction value of the sample contract based on the lease data, tenant characteristics and subject matter characteristics of the sample contract; The scoring module is used to score the value of the sample contract based on the renewal forecast value, tenant credit, contract amount and contract performance rate of the sample contract, and obtain the target sample contract based on the scoring results. The correction module is used to correct the contract elements in the target sample contract based on at least one of the scoring results, the tenant characteristics, and the subject matter characteristics, so as to obtain a lease contract.

10. An electronic device, characterized in that, include: One or more processors; Storage device for storing one or more programs. Wherein, when the one or more programs are executed by the one or more processors, the one or more processors perform the method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, It stores executable instructions that, when executed by a processor, cause the processor to perform the method according to any one of claims 1 to 8.

12. A computer program product, characterized in that, It includes a computer program that, when executed by a processor, implements the method according to any one of claims 1 to 8.

Images