Resource lending evaluation method and device, storage medium and computer program product
By acquiring vehicle traffic-related data and risk assessment models, and dynamically adjusting credit conditions, the accuracy problem of vehicle credit assessment in existing technologies has been solved, enabling more precise resource allocation services.
Patent Information
- Application Number
- CN202511694623.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-18
- Publication Date
- 2026-02-27
AI Technical Summary
Existing vehicle credit assessment methods rely too heavily on vehicle valuation and owner credit information, leading to over-crediting or under-crediting, and resulting in low assessment accuracy.
By acquiring traffic-related data of the target vehicle, the risk coefficient is determined. Combined with vehicle attribute data, behavioral data, and credit data, a risk assessment model is used to calculate the first risk coefficient, conduct credibility assessment and attribute value adjustment, and dynamically adjust credit conditions.
This improves the accuracy and efficiency of resource allocation services, avoids over-crediting or under-crediting, and ensures that credit decisions are matched with vehicle risk.
Smart Images
Figure CN121581986A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of computer technology, and more specifically, to an evaluation method and apparatus for resource borrowing, a storage medium, and a computer program product. Background Technology
[0002] With the booming development of the automotive industry and the deep integration of financial technology, vehicle resource allocation (such as credit) has become an indispensable part.
[0003] In the current vehicle loan model, credit assessment mainly relies on vehicle valuation and the owner's credit information. This kind of credit assessment depends too much on the vehicle valuation or the owner's credit score, which can easily lead to excessive risk exposure, resulting in over-credit or under-credit.
[0004] Therefore, there is an urgent need to provide an assessment method for resource borrowing in order to improve the accuracy of resource borrowing services. Summary of the Invention
[0005] This application provides a method and apparatus for evaluating resource borrowing, a storage medium, and a computer program product, in order to at least solve the problem of excessive credit or insufficient credit in related technologies, which leads to low accuracy of resource borrowing services.
[0006] According to one embodiment of this application, a method for assessing resource borrowing is provided, comprising: acquiring traffic-related data associated with a target vehicle, and determining a risk coefficient of the target vehicle; wherein the risk coefficient is used to characterize the degree of risk of borrowing resources from the target vehicle;
[0007] Based on the traffic-related data, a credibility assessment is performed on the target vehicle to determine the initial attribute value of the target vehicle for resource borrowing. Based on the risk coefficient and a pre-set attribute value adjustment strategy, an attribute value adjustment coefficient for the target vehicle is determined. The initial attribute value is adjusted based on the attribute value adjustment coefficient to determine the target attribute value of the target vehicle.
[0008] In an exemplary embodiment, determining the risk coefficient of the target vehicle includes: determining a first risk coefficient, a second risk coefficient, and a third risk coefficient of the target vehicle; wherein the first risk coefficient, the second risk coefficient, and the third risk coefficient are different; and weighting the first risk coefficient, the second risk coefficient, and the third risk coefficient to determine the risk coefficient of the target vehicle.
[0009] In an exemplary embodiment, the method for determining the first risk coefficient includes: acquiring vehicle attribute data, vehicle behavior data, and credit data of a target object associated with the target vehicle; fusing the vehicle attribute data, the vehicle behavior data, and the credit data to obtain fused data; and inputting the fused data into a pre-trained risk assessment model to obtain the first risk coefficient of the target vehicle.
[0010] In one exemplary embodiment, the method for determining the second risk coefficient includes: acquiring vehicle behavior data of the target vehicle; extracting multiple behavioral features of the target vehicle from the vehicle behavior data; and determining the second risk coefficient of the target vehicle based on the feature contribution values corresponding to the multiple behavioral features respectively.
[0011] In an exemplary embodiment, the step of assessing the credibility of the target vehicle based on the traffic-related data and determining the initial attribute values for resource borrowing of the target vehicle includes: performing continuity analysis, consistency analysis, anomaly analysis, and stability analysis on the traffic-related data to obtain continuity analysis results, consistency analysis results, anomaly analysis results, and stability analysis results; weighting the continuity analysis results, consistency analysis results, anomaly analysis results, and stability analysis results to determine the credibility level of the target vehicle; and determining the attribute values matching the credibility level as the initial attribute values for resource borrowing of the target vehicle.
[0012] In an exemplary embodiment, the attribute value adjustment strategy includes: a high-risk preference strategy or a low-risk preference strategy; determining the attribute value adjustment coefficient of the target vehicle based on the risk coefficient and the pre-set attribute value adjustment strategy includes: determining a first adjustment coefficient based on the risk coefficient and a first risk sensitivity indicated by the high-risk preference strategy, or determining a second adjustment coefficient based on the risk index and a second risk sensitivity indicated by the low-risk preference strategy; and determining the attribute value adjustment coefficient of the target vehicle based on the first adjustment coefficient or the second adjustment coefficient.
[0013] In one exemplary embodiment, adjusting the initial attribute value based on the attribute value adjustment coefficient to determine the target attribute value of the target vehicle includes: multiplying the attribute value adjustment coefficient and the initial attribute value to obtain a product; and determining the product as the target attribute value of the target vehicle.
[0014] According to another embodiment of this application, a resource borrowing assessment device is also provided, comprising: an acquisition module, configured to acquire traffic-related data associated with a target vehicle, and determine a risk coefficient of the target vehicle; wherein the risk coefficient is used to characterize the degree of risk of borrowing resources from the target vehicle; an assessment module, configured to perform a credibility assessment on the target vehicle based on the traffic-related data, determine an initial attribute value for resource borrowing of the target vehicle, and determine an attribute value adjustment coefficient for the target vehicle based on the risk coefficient and a pre-set attribute value adjustment strategy; and an adjustment module, configured to adjust the initial attribute value based on the attribute value adjustment coefficient, and determine a target attribute value for the target vehicle.
[0015] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, wherein a computer program is stored in the computer program, and the computer program is configured to execute the above-described method at runtime.
[0016] According to another aspect of the embodiments of this application, an electronic device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the above-described method through the computer program.
[0017] According to another aspect of the embodiments of this application, a computer program product is also provided, including a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.
[0018] In this embodiment, traffic-related data associated with the target vehicle is acquired, and the risk coefficient of the target vehicle is determined. Then, by analyzing the traffic-related data of the target vehicle, the initial attribute values for resource borrowing are determined, and an attribute value adjustment coefficient is calculated based on the risk coefficient of the target vehicle and a preset attribute value adjustment strategy. Finally, the initial attribute values are adjusted using the attribute value adjustment coefficient to obtain the target attribute values for the target vehicle. This ensures that resource borrowing matches the traffic-related data and risk coefficient of the target vehicle, avoiding over-crediting or under-crediting, and improving the accuracy and efficiency of resource borrowing services. Attached Figure Description
[0019] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:
[0020] Figure 1 This is a hardware structure block diagram of a computer device for evaluating a resource borrowing method according to an embodiment of this application;
[0021] Figure 2 This is a flowchart of an evaluation method for resource borrowing according to an embodiment of this application;
[0022] Figure 3 This is a flowchart of an evaluation method for resource borrowing according to an embodiment of this application;
[0023] Figure 4 This is a flowchart of an evaluation method for resource borrowing according to an embodiment of this application;
[0024] Figure 5 This is a structural block diagram of an evaluation device for resource borrowing according to an embodiment of this application. Detailed Implementation
[0025] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.
[0026] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus; "a plurality" means two or more.
[0027] The methods and embodiments provided in this application can be executed in a computer device or similar computing device, a cloud platform, a standalone physical server, or a software platform, wherein the aforementioned software platform runs through one or more servers. Taking running on a computer device as an example, Figure 1 This is a hardware structure block diagram of a resource borrowing evaluation method according to an embodiment of this application. For example... Figure 1 As shown, a computer device may include one or more ( Figure 1(Only one is shown in the image) A processor 102 and a memory 104 for storing data. In one exemplary embodiment, the computer device may further include a transmission device 106 for communication functions and an input / output device 108. The processor 102 may include, but is not limited to, a processing device such as a microprocessor (MCU) or a programmable logic device (FPGA). Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the computer device described above. For example, the computer device may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1 Equivalent functions or ratios shown Figure 1 The functions shown have more different configurations.
[0028] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the method in this embodiment. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, thus implementing the above-described method. The memory 104 may include high-speed random access memory and non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to computer devices via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.
[0029] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider for the computer equipment. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module used for wireless communication with the Internet.
[0030] This embodiment provides a method for evaluating resource borrowing, which is applied to the aforementioned computer equipment, which can be a server device. Figure 2 This is a flowchart of a resource borrowing assessment method according to an embodiment of this application, which includes the following steps:
[0031] Step S202: Obtain the traffic-related data associated with the target vehicle and determine the risk coefficient of the target vehicle; wherein, the risk coefficient is used to characterize the risk level of resource borrowing of the target vehicle.
[0032] It should be noted that toll-related data may include: ETC (Electronic Toll Collection System) toll records, GPS (Global Positioning System) location data, payment behavior data (such as refueling / charging records), information on the target vehicle entering parking lots or repair shops, and details such as the number of times, time, location, and amount of passage on possible toll roads or highways. Toll-related data can reflect key characteristics of the target vehicle, such as usage frequency, travel routes, and payment behavior, and is crucial for assessing the vehicle's credit risk.
[0033] The risk coefficient is a quantitative indicator used to assess the level of risk in providing resource lending services to the target vehicle. It integrates various factors such as the target vehicle's usage behavior, the creditworthiness of the associated target entities, and changes in the target vehicle's value; it is a dynamically calculated result. Resource lending refers to credit services provided to the target vehicle, including but not limited to vehicle loans and financial leasing.
[0034] Step S204: Based on traffic-related data, conduct a credibility assessment of the target vehicle, determine the initial attribute value of the target vehicle for resource borrowing, and determine the attribute value adjustment coefficient of the target vehicle based on the risk coefficient and a pre-set attribute value adjustment strategy.
[0035] Credibility assessment refers to the process of quantifying the authenticity and credibility of a specific data source or identity identifier. In this application, the assessment primarily targets toll-related data collected through the ETC system to determine the reliability of a vehicle's ETC digital identity.
[0036] Initial attribute values refer to the credit limit, interest rate, etc., set based on the comprehensive situation of the target vehicle (such as the credibility of the ETC digital identity, vehicle valuation, and owner credit score) before risk adjustment.
[0037] Understandably, after completing the credibility assessment, the initial attribute values for resource borrowing, such as the initial credit limit and interest rate, can be determined based on the assessment results and other attributes of the target vehicle (such as vehicle value, owner credit score, etc.). Next, according to the target vehicle's risk coefficient and a pre-defined attribute value adjustment strategy, an attribute value adjustment coefficient is calculated to dynamically adjust credit conditions, ensuring risk control while maximizing the efficiency of fund utilization. The attribute value adjustment coefficient is a value between 0 and 1, used to dynamically adjust the initial attribute values of the target vehicle based on the risk coefficient.
[0038] Step S206: Adjust the initial attribute values based on the attribute value adjustment coefficient to determine the target attribute values of the target vehicle.
[0039] Steps S202-S206 above involve acquiring traffic-related data associated with the target vehicle and determining its risk coefficient. By analyzing this traffic-related data, the initial attribute values for resource borrowing are determined. Based on the target vehicle's risk coefficient and a pre-defined attribute adjustment strategy, an attribute adjustment coefficient is calculated. Finally, the initial attribute values are adjusted using this coefficient to arrive at the target attribute values for the target vehicle. This ensures that resource borrowing matches the target vehicle's traffic-related data and risk coefficient, avoiding over-crediting or under-crediting, and improving the accuracy and efficiency of resource borrowing services.
[0040] In an exemplary embodiment, determining the risk coefficient of the target vehicle includes: determining a first risk coefficient, a second risk coefficient, and a third risk coefficient of the target vehicle; wherein the first risk coefficient, the second risk coefficient, and the third risk coefficient are different; and weighting the first risk coefficient, the second risk coefficient, and the third risk coefficient to determine the risk coefficient of the target vehicle.
[0041] It should be noted that the first, second, and third risk coefficients represent different risk dimensions, each with varying degrees of influence on the final credit decision. Therefore, the first, second, and third risk coefficients need to be weighted to determine their relative importance in the overall risk score. The weighting can be based on historical data analysis, industry standards, etc.
[0042] In some embodiments, the method for determining the first risk coefficient includes: acquiring vehicle attribute data, vehicle behavior data, and credit data of a target object associated with the target vehicle; fusing the vehicle attribute data, the vehicle behavior data, and the credit data to obtain fused data; and inputting the fused data into a pre-trained risk assessment model to obtain the first risk coefficient of the target vehicle.
[0043] Vehicle attribute data may include: the target vehicle's model, year, mileage, maintenance history, insurance records, etc. Obtaining vehicle behavior data involves the target vehicle's recent traffic records, driving habits (such as frequency of rapid acceleration and braking), mileage, and vehicle location records. This type of data can be collected through ETC systems, on-board OBD devices, or vehicle-to-everything (V2X) platforms. Credit data for the target entity refers to the vehicle owner's credit history, repayment ability, and debt situation.
[0044] After acquiring the aforementioned data, these scattered, multi-dimensional data need to be fused to form a comprehensive dataset containing both static vehicle attributes and dynamic behavioral characteristics—this is the fused data. Further, a risk assessment model, such as a random forest, XGBoost, or neural network, is selected. The model is trained using historical credit data and other relevant information to identify patterns or characteristics associated with high-risk vehicles. The fused data, including all relevant vehicle attributes, behavioral characteristics, and credit information, is used as input to the risk assessment model, ensuring that the model can comprehensively assess risk based on this information. The risk assessment model calculates a first risk coefficient based on the input fused data. This is typically a numerical value reflecting the credit risk level of the target vehicle. Lower values represent lower risk, while higher values indicate higher risk.
[0045] In some embodiments, the method for determining the second risk coefficient includes: acquiring vehicle behavior data of the target vehicle; extracting multiple behavioral features of the target vehicle from the vehicle behavior data; and determining the second risk coefficient of the target vehicle based on the feature contribution values corresponding to the multiple behavioral features respectively.
[0046] This involves acquiring real-time dynamic data of the target vehicle from channels such as the ETC system, on-board OBD devices, and vehicle-to-everything (V2X) platforms. This data includes, but is not limited to, the target vehicle's driving trajectory, driving behavior (such as rapid acceleration and braking), traffic frequency, and stopping locations. Historical driving data of the target vehicle, such as historical mileage, regular maintenance records, and accident reports, is also collected to construct a more comprehensive behavioral pattern.
[0047] From the collected vehicle behavior data, key behavioral characteristics that significantly impact risk assessment are extracted. These characteristics may include: Mileage: Recent average mileage; longer mileage may indicate higher usage frequency and faster vehicle wear, thus increasing risk. Nighttime Driving Ratio: The proportion of time spent driving at night to total driving time; a higher nighttime driving ratio may be associated with unsafe driving habits or potential illegal activities. Frequency of Sudden Acceleration and Braking: The number of times sudden acceleration and braking occur; frequent sudden acceleration and braking reflect potentially aggressive driving habits, increasing the risk of accidents. Abnormal Traffic Records: Including records such as frequent entry into used car markets or failed cross-verification; these abnormal behaviors may be associated with fraud risk.
[0048] Understandably, machine learning models (such as random forests and XGBoost) can be used to analyze vehicle behavior data to determine which behavioral features contribute most to risk assessment. The model can quantify this contribution using methods such as SHAP values and feature importance scores. Based on the results of feature importance analysis, a feature weight is assigned to each behavioral feature, representing its relative importance in risk assessment. The value of each behavioral feature is standardized and then multiplied by its corresponding feature weight to obtain the feature contribution value. The feature contribution values of all behavioral features are summed, and then a weighted average of the sum and feature weights is used to obtain the second risk coefficient of the target vehicle.
[0049] In some embodiments, the third risk coefficient refers to a quantitative indicator reflecting the credit status of a target vehicle, derived from the analysis and evaluation of its credit information. The third risk coefficient is calculated based on credit behavior data, including but not limited to loan repayment records, credit card usage, default history, and debt levels.
[0050] In the above embodiments, different risk dimensions have different impacts on the final credit decision. Therefore, by determining the first risk coefficient, the second risk coefficient, and the third risk coefficient, and then weighting the first risk coefficient, the second risk coefficient, and the third risk coefficient, their relative importance in the total risk score is determined, thereby improving the accuracy of the risk coefficient.
[0051] In an exemplary embodiment, the step of assessing the credibility of the target vehicle based on the access-related data and determining the initial attribute values for resource borrowing of the target vehicle includes: performing continuity analysis, consistency analysis, anomaly analysis, and stability analysis on the access-related data to obtain continuity analysis results, consistency analysis results, anomaly analysis results, and stability analysis results; weighting the continuity analysis results, consistency analysis results, anomaly analysis results, and stability analysis results to determine the credibility level of the target vehicle; and determining the attribute values matching the credibility level as the initial attribute values for resource borrowing of the target vehicle.
[0052] Continuity analysis is used to check whether the traffic data of the target vehicle is consistent, that is, whether there are stable traffic records within a certain time period (such as the most recent year). This helps to determine whether the vehicle is in normal operating condition and whether the owner is using the vehicle consistently.
[0053] Consistency analysis is used to compare location information in traffic data with the target individual's daily activity areas. For example, does the traffic data match the vehicle owner's home address, workplace, and frequently used routes? This can verify the authenticity of the data and prevent fraudulent activities.
[0054] Anomaly analysis is used to identify abnormal patterns in traffic data, such as sudden high toll fees or frequent cross-regional travel. These abnormal patterns may indicate risks, such as cloned vehicles or vehicles being used for illegal activities.
[0055] Stability analysis is used to assess the stability and reliability of payment channels, including but not limited to the consistency of ETC payments, the success rate of deductions, and the existence of frequent payment problems. A stable payment pattern usually reflects the financial stability of the vehicle owner.
[0056] After completing the above analysis, the results of continuity analysis, consistency analysis, anomaly analysis, and stability analysis need to be weighted and integrated to determine the credibility level of the target vehicle. Based on business experience and historical data analysis, a weight is assigned to each analysis result to reflect its relative importance in the credibility assessment. For example, continuity analysis may have the highest weight because a stable and consistent passage record is a key indicator of credibility.
[0057] Furthermore, the credibility level is mapped to a predefined rating system, such as A, B, and C levels. These levels reflect the credit risk level of the target vehicle, with A indicating low risk and C indicating high risk. Once the credibility level of the target vehicle is determined, the initial attribute values for resource borrowing, including the initial credit limit and loan interest rate, can be set based on this level. Each credibility level corresponds to a range of initial attribute values. For example, an A-level vehicle might correspond to a higher initial credit limit and a lower loan interest rate, while a C-level vehicle might correspond to a lower initial credit limit and a higher loan interest rate. Although the initial attribute values are determined based on the credibility level, financial institutions will also dynamically adjust them based on other risk assessment results of the target vehicle (such as the owner's credit history, vehicle value, etc.) to ensure that the attribute values are set more in line with the actual risk situation.
[0058] In the above embodiments, by converting the analysis results of access-related data into a credibility level and setting the initial attribute value of resource borrowing accordingly, financial institutions can more effectively manage credit risk and provide financial services that better meet market and user needs.
[0059] In an exemplary embodiment, the attribute value adjustment strategy includes: a high-risk preference strategy or a low-risk preference strategy; determining the attribute value adjustment coefficient of the target vehicle based on the risk coefficient and the pre-set attribute value adjustment strategy includes: determining a first adjustment coefficient based on the risk coefficient and a first risk sensitivity indicated by the high-risk preference strategy, or determining a second adjustment coefficient based on the risk index and a second risk sensitivity indicated by the low-risk preference strategy; and determining the attribute value adjustment coefficient of the target vehicle based on the first adjustment coefficient or the second adjustment coefficient.
[0060] It's important to note that a high-risk-preference strategy tends to involve taking on higher risks in exchange for potentially higher returns. This strategy provides higher-risk clients with more financing opportunities, or sets lower risk adjustment factors, allowing for larger credit lines or lower interest rates, but simultaneously strengthens post-loan management and sets stricter default clauses.
[0061] Conversely, low-risk preference strategies are more cautious, aiming to reduce losses and ensure capital safety. Under this strategy, even lower-risk clients may face higher risk adjustment factors, leading to limited credit lines or higher interest rates to compensate for potential risk exposure.
[0062] Upon receiving the risk coefficient of the target vehicle, a specific adjustment coefficient is determined based on a pre-set attribute value adjustment strategy. This process involves two different risk sensitivity assessments: primary risk sensitivity and secondary risk sensitivity. Under a high-risk-preference strategy, a more optimistic attitude towards risk is adopted. Even with a high risk coefficient, a relatively small primary adjustment coefficient is set, meaning the adjustment range for the target vehicle's credit attribute value is smaller, tending to maintain or slightly increase its initial attribute value. Under a low-risk-preference strategy, a more cautious attitude towards risk is adopted. Any increase in the risk coefficient will lead to a larger secondary adjustment coefficient, thereby significantly reducing the target vehicle's credit attribute value, such as reducing the credit limit or increasing the interest rate.
[0063] In the above embodiments, by flexibly applying high-risk or low-risk preference strategies, the initial attribute values of the target vehicle can be dynamically adjusted according to different management strategies. This attribute value adjustment mechanism based on risk coefficients not only enhances the flexibility of resource borrowing products but also effectively controls risks and ensures asset security.
[0064] In an exemplary embodiment, adjusting the initial attribute value based on the attribute value adjustment coefficient to determine the target attribute value of the target vehicle includes: multiplying the attribute value adjustment coefficient and the initial attribute value to obtain a product; and determining the product as the target attribute value of the target vehicle.
[0065] To better understand the above method, the following embodiments are used to illustrate the process, but these are not intended to limit the technical solutions of the embodiments of this application. Specifically:
[0066] The resource borrowing assessment method provided in this application can be divided into the following stages in its specific implementation:
[0067] System deployment preparation phase: Define system boundaries, build basic technical architecture, and complete compliance design.
[0068] Demand Survey: We held a demand seminar with financial institutions, ETC issuers, and automakers to clarify core functions. For example, for C-end car owners, we provide unsecured credit, dynamic credit limit adjustment, and targeted fund payments (fueling / maintenance / insurance); for B-end financial institutions, we provide pre-loan risk profiling, mid-loan credit limit management, and post-loan risk warning; and for regulators, we provide data storage and compliance auditing.
[0069] The technical architecture design adopts a cloud-edge-device layered architecture, with the following layers: Terminal layer: vehicle owner side (on-board OBU device, mobile APP), financial institution side (credit system interface); Edge layer: ETC issuer data interface gateway (responsible for ETC data cleaning and de-identification), financial institution risk control engine (real-time calculation of risk index); Cloud layer: blockchain evidence storage platform (stores key data hash values), federated learning platform (multi-source data joint modeling), big data platform (stores and analyzes full data).
[0070] During the core module development phase, the code development and unit testing of four core modules were completed: ETC digital identity construction, multi-source data fusion, dynamic risk control model, and targeted fund payment.
[0071] ETC Digital Identity Construction Module: Data Collection Interface Development: Connects to the ETC issuer's API to obtain basic ETC card information (license plate number, OBU device number, issuer code, vehicle owner's ID number hash value), toll transaction history (timestamp, entrance / exit toll station ID, amount, transaction type), and payment characteristics (payment account type: bank card / Alipay / WeChat linked to ETC; deduction status: successful / failed / repayment); Connects to the traffic management system API to obtain vehicle registration information (vehicle identification number, engine number, registration date, annual inspection status).
[0072] Data cleaning: Filter abnormal transactions (such as transactions with an amount of 0 yuan or repeated transactions at the same toll station within 10 minutes) using regular expressions, and exclude cross-day transaction errors using timestamp verification.
[0073] ETC Digital Identity ID Generation: The SHA-256 hash algorithm is used to combine the "ETC card number + OBU device number + vehicle frame number" to generate a unique identity ID (e.g., a1b2c3d4e5f6), ensuring that it cannot be tampered with; an ETC digital identity file is established to store the basic information associated with the identity ID, toll transaction history, and payment characteristics (structured storage in MySQL, and unstructured data storage in HDFS).
[0074] Credibility Assessment Submodule: Set assessment indicators (as shown in Table 1), calculate credibility scores (0-100 points) through the rule engine (Drools), and classify them into three levels: A (≥80), B (60-79), and C (<60). Level A can directly pass the initial identity credibility review and enter the fully automated credit process; Level B will trigger an identity verification mechanism, that is, the system automatically retrieves the vehicle photo from the traffic management system, compares it with the VIN code of the vehicle bound to ETC, and can only enter the credit granting process after manual verification; Level C will refuse unsecured credit or guide to the collateral supplementary material process.
[0075] Table 1
[0076]
[0077] For the multi-source data fusion module, refer to Figure 3 The diagram shown is a structural block diagram of multi-source data fusion, including: external data interface development: credit data interface, connecting to the credit reporting center API to obtain the car owner's personal credit report (total debt, number of overdue payments, credit card usage rate); vehicle attribute data, obtaining car insurance records (number of accidents, repair costs), obtaining maintenance records (last maintenance time, mileage); behavioral data, connecting to the vehicle networking platform to obtain driving behavior data (number of rapid accelerations / 100 km / h, average vehicle speed, percentage of nighttime driving).
[0078] Data Synchronization: Real-time data synchronization is achieved using Kafka message queues (e.g., ETC toll records are pushed every 5 minutes), and historical data is synchronized to the big data platform through daily incremental backups. Storage Structure: Relational data (e.g., vehicle owner basic information): MySQL (database sharding, partitioned by vehicle owner ID hash); Time-series data (e.g., toll records, driving behavior): DB (supporting efficient time range queries); Unstructured data (e.g., maintenance record images): Object storage (e.g., Alibaba Cloud OSS). Feature Engineering Development: Feature extraction is performed on multi-source data based on Spark SQL, with basic category weights for the three data types during feature extraction (pre-loan scenario). As shown in Table 2, these weights can be dynamically adjusted based on user information using a federated learning framework.
[0079] Table 2
[0080]
[0081] Dynamic Risk Control Model Development: Model Selection and Training: XGBoost (for structured data) and LSTM (for time-series data, such as traffic flow trends over the past 6 months) are used to train sub-models separately, which are then integrated via Stacking. Training Data: Historical credit samples (normal repayment / overdue for 90+ days) are used, labeled with default tags (Y=1 indicates overdue, Y=0 indicates normal). Feature Importance Analysis: Key features are determined using SHAP values (e.g., the proportion of nighttime traffic over the past 3 months has the highest impact). Real-time Risk Calculation: A Flink real-time computing engine is deployed to receive real-time data such as ETC passage and driving behavior pushed by Kafka, updating the vehicle owner risk index hourly (range 0-100, higher values indicate greater risk).
[0082] refer to Figure 4 The diagram shown illustrates the flowchart involved in calculating the risk coefficient. By inputting multi-source data and ETC information into the risk control model, multiple modules of the model process the input, output the model, and ultimately determine the risk index. The following is the formula for calculating the risk coefficient, i.e., the risk index:
[0083] RiskScore=0.4×ModelScore+0.3×CreditScore+0.3×BehaviorScore;
[0084] Here, ModelScore is the output of the integrated model, i.e., the first risk coefficient; BehaviorScore is the driving behavior score, i.e., the second risk coefficient; and CreditScore is the credit score, i.e., the third risk coefficient. BehaviorScore is the quantitative score after detailed extraction of behavioral features. It is the result of cleaning, standardizing, and weighting the behavioral data, transforming scattered behavioral signals into comparable risk values. The calculation logic is as follows: BehaviorScore = ∑(Behavioral Feature Contribution Value i × Behavioral Feature Weight). The behavioral feature contribution value is the standardized value of the original data, and the behavioral feature weight is determined through business experience. In the risk index calculation formula, if ModelScore already includes behavioral data features fused from multiple sources, but BehaviorScore is used again, it's because ModelScore is the comprehensive output, while BehaviorScore is an intermediate input and also an independent quantitative result of the behavioral data. Introducing BehaviorScore allows for a more comprehensive consideration of the impact of user behavior. Risk coefficients can be used to map credit limit adjustment coefficients, thereby calculating the final credit limit based on the coefficients.
[0085] Furthermore, a preset risk index-credit limit coefficient mapping rule (dynamically adjustable) is established to convert the comprehensive risk index into a credit limit adjustment ratio. The risk sensitivity is set according to business objectives, such as 1.2 for a conservative strategy and 0.8 for an aggressive strategy. The calculation method for the credit limit adjustment coefficient is as follows:
[0086] Credit limit adjustment factor = 1 100 Risk Index × Risk Sensitivity;
[0087] The final credit limit is then: base credit limit (determined by the credibility rating) × credit limit adjustment coefficient, and is also subject to "maximum / minimum credit limit".
[0088] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of this application.
[0089] This application also provides a structural block diagram of a resource borrowing assessment device in its embodiments. Figure 5 This is a structural block diagram of an evaluation apparatus for resource borrowing according to an embodiment of this application; as shown below. Figure 5 As shown, it includes:
[0090] The acquisition module 52 is used to acquire traffic-related data associated with the target vehicle and to determine the risk coefficient of the target vehicle; wherein the risk coefficient is used to characterize the degree of risk of resource borrowing of the target vehicle;
[0091] The evaluation module 54 is used to evaluate the credibility of the target vehicle based on the traffic-related data, determine the initial attribute value of the target vehicle for resource borrowing, and determine the attribute value adjustment coefficient of the target vehicle based on the risk coefficient and the pre-set attribute value adjustment strategy.
[0092] The adjustment module 56 is used to adjust the initial attribute value based on the attribute value adjustment coefficient to determine the target attribute value of the target vehicle.
[0093] The aforementioned device acquires traffic-related data associated with the target vehicle and determines its risk coefficient. By analyzing this data, the initial attribute values for resource allocation are determined, and an attribute adjustment coefficient is calculated based on the vehicle's risk coefficient and a pre-defined adjustment strategy. Finally, the initial attribute values are adjusted using this coefficient to arrive at the target vehicle's target attribute values. This ensures that resource allocation matches the target vehicle's traffic-related data and risk coefficient, preventing over-crediting or under-crediting and improving the accuracy and efficiency of resource allocation services.
[0094] In an exemplary embodiment, the evaluation module 54 is further configured to determine a first risk coefficient, a second risk coefficient, and a third risk coefficient of the target vehicle; wherein the first risk coefficient, the second risk coefficient, and the third risk coefficient are different; and to determine the risk coefficient of the target vehicle by weighting the first risk coefficient, the second risk coefficient, and the third risk coefficient.
[0095] In an exemplary embodiment, the evaluation module 54 is further configured to acquire vehicle attribute data, vehicle behavior data, and credit data of target objects associated with the target vehicle; fuse the vehicle attribute data, the vehicle behavior data, and the credit data to obtain fused data; and input the fused data into a pre-trained risk assessment model to obtain a first risk coefficient of the target vehicle.
[0096] In an exemplary embodiment, the evaluation module 54 is further configured to acquire vehicle behavior data of the target vehicle; extract multiple behavioral features of the target vehicle from the vehicle behavior data; and determine a second risk coefficient of the target vehicle based on the feature contribution values corresponding to the multiple behavioral features respectively.
[0097] In an exemplary embodiment, the evaluation module 54 is further configured to perform continuity analysis, consistency analysis, anomaly analysis, and stability analysis on the passage-related data to obtain continuity analysis results, consistency analysis results, anomaly analysis results, and stability analysis results; to determine the credibility level of the target vehicle by weighting the continuity analysis results, consistency analysis results, anomaly analysis results, and stability analysis results; and to determine the attribute value matching the credibility level as the initial attribute value for resource borrowing of the target vehicle.
[0098] In an exemplary embodiment, the attribute value adjustment strategy includes a high-risk preference strategy or a low-risk preference strategy; the adjustment module 56 is further configured to determine a first adjustment coefficient based on the risk coefficient and a first risk sensitivity indicated by the high-risk preference strategy, or to determine a second adjustment coefficient based on the risk index and a second risk sensitivity indicated by the low-risk preference strategy; and to determine the attribute value adjustment coefficient of the target vehicle based on the first adjustment coefficient or the second adjustment coefficient.
[0099] In an exemplary embodiment, the adjustment module 56 is further configured to multiply the attribute value adjustment coefficient and the initial attribute value to obtain a product; and to determine the product as the target attribute value of the target vehicle.
[0100] Embodiments of this application also provide a storage medium including a stored program, wherein the program executes any of the methods described above when it is run.
[0101] Optionally, in this embodiment, the storage medium may be configured to store program code for performing the following steps:
[0102] S1, acquire traffic-related data associated with the target vehicle, and determine the risk coefficient of the target vehicle; wherein, the risk coefficient is used to characterize the degree of risk of resource borrowing of the target vehicle;
[0103] S2, based on the traffic-related data, perform a credibility assessment on the target vehicle, determine the initial attribute value of the target vehicle for resource borrowing, and based on the risk coefficient and a pre-set attribute value adjustment strategy, determine the attribute value adjustment coefficient of the target vehicle.
[0104] S3, adjust the initial attribute value based on the attribute value adjustment coefficient to determine the target attribute value of the target vehicle.
[0105] Embodiments of this application also provide an electronic device including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.
[0106] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.
[0107] Optionally, in this embodiment, the processor can be configured to perform the following steps via a computer program:
[0108] S1, acquire traffic-related data associated with the target vehicle, and determine the risk coefficient of the target vehicle; wherein, the risk coefficient is used to characterize the degree of risk of resource borrowing of the target vehicle;
[0109] S2, based on the traffic-related data, perform a credibility assessment on the target vehicle, determine the initial attribute value of the target vehicle for resource borrowing, and based on the risk coefficient and a pre-set attribute value adjustment strategy, determine the attribute value adjustment coefficient of the target vehicle.
[0110] S3, adjust the initial attribute value based on the attribute value adjustment coefficient to determine the target attribute value of the target vehicle.
[0111] Optionally, in this embodiment, the storage medium may include, but is not limited to, various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.
[0112] The collection, storage, use, processing, transmission, provision, and disclosure of financial data or user data involved in the technical solution of this application all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.
[0113] It should be noted that in the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, it does not mean that the applicant has used or necessarily used the solution.
[0114] Embodiments of this application also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.
[0115] Embodiments of this application also provide another computer program product, including a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.
[0116] Embodiments of this application also provide a computer program that includes computer instructions stored in a computer-readable storage medium; a processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps in any of the above method embodiments.
[0117] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementations, and will not be repeated here.
[0118] Obviously, those skilled in the art should understand that the modules or steps of this application described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented here, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.
[0119] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.
Claims
1. A method for evaluating resource borrowing, characterized in that, include: Acquire traffic-related data associated with the target vehicle and determine the risk coefficient of the target vehicle; wherein the risk coefficient is used to characterize the degree of risk of resource borrowing of the target vehicle; Based on the traffic-related data, the credibility of the target vehicle is assessed, the initial attribute value of the target vehicle for resource borrowing is determined, and the attribute value adjustment coefficient of the target vehicle is determined based on the risk coefficient and the pre-set attribute value adjustment strategy. The initial attribute value is adjusted based on the attribute value adjustment coefficient to determine the target attribute value of the target vehicle.
2. The method according to claim 1, characterized in that, Determining the risk coefficient of the target vehicle includes: A first risk coefficient, a second risk coefficient, and a third risk coefficient are determined for the target vehicle; wherein the first risk coefficient, the second risk coefficient, and the third risk coefficient are different. The risk coefficient of the target vehicle is determined by weighting the first risk coefficient, the second risk coefficient, and the third risk coefficient.
3. The method according to claim 2, characterized in that, The method for determining the first risk coefficient includes: Obtain vehicle attribute data, vehicle behavior data, and credit data of target objects associated with the target vehicle; The vehicle attribute data, the vehicle behavior data, and the credit data are combined to obtain fused data; The fused data is input into a pre-trained risk assessment model to obtain the first risk coefficient of the target vehicle.
4. The method according to claim 2, characterized in that, The methods for determining the second risk factor include: Obtain vehicle behavior data of the target vehicle; Extract multiple behavioral features of the target vehicle from the vehicle behavior data; Based on the feature contribution values corresponding to the multiple behavioral characteristics, the second risk coefficient of the target vehicle is determined.
5. The method according to claim 1, characterized in that, The step of assessing the credibility of the target vehicle based on the traffic-related data and determining the initial attribute values for resource borrowing of the target vehicle includes: The passage-related data are subjected to continuity analysis, consistency analysis, anomaly analysis, and stability analysis to obtain the results of continuity analysis, consistency analysis, anomaly analysis, and stability analysis. The credibility level of the target vehicle is determined by weighting the results of the continuity analysis, consistency analysis, anomaly analysis, and stability analysis. The attribute values that match the credibility level are determined as the initial attribute values for the resource borrowing of the target vehicle.
6. The method according to claim 1, characterized in that, The attribute value adjustment strategy includes: a high-risk preference strategy or a low-risk preference strategy; determining the attribute value adjustment coefficient of the target vehicle based on the risk coefficient and the pre-set attribute value adjustment strategy includes: A first adjustment coefficient is determined based on the risk coefficient and the first risk sensitivity indicated by the high-risk preference strategy, or a second adjustment coefficient is determined based on the risk index and the second risk sensitivity indicated by the low-risk preference strategy. The attribute value adjustment coefficient of the target vehicle is determined based on the first adjustment coefficient or the second adjustment coefficient.
7. The method according to claim 1, characterized in that, The step of adjusting the initial attribute value based on the attribute value adjustment coefficient to determine the target attribute value of the target vehicle includes: The product is obtained by multiplying the attribute value adjustment coefficient and the initial attribute value; The product is determined as the target attribute value of the target vehicle.
8. An assessment device for resource borrowing, characterized in that, include: The acquisition module is used to acquire traffic-related data associated with the target vehicle and to determine the risk coefficient of the target vehicle; wherein, the risk coefficient is used to characterize the degree of risk of resource borrowing of the target vehicle; The evaluation module is used to evaluate the credibility of the target vehicle based on the traffic-related data, determine the initial attribute value of the target vehicle for resource borrowing, and determine the attribute value adjustment coefficient of the target vehicle based on the risk coefficient and the pre-set attribute value adjustment strategy. An adjustment module is used to adjust the initial attribute value based on the attribute value adjustment coefficient to determine the target attribute value of the target vehicle.
9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein the program, when executed, performs the method described in any one of claims 1 to 7.
10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1 to 7.