A cross-industry asset management method and system for financial leasing

Abstract

The application provides a cross-industry asset management method and system for financial leasing, which is applied to the field of asset data processing; the application generates asset data standards and their description information, ensures that the integration of cross-industry data has a clear structure and unified standards, and is crucial for subsequent data analysis, risk assessment and decision support; meanwhile, through analysis of financial leasing information, cross-industry asset data related to the financial leasing information is identified and integrated into the decision-making process, breaking the barriers between industries and allowing more comprehensive asset data to be used for evaluation; and through a cross-industry collaborative decision-making mechanism, data and insights from multiple fields can be combined to generate more valuable decision-making information; this mechanism not only improves the efficiency of financial leasing approval, but also effectively reduces manual intervention and decision-making delay.

Description

Technical Field

[0001] This invention relates to the field of asset data processing, and in particular to a cross-industry asset management method and system for financial leasing. Background Technology

[0002] Financial leasing, as a flexible financing method, is widely used in various industries, providing enterprises with the right to use equipment and assets while reducing their capital expenditures. However, the financial leasing process involves multiple participants, including leasing companies, customers, and equipment suppliers, and the entire process is accompanied by various risks.

[0003] In the financial leasing industry, data from various parties is often scattered across different systems and platforms, resulting in information silos and hindering effective data sharing and integration, which in turn affects the accuracy and timeliness of risk assessment. Summary of the Invention

[0004] This invention aims to address the problem of how to achieve effective sharing and integration of cross-industry asset data, improve the accuracy and timeliness of risk assessment, and provides a cross-industry asset management method and system for financial leasing.

[0005] The present invention employs the following technical means to solve the technical problem:

[0006] This invention provides a cross-industry asset management method for financial leasing, comprising:

[0007] Based on pre-input business data requirements, the business data requirements are entered into a preset data lake, and the data lake generates the data type corresponding to the business data requirements. Specifically, the data type includes text, image, video, and log file.

[0008] Determine whether the data type is compatible with the preset data lake platform;

[0009] If possible, based on the preset data lake architecture, the business data requirements are organized in the data lake using a preset layered structure, asset data standards corresponding to the data types are constructed, descriptive information of the asset data standards is generated, the business data requirements are migrated from the preset data center to the data lake, and the financial leasing information of the business data requirements is divided. Specifically, the data lake architecture includes a data storage layer, a data processing layer, and a data access layer. The descriptive information specifically includes the source, format, and creation time. The financial leasing information specifically includes financial leasing contracts, asset valuation, and risk management.

[0010] Determine whether the finance lease information meets the preset completeness requirements;

[0011] If the conditions are met, the user's cross-industry asset data is identified based on the financial leasing information. The financial leasing information and the cross-industry asset data are then input into a pre-trained risk dimension data model to update the user's cross-industry collaborative decision-making mechanism. The financial leasing approval information corresponding to the business data requirements is then generated through the cross-industry collaborative decision-making mechanism. Specifically, the cross-industry asset data includes asset scope, customer groups, and market popularity.

[0012] Furthermore, before the step of organizing the business data requirements in the data lake according to the preset data lake architecture and using a preset hierarchical structure, the method further includes:

[0013] Based on the source data pre-collected by the data lake, the industry type corresponding to the business data requirement is identified, wherein the source data specifically includes structured data, semi-structured data and unstructured data;

[0014] Determine whether the industry type covers the source data;

[0015] If not, then based on the industry type, obtain the data source for the business data requirement, extract the corresponding business data from the data source, convert the business data into the preset data format of the data lake, and periodically update the new data of the business data in the data lake within a preset time period. The data source specifically includes production equipment sensor data, inventory management data, and supply chain data.

[0016] Furthermore, the step of migrating the business data requirements from the preset data center to the data lake and classifying the financial leasing information of the business data requirements further includes:

[0017] Based on the preset data fields of the financial leasing information, key fields in the data fields are identified, wherein the key fields specifically include contract number, lease amount, lease term, customer ID and asset number;

[0018] Determine whether the key fields match a preset uniform format;

[0019] If so, then based on the key fields, a preset identifier is applied to associate the financial leasing information with the user, establish index information of the financial leasing information in the data lake, and grant the user access rights to the index information.

[0020] Furthermore, after the step of inputting the financial lease information and the cross-industry asset data into the pre-trained risk dimension data model, the method further includes:

[0021] Based on preset risk factors, corresponding risk levels are divided, and a risk dimension data model is constructed according to the risk levels. Specifically, the risk factors include asset risk, customer credit risk, and market risk, and the risk levels include customer layer, contract layer, asset layer, and market layer.

[0022] Determine whether the risk dimension data model can generate risk assessment results;

[0023] If not, obtain the error log of the risk dimension data model during runtime, apply the preset outlier filter to identify extreme values ​​of the cross-industry asset data from the error log, and restrict the user from conducting risk assessment based on the extreme values. Specifically, the extreme values ​​include extremely large lease amounts, excessively low customer credit scores, and excessively long lease terms.

[0024] Furthermore, the step of updating the user's cross-industry collaborative decision-making mechanism and generating the financing lease approval information corresponding to the business data requirement through the cross-industry collaborative decision-making mechanism further includes:

[0025] Based on the decision variable information preset by the data center for the business data requirements, the decision variable information is synchronized to the preset industry departments, and the approval standard value of the user is established through the industry departments. The decision variable information specifically includes customer credit score, asset valuation and contract terms.

[0026] Determine whether the approval standard value meets the preset review benchmark;

[0027] If so, the user and their financial leasing information will be shared with the industry department's digital collaboration platform according to the preset execution permissions of the data center, and the user's financial leasing approval information will be constructed in real time on the digital collaboration platform.

[0028] Furthermore, the step of determining whether the data type is compatible with the preset data lake platform also includes:

[0029] Obtain the preset time field for the business data requirements;

[0030] Determine whether the time field matches the preset timestamp unit of the data lake platform;

[0031] If not, then unify the time field according to the timestamp unit, and reorder the timestamp partitions of the business data requirements according to the time window of the timestamp unit.

[0032] Furthermore, the step of inputting the pre-input business data requirements into a preset data lake and generating the data type corresponding to the business data requirements through the data lake further includes:

[0033] Based on the pre-set cleaning measures of the data center, detect the items to be cleaned in the business data requirements;

[0034] Determine whether the item to be cleaned matches a preset cleanable item;

[0035] If so, the business data requirements will be pre-cleaned to generate data to be confirmed. The data to be confirmed will be synchronized to the user's preset terminal. Through the data center, the operation permissions for the business data requirements provided by the user will be obtained. Specifically, the data pre-cleaning includes removing useless fields, processing missing values, and correcting erroneous data.

[0036] This invention also provides a cross-industry asset management system for financial leasing, comprising:

[0037] The generation module is used to input the pre-input business data requirements into a preset data lake based on the pre-input business data requirements, and generate the data type corresponding to the business data requirements through the data lake. The data type specifically includes text, image, video and log file.

[0038] The judgment module is used to determine whether the data type can be adapted to the preset data lake platform;

[0039] The execution module is used, if possible, to organize the business data requirements in the data lake according to the preset data lake architecture and the preset layered structure, construct the asset data standard corresponding to the data type, generate the description information of the asset data standard, migrate the business data requirements from the preset data center to the data lake, and divide the financial leasing information of the business data requirements. The data lake architecture specifically includes a data storage layer, a data processing layer, and a data access layer. The description information specifically includes the source, format, and creation time. The financial leasing information specifically includes financial leasing contracts, asset valuation, and risk management.

[0040] The second judgment module is used to determine whether the financial lease information meets the preset completeness requirements;

[0041] The second execution module is used to identify the user's cross-industry asset data based on the financing lease information if the conditions are met, input the financing lease information and the cross-industry asset data into a pre-trained risk dimension data model, update the user's cross-industry collaborative decision-making mechanism, and generate financing lease approval information corresponding to the business data requirements through the cross-industry collaborative decision-making mechanism. The cross-industry asset data specifically includes asset scope, customer groups, and market popularity.

[0042] Furthermore, it also includes:

[0043] The identification module is used to identify the industry type corresponding to the business data requirement based on the source data pre-collected in the data lake, wherein the source data specifically includes structured data, semi-structured data and unstructured data;

[0044] The third judgment module is used to determine whether the industry type covers the source data;

[0045] The third execution module is used to, if not, obtain the data source of the business data requirement according to the industry type, extract the corresponding business data from the data source, convert the business data into the data format preset by the data lake, and periodically update the new data of the business data in the data lake within a preset time period. The data source specifically includes production equipment sensor data, inventory management data, and supply chain data.

[0046] Furthermore, the execution module also includes:

[0047] The identification unit is used to identify key fields in the data fields based on the preset data fields of the financial leasing information, wherein the key fields specifically include contract number, lease amount, lease term, customer ID and asset number;

[0048] The judgment unit is used to determine whether the key field matches a preset unified format;

[0049] An execution unit is configured to, if so, associate the financial leasing information with the user using a preset identifier based on the key field, establish index information of the financial leasing information in the data lake, and grant the user access rights to the index information.

[0050] This invention provides a cross-industry asset management method and system for financial leasing, which has the following beneficial effects:

[0051] This invention ensures that the integration of cross-industry data has a clear structure and unified standards by generating asset data standards and their descriptive information. This is crucial for subsequent data analysis, risk assessment, and decision support. At the same time, by analyzing financial leasing information, it identifies relevant cross-industry asset data and integrates it into the decision-making process, breaking down barriers between industries and allowing more comprehensive asset data to be used for evaluation. Furthermore, through a cross-industry collaborative decision-making mechanism, it can combine data and insights from multiple fields to generate more valuable decision information. This mechanism not only improves the efficiency of financial leasing approval but also effectively reduces human intervention and decision delays. Attached Figure Description

[0052] Figure 1This is a flowchart illustrating one embodiment of the cross-industry asset management method for financial leasing according to the present invention.

[0053] Figure 2 This is a structural block diagram of an embodiment of the cross-industry asset management system for financial leasing of the present invention. Detailed Implementation

[0054] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. The realization of the purpose, functional features, and advantages of the invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings.

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] Reference Appendix Figure 1 The cross-industry asset management method for financial leasing, as described in one embodiment of the present invention, includes:

[0057] S: Based on the pre-input business data requirements, the business data requirements are entered into a preset data lake, and the data lake generates the data type corresponding to the business data requirements. Specifically, the data type includes text, image, video and log file.

[0058] S2: Determine whether the data type is compatible with the preset data lake platform;

[0059] S3: If possible, then according to the preset data lake architecture, the business data requirements are organized in the data lake using a preset layered structure, asset data standards corresponding to the data types are constructed, description information of the asset data standards is generated, the business data requirements are migrated from the preset data center to the data lake, and the financial leasing information of the business data requirements is divided. The data lake architecture specifically includes a data storage layer, a data processing layer, and a data access layer. The description information specifically includes the source, format, and creation time. The financial leasing information specifically includes financial leasing contracts, asset valuation, and risk management.

[0060] S4: Determine whether the finance lease information meets the preset completeness requirements;

[0061] S5: If the conditions are met, then based on the financial leasing information, identify the user's cross-industry asset data, input the financial leasing information and the cross-industry asset data into the pre-trained risk dimension data model, update the user's cross-industry collaborative decision-making mechanism, and generate financial leasing approval information corresponding to the business data requirements through the cross-industry collaborative decision-making mechanism. The cross-industry asset data specifically includes asset scope, customer groups and market popularity.

[0062] In this embodiment, the system inputs pre-entered business data requirements into a pre-defined data lake based on these requirements. The data lake then generates data types corresponding to these requirements, including asset text, asset images, asset videos, and asset log files. The system then determines whether these data types are compatible with the pre-defined data lake platform and executes corresponding steps accordingly. For example, if the system determines that the data types corresponding to the business data requirements are incompatible with the pre-defined data lake platform, it assumes that the data types may lack necessary tags or metadata, leading to ineffective indexing or classification. The system then performs data cleaning, tag supplementation, and metadata addition to ensure that the data structure meets the requirements of the data lake platform. For files lacking tags, the system uses automated tools or manual intervention to add necessary tags and metadata so that the data lake can effectively identify and process them. Finally, this data is stored in an external database or cloud platform. The system then uses indexing or association mechanisms to enable the data lake to access and call the data, thus avoiding direct modifications to the data lake structure and ensuring data integrity and accessibility. For example, when the system determines that the data type corresponding to the business data requirement is compatible with the pre-defined data lake platform, the system will consider that the data type can be effectively indexed or classified in the data lake. Based on the pre-defined data lake architecture, which specifically includes a data storage layer, a data processing layer, and a data access layer, the system will organize the business data requirements in the data lake using a pre-defined layered structure, construct asset data standards corresponding to the data types, and generate descriptive information for the asset data standards, including source, format, and creation time. The system will then migrate these business data requirements from the pre-defined data center to the data lake and classify the business data requirements into financial leasing information, which specifically includes financial leasing contracts, asset valuation, and risk management.The system manages business data using a layered architecture of a data lake (data storage layer, data processing layer, and data access layer). This allows different types of data to be processed and optimized at each layer. The data storage layer centrally manages data storage and backup, the data processing layer is responsible for data cleaning and analysis, and the data access layer supports the data access needs of different users. It also establishes asset data standards corresponding to different data types, standardizing data structure and attributes through unified descriptive information to ensure data consistency within the data lake. This standardization ensures data quality, enabling cross-industry data to be stored and analyzed in a unified format, providing a reliable data foundation for subsequent risk assessment and decision support. Furthermore, through effective indexing and classification of data types, the data lake can quickly find data that meets specific query conditions, allowing the system to query, classify, and filter data more flexibly. Especially when cross-industry data integration and analysis are required, it can quickly obtain relevant information and achieve data correlation, greatly improving data access efficiency. Dividing business data into specific categories such as financial leasing contracts, asset valuation, and risk management facilitates refined management of financial leasing business data, making the data more aligned with business needs. This helps the system conduct more accurate risk analysis, customer assessment, and... Asset management; then the system determines whether this financing lease information meets the pre-set completeness requirements to execute corresponding steps; for example, when the system determines that the financing lease information required by the business data does not meet the pre-set completeness requirements, the system will consider the information to be incomplete or inaccurate. The system will format the data, unify the date, currency, and text encoding formats, so that all financing lease information conforms to the preset format standards. At the same time, metadata such as source and creation time will be added to the incomplete data to ensure the traceability and reliability of the data in subsequent processing, and data that does not meet the completeness requirements will be marked so that it can be identified and excluded in the risk assessment process to avoid negative impacts on the overall assessment results; for example, when the system determines that the financing lease information required by the business data meets the pre-set completeness requirements, the system will consider that the information does not have the problem of data incompleteness. The system will identify the user's cross-industry asset data based on the financing lease information. Cross-industry asset data specifically includes asset scope, customer groups, and market popularity. This financing lease information and cross-industry asset data will be input into a pre-trained risk dimension data model to update the user's cross-industry collaborative decision-making mechanism, and generate financing lease approval information corresponding to the business data requirements through the cross-industry collaborative decision-making mechanism.The system combines financial leasing information with cross-industry asset data, such as asset scope, customer groups, and market trends, providing richer references for risk assessment. Compared to single-dimensional data analysis, comprehensive multi-dimensional information helps the model more comprehensively identify potential risks, especially in complex market environments where this multi-dimensional data integration is particularly important. Furthermore, by updating the user's cross-industry collaborative decision-making mechanism, the system can collaboratively utilize information, analysis, and decision-making in multi-departmental and cross-industry collaborations, providing management with more intelligent decision support and improving the efficiency and accuracy of financial leasing approvals. Based on intelligent risk models and cross-industry collaborative mechanisms, the system can quickly analyze business data requirements and generate financial leasing approval information, shortening approval processes and response times, helping companies quickly respond to market changes and improving the market responsiveness of financial leasing business.

[0063] It should be noted that, based on the aforementioned financial leasing information, the user's cross-industry asset data is identified. The financial leasing information and the cross-industry asset data are then input into a pre-trained risk dimension data model to update the user's cross-industry collaborative decision-making mechanism. This mechanism then generates financial leasing approval information corresponding to the business data requirements. A specific example is as follows:

[0064] Assume the scenario involves Company X, a large manufacturing enterprise, which applies for a financial lease of a batch of high-value heavy equipment; the specific process is as follows:

[0065] First, the system identifies cross-industry asset data. It reads Company X's finance lease application to obtain basic information, including the application amount, lease term, and equipment type. Simultaneously, the system uses a data lake to identify relevant cross-industry asset data for Company X. This data includes not only equipment leasing data from the manufacturing sector but also multi-dimensional data such as Company X's financial health, credit score, and market activity within the industry. The system finds that Company X primarily operates in the manufacturing sector and has some global business coverage. By reading Company X's financial statements and third-party credit scores, the system understands that Company X's credit score is above average within the industry. Furthermore, the system observes a recent increase in demand for heavy equipment in the manufacturing sector, but also high maintenance costs and fluctuating average profit margins in the industry. This information, as part of the asset data, will be used in subsequent risk assessments.

[0066] Then, the data is input into the risk dimension data model. After obtaining complete financial leasing information and cross-industry asset data, the system inputs all relevant data into the pre-trained risk dimension data model for evaluation. The model analyzes multiple risk dimensions, such as the customer's financial health, past lease performance records, and industry market risks, combined with external factors, such as global market conditions and industry trends. The model analysis shows that Company X's financial situation is relatively stable, its past leasing record is good, and its performance is reliable. The industry dimension shows that the current demand for heavy equipment is high, but prices fluctuate greatly and are significantly affected by economic cycles. Based on these factors, the model determines that Company X is in the "medium risk" category, but has high performance potential. At this point, the system may mark the lease as "controllable risk" and suggest adding some risk mitigation measures.

[0067] Then, the cross-industry collaborative decision-making mechanism is updated. After the model completes the risk level assessment, the system automatically updates the cross-industry collaborative decision-making mechanism, synchronizing the assessment results with relevant departments (such as finance, risk control, and market analysis), and integrating cross-industry data to provide these departments with references so that they can make more comprehensive suggestions for financing decisions based on their own responsibilities. In this case, after analyzing the financial data of Company X, the finance department believes that its creditworthiness is good, but attention should be paid to the risk of equipment price fluctuations, and suggests shortening the lease period or increasing the lease interest rate. The risk control department suggests adding a regular inspection clause to reduce the risk of fluctuations in equipment maintenance costs during the lease period. The market analysis department, through industry data and market demand, judges that the lease demand for the equipment has seasonal fluctuations, and therefore suggests dynamically adjusting the lease amount according to the market conditions during the lease period.

[0068] Finally, the system generates financial lease approval information by integrating the assessments and suggestions from various departments and incorporating it into business decisions. This approval information details the approved lease amount, lease term, risk control measures, etc., to ensure that financial lease operations can proceed smoothly while controlling risks.

[0069] Ultimately, the approval information generated by the system includes the following:

[0070] Approved lease amount: 80% of the equipment value, i.e., the financing amount in line with the market average;

[0071] Lease term: One year, with flexible renewal terms;

[0072] Lease interest rate: Increased by 3% on top of the standard rate to mitigate the risk of equipment price fluctuations;

[0073] Risk control clauses: Includes quarterly inspections, equipment maintenance costs during the lease term are borne by the lessee, and the lease contract can be terminated early if market demand fluctuates significantly; this approval information will be directly used for customer communication and contract generation on the business side, helping business personnel to provide customer X company with a financing lease plan that meets its needs while taking into account risk management.

[0074] In summary, through the aforementioned processes and cross-industry data collaboration, the system-generated financing lease solutions can comprehensively assess risks, rationally allocate resources, and ensure the accuracy of business decisions. In the case of Company X, the system combined financial, market, and risk information to provide a flexible solution with risk control clauses, thereby improving the security of the financing lease business and customer satisfaction.

[0075] In this embodiment, before step S3, which involves organizing the business data requirements in the data lake according to a preset data lake architecture and a preset hierarchical structure, the method further includes:

[0076] S301: Based on the source data pre-collected by the data lake, identify the industry type corresponding to the business data requirement, wherein the source data specifically includes structured data, semi-structured data and unstructured data;

[0077] S302: Determine whether the industry type covers the source data;

[0078] S303: If not, then according to the industry type, obtain the data source of the business data requirement, extract the corresponding business data from the data source, convert the business data into the preset data format of the data lake, and periodically update the new data of the business data in the data lake within a preset time period. The data source specifically includes production equipment sensor data, inventory management data, and supply chain data.

[0079] In this embodiment, the system identifies the industry type corresponding to the business data requirement based on the pre-collected source data in the data lake. This source data specifically includes structured, semi-structured, and unstructured data. The system then determines whether these industry types cover the source data and executes corresponding steps accordingly. For example, when the system determines that the industry type corresponding to the business data requirement can cover the source data, it assumes that a comprehensive data source related to that industry can be found in the data lake. The system automatically retrieves the source data related to the business requirement from the data lake, classifies it as structured, semi-structured, or unstructured data, and performs preliminary data integration, including sorting the source data according to the business requirement type. The data is filtered, for example, by selecting financial data, customer profiles, and market dynamics that match the industry type. Simultaneously, using the data lake's layered structure, the extracted source data is categorized into appropriate levels based on data type and content, stored in the data storage layer, data processing layer, and data access layer to support subsequent analysis and access operations. The pre-processed source data is then input into the corresponding analytical model. Combined with the specific objectives of the business data requirements, multi-dimensional industry data is analyzed, including market trend forecasting, customer demand assessment, and financial risk analysis, to provide reliable data support for business decisions. For example, if the system determines that the industry type corresponding to the business data requirement cannot be covered by the source data, then... If the system determines that a comprehensive data source relevant to the industry cannot be found in the data lake, it will acquire data sources based on the industry type to meet business data needs. These data sources specifically include production equipment sensor data, inventory management data, and supply chain data. The system will then extract corresponding business data from these sources, convert this business data into a pre-defined data format for the data lake, and periodically update the data lake with new business data within a pre-defined timeframe. By extracting data from external data sources such as production equipment sensor data, inventory management data, and supply chain data, the system can expand the coverage of the data lake, enabling it to encompass data needs from more industry types. This way, even if there is no direct data source in the data lake... By connecting the data, the system can flexibly respond to various business needs, improving the adaptability and scalability of the data lake. At the same time, after external data is converted into the data lake's preset format, it is managed uniformly with the existing data in the data lake. This standardization of the format allows data from different sources to be accessed and analyzed on a unified platform, ensuring data consistency and ease of access. This lays the foundation for cross-industry analysis and subsequent integration. Furthermore, business data is regularly synchronized to the data lake, keeping the data up-to-date and reflecting changes in the external environment in a timely manner. This avoids analytical errors caused by data lag, ensuring that risk assessment and business decisions are based on the latest information, thereby improving the accuracy of risk assessment and the timeliness of decision-making.

[0080] In this embodiment, step S3, which involves migrating the business data requirements from the preset data center to the data lake and segmenting the financial leasing information of the business data requirements, further includes:

[0081] S31: Based on the preset data fields of the financial leasing information, identify the key fields in the data fields, wherein the key fields specifically include contract number, lease amount, lease term, customer ID and asset number;

[0082] S32: Determine whether the key field matches the preset unified format;

[0083] S33: If so, then based on the key field, apply a preset identifier to associate the financial leasing information with the user, establish index information of the financial leasing information in the data lake, and grant the user access rights to the index information.

[0084] In this embodiment, the system identifies key fields in the pre-defined data fields of the financial leasing information. These key fields include contract number, lease amount, lease term, customer ID, and asset number. The system then determines whether these key fields match a pre-defined uniform format to execute corresponding steps. For example, if the system determines that a key field in the data does not match the pre-defined uniform format, it considers the data format inconsistent or non-compliant, potentially leading to ineffective data processing or analysis. The system provides detailed error feedback, indicating which key fields do not conform to the preset standards. For example, the contract number field may contain non-numeric characters, the lease amount field may contain illegal symbols, and the customer ID field may be empty or formatted incorrectly. The system also records which data fields have undergone format correction and provides corresponding traceability functions to ensure the data processing process is transparent and auditable. This not only aids in subsequent error investigation but also ensures that critical information is not lost during processing. Furthermore, in subsequent data entry and processing stages, the system can automatically detect and prevent any non-compliant data from entering the system by setting stricter format validation rules, improving data quality from the source. For example, if the system determines that a key field in the data matches a pre-defined uniform format... The system adopts a unified data format, recognizing the data as consistent and capable of effective processing and analysis. Based on key fields and pre-defined identifiers, the system associates financial leasing information with user data, establishing an index of financial leasing information within the data lake and granting users access to this index. By creating an index based on key fields, the system enables more efficient data retrieval and management. Each piece of financial leasing information is linked to related user data through the index, significantly improving data search and access speed, especially with large datasets, drastically reducing retrieval time. Furthermore, when granting users access to the index, the system controls access permissions, ensuring only authorized users can access relevant data, enhancing data security and enabling fine-grained access control to prevent unauthorized access or data leakage. The data association and index construction strengthen the mining of relationships between different data sets, allowing the system to extract valuable information from various data sources, optimizing the accuracy and timeliness of business decisions. The establishment of data indexes and user permissions facilitates cross-departmental or cross-industry data sharing and collaboration, enabling different departments to work collaboratively on a unified data platform and quickly locate required information through the index, thereby improving work efficiency.

[0085] In this embodiment, after step S5 of inputting the financial lease information and the cross-industry asset data into the pre-trained risk dimension data model, the method further includes:

[0086] S501: Based on preset risk factors, corresponding risk levels are divided, and a risk dimension data model is constructed according to the risk levels. Specifically, the risk factors include asset risk, customer credit risk and market risk, and the risk levels include customer layer, contract layer, asset layer and market layer.

[0087] S502: Determine whether the risk dimension data model can generate risk assessment results;

[0088] S503: If not, obtain the error log of the risk dimension data model during runtime, apply the preset outlier filter to identify the extreme values ​​of the cross-industry asset data from the error log, and restrict the user from conducting risk assessment based on the extreme values. Specifically, the extreme values ​​include extremely large lease amounts, excessively low customer credit scores, and excessively long lease terms.

[0089] In this embodiment, the system divides risk factors into corresponding risk levels based on pre-defined risk factors. These risk factors specifically include asset risk, customer credit risk, and market risk. The risk levels specifically include customer level, contract level, asset level, and market level. A risk dimension data model is constructed based on these risk levels. The system then determines whether this risk dimension data model can generate risk assessment results to execute corresponding steps. For example, when the system determines that the risk dimension data model can generate risk assessment results, it considers the model to have the ability to generate effective and reliable risk assessment results, accurately assessing various risk factors. The system will then conduct a comprehensive analysis of the business based on the pre-defined risk factors and corresponding risk levels. The risk assessment will integrate these risk factors and levels into the model to form a comprehensive risk assessment report. This report will include assessment results from different risk dimensions. Once the risk assessment results are generated, the system will promote cross-departmental collaboration, particularly among business, risk control, and finance departments. By sharing the risk assessment report, each party can conduct more targeted work based on the assessment results. For example, the finance department can adjust financing amounts based on the risk assessment results, the risk department can adjust risk control strategies, and the business department can optimize customer communication. Furthermore, the system can adjust the parameters of the risk dimension data model in real time according to market changes, asset value fluctuations, or changes in customer behavior, ensuring that the assessment results remain synchronized with the external environment. When the system determines that the risk dimension data model cannot generate a risk assessment result, it considers the user's financing lease information and cross-industry asset data to be abnormal. The system will retrieve the error logs of the risk dimension data model during runtime and apply pre-set outlier filters to identify extreme values ​​in the cross-industry asset data. These extreme values ​​include extremely high lease amounts, excessively low customer credit scores, and excessively long lease terms. Based on these extreme values, the system restricts the user's risk assessment. By retrieving the error logs of the risk dimension data model during runtime, the system can quickly locate and identify anomalies in the data processing process, enabling it to accurately identify which data items or datasets have problems, thereby effectively reducing [the risk]. By assessing erroneous or invalid outputs and introducing an outlier filtering mechanism, the system's fault tolerance is enhanced. When abnormal data occurs, the system can identify and correct it in a timely manner, ensuring that its operation will not fail due to data problems. This guarantees that the system can still output effective risk assessment results stably and efficiently when processing large-scale data. Furthermore, by setting different limits for different extreme values, the system can flexibly adjust risk management strategies. For example, for extremely large lease amounts, the system may automatically set stricter approval processes, while for customers with excessively low credit scores, it may require additional guarantees or other risk mitigation measures. This flexibility enhances the accuracy and timeliness of risk control.

[0090] It should be noted that the application uses a preset outlier filter to identify extreme values ​​in the cross-industry asset data from the error logs, and restricts the user from conducting risk assessments based on these extreme values. A specific example is as follows:

[0091] Suppose this financial leasing company is conducting a cross-industry asset financial leasing risk assessment and has already collected relevant data on customers and assets from the healthcare, manufacturing, and energy industries. This data includes asset value, lease term, customer credit score, etc. The company is using an advanced risk assessment model that relies on this data to make reasonable decisions. However, the data may contain some extreme values, which, if not processed, may lead to incorrect risk assessment results.

[0092] 1. Healthcare industry:

[0093] Lease amount: 5 million yuan;

[0094] Lease term: 5 years;

[0095] Customer credit score: 750 (Good);

[0096] 2. Manufacturing:

[0097] Lease amount: 20 million yuan;

[0098] Lease term: 8 years;

[0099] Customer credit score: 680 (Medium);

[0100] 3. Energy industry:

[0101] Lease amount: 100 million yuan;

[0102] Lease term: 15 years;

[0103] Customer credit score: 450 (low);

[0104] Among these data, the third customer information from the energy industry is clearly problematic; the customer's credit score is 450, far below the minimum credit score requirement (600) stipulated by the financial leasing company, and the lease amount is as high as 100 million yuan, which significantly exceeds the range of amounts that the company usually handles (usually not exceeding 50 million yuan); these data items may cause incorrect predictions in the risk assessment model, because a low credit score and a very high lease amount are likely to mean high risk.

[0105] When the system is running, it will generate an error log according to preset rules, recording the following content:

[0106] Excessive lease amount: The lease amount of 100 million yuan exceeded the scope of the company's normal business operations;

[0107] Credit score too low: Your credit score of 450 is far below the minimum credit score required by the company.

[0108] Therefore, these error messages will be logged in the system's log file;

[0109] The outlier filtering mechanism first scans the data according to preset outlier standards. For lease amounts and credit scores, it compares these values ​​with the set ranges: lease amounts exceeding the upper limit of 50 million yuan and customer credit scores below the minimum standard of 600. Therefore, this lease data from the energy industry will be marked as an "extreme value" by the system.

[0110] For handling extreme values, once the system identifies an extreme value, it will immediately take the following steps to restrict the data from entering the risk assessment model: To avoid incorrectly inputting this abnormal data into the risk assessment model, the system will refuse to continue processing the data; the system will prompt business personnel, indicating that the data has a problem and cannot enter the assessment process; for such problems, the system will suggest that business personnel contact the customer to correct their credit score or add more collateral to reduce risk; for example, the system can ask the customer to provide additional financial statements, cash flow proof, or more assets as collateral to mitigate the risk of a low credit score; and if the lease amount is too high, the system will suggest that business personnel consider adjusting the lease amount to a reasonable range or ask the customer to provide more explanations of the source of funds.

[0111] Dynamic risk assessment adjustments: If business personnel correct the data, the system will update the data in real time and re-enter it into the risk assessment model. The system will recalculate the risk assessment based on the new data (such as an updated credit score or an adjusted lease amount). The system generates a new risk assessment result for each customer based on different industry types, customer credit, and other information. For example, after correction, if a customer's credit score increases to 600 and the lease amount decreases to 40 million yuan, the system will reassess the customer's risk and provide a more reasonable financing plan.

[0112] Continuous monitoring and feedback: The system continuously monitors data sources to ensure data accuracy; when new outliers occur, the system automatically restarts the outlier detection mechanism and provides feedback to users so they can take timely measures; if new data is subsequently available (such as updates to customer credit scores or changes in lease amounts), the system will reassess and update the risk assessment results to ensure that the approval information for financial leases is based on the latest and corrected data.

[0113] In summary, through the aforementioned outlier filtering mechanism, the system can effectively identify and process extreme values ​​in the data, avoiding erroneous risk assessment results. This not only ensures the accuracy of risk assessment but also allows for real-time adjustments to the decision-making process when data anomalies occur, reducing unnecessary risks. For example, if an energy industry customer's credit score remains at 450 and the lease amount is 100 million yuan, without adjustment through the outlier filtering mechanism, the customer might be incorrectly considered low-risk, potentially leading to default risk in the final approved finance lease agreement. If anomalies are identified promptly through a pre-set filtering mechanism, and measures are taken to correct the data or increase collateral, the final financing decision will be more reliable, thereby reducing potential losses. Through this dynamic and automated outlier detection and correction mechanism, finance leasing companies can maintain efficient risk management in cross-industry asset risk assessment, ensuring the rationality and security of the decision-making process.

[0114] In this embodiment, step S5, which updates the user's cross-industry collaborative decision-making mechanism and generates the financing lease approval information corresponding to the business data requirement through the cross-industry collaborative decision-making mechanism, further includes:

[0115] S51: Based on the decision variable information preset by the data center for the business data requirements, the decision variable information is synchronized to the preset industry department, and the approval standard value of the user is established through the industry department. The decision variable information specifically includes customer credit score, asset valuation and contract terms.

[0116] S52: Determine whether the approval standard value meets the preset review benchmark;

[0117] S53: If so, then according to the execution permissions preset by the data center, the user and their financial leasing information are shared to the digital collaboration platform of the industry department, and the financial leasing approval information of the user is constructed in real time on the digital collaboration platform.

[0118] In this embodiment, the system, based on decision variables pre-set by the data center for business data requirements, including customer credit scores, asset valuations, and contract terms, synchronizes these decision variables to pre-defined industry departments. Different industry departments establish approval standard values ​​for each user. The system then determines whether the user's approval standard value meets the pre-set review benchmark and executes the corresponding steps. For example, if the system determines that the user's approval standard value does not meet the pre-set review benchmark, the system considers the user's financial leasing application to have a certain risk and cannot meet the company's financing requirements or credit standards. The system automatically generates an error or warning message, marking that the user's approval standard value has not met the preset benchmark. The feedback clearly indicates which decision variable caused the approval standard value to fail to meet the benchmark. A buffer period is set, allowing the customer to adjust their financial situation, provide supplementary information, or modify contract terms within a specified time to improve the approval standard value. Furthermore, the system regularly optimizes the approval standards based on historical review results, market changes, and customer feedback. Different approval benchmarks can be set for different industries and different customer types, making the approval process more flexible and adaptable to market demands. For example, if the system determines that the user's approval standard value has met the pre-set review benchmark, then... If the system determines that no risks have been detected in a user's financing lease application, it will share the user and their financing lease information with the industry department's digital collaboration platform according to the pre-set execution permissions in the data center. The platform will then generate the user's financing lease approval information in real time. The system can obtain user financing lease information in real time through different industry departments, collaborating on the digital platform to avoid information silos and improve communication efficiency between departments. All relevant departments can view and update information on the same platform, ensuring transparency and consistency in the financing lease approval process, thereby reducing communication errors and information loss. Simultaneously, data judgment based on preset review criteria reduces bias caused by human factors. Sharing confirmed compliant user information with the digital platform ensures that all decisions and approvals are based on the same, unified data, guaranteeing the accuracy of decisions. Furthermore, after sharing financing lease information and generating approval information in real time, the digital platform can collect data feedback to help companies monitor bottlenecks and potential problems in the approval process. The system can further optimize approval standards and decision-making models based on this feedback. By setting standardized approval criteria and data sharing mechanisms, the approval process for all financing lease applications becomes more standardized and consistent, thereby promoting the standardization and continuous optimization of enterprise business processes.

[0119] In this embodiment, step S2, which determines whether the data type is compatible with the preset data lake platform, further includes:

[0120] S21: Obtain the preset time field of the business data requirement;

[0121] S22: Determine whether the time field matches the preset timestamp unit of the data lake platform;

[0122] S23: If not, then unify the time field according to the timestamp unit, and reorder the timestamp partitions of the business data requirements according to the time window of the timestamp unit.

[0123] In this embodiment, the system obtains pre-defined time fields from business data requirements and then determines whether these time fields match the timestamp units pre-defined by the data lake platform to execute corresponding steps. For example, when the system determines that the pre-defined time fields in the business data requirements match the timestamp units pre-defined by the data lake platform, the system considers the time field format in the business data requirements to be consistent with the timestamp units of the data lake platform. This means the system can successfully parse these time data without additional conversion or processing steps. The system will directly input the time field information in the business data requirements into the data lake platform without additional conversion, ensuring the originality and accuracy of the time data. At the same time, a time series index is created based on these time fields. The time series index helps to quickly locate data in a specific time period, improving the efficiency of data query and analysis. Especially when time period comparison, trend analysis, or anomaly detection is required, it can significantly speed up the processing speed. Furthermore, based on the matched timestamps, a synchronization strategy with other data sources is established to ensure that the business data in the data lake is consistent with external data sources. This ensures that all time-related data is kept synchronized and updated in multi-system or cross-departmental collaborative work. For example, when the system determines that the pre-defined time fields in the business data requirements match the timestamp units pre-defined by the data lake platform, the system will execute corresponding steps to ... If a good time field cannot match the timestamp unit pre-defined by the data lake platform, the system will consider the time field format in the business data requirement to be inconsistent with the timestamp unit of the data lake platform, and will be unable to parse this time data. The system will then unify the time field according to the timestamp unit and reorder the timestamp partitions of the business data requirement based on the time windows of these timestamp units. By unifying the time field to the timestamp unit of the data lake platform, the system can ensure that the time field in the business data requirement is consistent with other stored data. This unified format allows data from different data sources to be directly integrated, improving overall data consistency and reducing ambiguity in the analysis and query process. At the same time, after the timestamp unit is consistent and reordered according to the time window, the system can efficiently build a time partition index. This index helps to accelerate data query and retrieval based on the time field, especially in a data lake environment with large data volumes, significantly improving the efficiency of data processing and analysis. Furthermore, the unified timestamp unit makes time series analysis more accurate, avoiding errors caused by inconsistent time fields. For example, in time-based analyses such as trend analysis and anomaly detection, consistent time fields ensure that the data can correctly reflect the correlation between different points in time, enhancing the reliability of the analysis results.

[0124] In this embodiment, step S1, which involves inputting pre-input business data requirements into a preset data lake and generating the data type corresponding to the business data requirements through the data lake, further includes:

[0125] S11: Based on the cleaning measures preset in the data center, detect the items to be cleaned in the business data requirements;

[0126] S12: Determine whether the item to be cleaned matches a preset cleanable item;

[0127] S3: If so, the business data requirements will be pre-cleaned to generate data to be confirmed. The data to be confirmed will be synchronized to the user's preset terminal. The operation permissions for the business data requirements provided by the user will be obtained through the data center. Specifically, the data pre-cleaning includes removing useless fields, processing missing values, and correcting erroneous data.

[0128] In this embodiment, the system detects items to be cleaned in business data requirements based on pre-established cleaning measures in the data center. The system then determines whether these items match pre-defined cleanable items and executes corresponding steps accordingly. For example, if the system determines that a business data requirement cannot match a pre-defined cleanable item, it considers that some data in the requirement has a format or content type that does not conform to cleaning standards. This data may lead to errors or omissions during cleaning and is difficult to standardize. The system marks the unmatched items and records their specific content and source for further analysis or manual processing. This approach prevents data that doesn't meet cleaning standards from entering subsequent processes, thereby improving overall data quality. Furthermore, based on actual business needs, the system can dynamically adjust cleaning rules. For example, if the formats of items to be cleaned differ slightly (such as slightly different date formats), the system can expand or relax the cleaning rules, allowing more formats of items to match the requirements for cleanable items. This allows for more data to be accommodated during automated cleaning. For data items that cannot be automatically cleaned, the system can notify data maintenance personnel for review to confirm whether manual supplementation or transformation is necessary, ensuring data integrity. Especially when important fields are involved, manual intervention can improve the data cleaning process. Accuracy; for example, when the system determines that the items to be cleaned in the business data requirement match the pre-defined cleanable items, the system assumes that these data can be standardized during the cleaning process. The system will then perform pre-cleaning of the business data requirement, specifically including removing useless fields, handling missing values, and correcting erroneous data. This generates data to be confirmed by the user for secondary confirmation. This data is then synchronized to the user's pre-defined terminal, and the system obtains the user's access permissions for the business data requirement through the data center. By removing useless fields, handling missing values, and correcting erroneous data, the system can eliminate interfering information in the early stages of cleaning, ensuring the accuracy and consistency of the business data, and enabling subsequent... Analysis and decision-making are based on more reliable data sources, thereby improving the accuracy of data applications. At the same time, pre-cleaned data to be confirmed is synchronized to the user terminal for secondary confirmation. This approach not only allows users to control the final data quality but also reduces the risk of misoperation due to inconsistent data formats or data anomalies, ensuring data security and the robustness of business decisions. Furthermore, after synchronizing the data to be confirmed and obtaining the user's operation permissions, the system provides the user with the opportunity to directly review and confirm, enabling the user to control the final data status. In different stages of data processing, the user's visibility and control over the data are improved, which helps to enhance the user's sense of participation and trust in data quality.

[0129] Reference Appendix Figure 2 This invention provides a cross-industry asset management system for financial leasing, comprising:

[0130] The generation module 10 is used to input the pre-input business data requirements into a preset data lake based on the pre-input business data requirements, and generate the data type corresponding to the business data requirements through the data lake. The data type specifically includes text, image, video and log file.

[0131] The judgment module 20 is used to determine whether the data type can be adapted to the preset data lake platform;

[0132] The execution module 30 is used, if possible, to organize the business data requirements in the data lake according to the preset data lake architecture and the preset layered structure, construct the asset data standard corresponding to the data type, generate the description information of the asset data standard, migrate the business data requirements from the preset data center to the data lake, and divide the financial leasing information of the business data requirements. The data lake architecture specifically includes a data storage layer, a data processing layer and a data access layer. The description information specifically includes the source, format and creation time. The financial leasing information specifically includes financial leasing contracts, asset valuation and risk management.

[0133] The second judgment module 40 is used to determine whether the financial lease information meets the preset completeness requirement;

[0134] The second execution module 50 is used to identify the user's cross-industry asset data based on the financing lease information if the conditions are met, input the financing lease information and the cross-industry asset data into a pre-trained risk dimension data model, update the user's cross-industry collaborative decision-making mechanism, and generate financing lease approval information corresponding to the business data requirements through the cross-industry collaborative decision-making mechanism. The cross-industry asset data specifically includes asset scope, customer groups, and market popularity.

[0135] In this embodiment, the generation module 10 inputs pre-entered business data requirements into a pre-defined data lake based on these requirements. The data lake then generates data types corresponding to these requirements, including asset text, asset images, asset videos, and asset log files. The judgment module 20 then determines whether these data types are compatible with the pre-defined data lake platform and executes corresponding steps accordingly. For example, if the system determines that the data types corresponding to the business data requirements are incompatible with the pre-defined data lake platform, the system assumes that the data types may lack necessary tags or metadata, leading to ineffective indexing or classification. The system then performs data cleaning, tag supplementation, and metadata addition to ensure that the data structure meets the requirements of the data lake platform. For files lacking tags, the system uses automated tools or manual intervention to add necessary tags and metadata so that the data lake can effectively identify and process them. Finally, these data are stored in an external database or cloud. On the platform, the data lake can access and call the data through indexing or association mechanisms. This avoids direct modification of the data lake structure and ensures data integrity and accessibility. For example, when the system determines that the data type corresponding to the business data requirement can be adapted to the pre-defined data lake platform, the execution module 30 will consider that the data type can be effectively indexed or classified in the data lake. The system will organize the business data requirements in the data lake according to the pre-defined data lake architecture, which specifically includes a data storage layer, a data processing layer, and a data access layer. It will construct asset data standards corresponding to the data type, generate descriptive information for the asset data standards, including source, format, and creation time. It will then migrate these business data requirements from the pre-defined data center to the data lake and classify the financial leasing information of the business data requirements, including financial leasing contracts, asset valuation, and risk management.The system manages business data using a layered architecture of a data lake (data storage layer, data processing layer, and data access layer). This allows different types of data to be processed and optimized at each layer. The data storage layer centrally manages data storage and backup, the data processing layer cleanses and analyzes the data, and the data access layer supports the data access needs of different users. It also establishes asset data standards corresponding to different data types, standardizing data structure and attributes through unified descriptive information to ensure data consistency within the data lake. This standardization ensures data quality, enabling cross-industry data to be stored and analyzed in a unified format, providing a reliable data foundation for subsequent risk assessment and decision support. Furthermore, through effective indexing and classification of data types, the data lake can quickly find data that meets specific query conditions, allowing the system to query, classify, and filter data more flexibly. Especially when cross-industry data integration and analysis are required, it can quickly obtain relevant information and achieve data correlation, greatly improving data access efficiency. Dividing business data into specific categories such as financial leasing contracts, asset valuation, and risk management facilitates refined management of financial leasing business data, making the data more aligned with business needs and helping the system to conduct risk analysis, customer assessment, and asset management more accurately. The second judgment module 40 then determines whether the financing lease information meets the pre-set completeness requirements to execute corresponding steps. For example, if the system determines that the financing lease information required by the business data does not meet the pre-set completeness requirements, the system will consider the information to be incomplete or inaccurate. The system will then format the data, unifying the date, currency, and text encoding formats to ensure that all financing lease information conforms to the preset format standards. Simultaneously, metadata such as source and creation time will be added to incomplete data to ensure traceability and reliability in subsequent processing. Furthermore, data that does not meet the completeness requirements will be marked for identification during subsequent risk assessment. To avoid negative impacts on the overall assessment results, the system excludes or removes this data. For example, when the system determines that the financing lease information required for the business data meets the pre-set completeness requirements, the second execution module 50 will assume that the information does not have a data incompleteness problem. Based on the financing lease information, the system will identify the user's cross-industry asset data, which specifically includes asset scope, customer groups, and market popularity. This financing lease information and cross-industry asset data will be input into a pre-trained risk dimension data model to update the user's cross-industry collaborative decision-making mechanism. Through the cross-industry collaborative decision-making mechanism, financing lease approval information corresponding to the business data requirements will be generated.The system combines financial leasing information with cross-industry asset data, such as asset scope, customer groups, and market trends, providing richer references for risk assessment. Compared to single-dimensional data analysis, comprehensive multi-dimensional information helps the model more comprehensively identify potential risks, especially in complex market environments where this multi-dimensional data integration is particularly important. Furthermore, by updating the user's cross-industry collaborative decision-making mechanism, the system can collaboratively utilize information, analysis, and decision-making in multi-departmental and cross-industry collaborations, providing management with more intelligent decision support and improving the efficiency and accuracy of financial leasing approvals. Based on intelligent risk models and cross-industry collaborative mechanisms, the system can quickly analyze business data requirements and generate financial leasing approval information, shortening approval processes and response times, helping companies quickly respond to market changes and improving the market responsiveness of financial leasing business.

[0136] In this embodiment, it also includes:

[0137] The identification module is used to identify the industry type corresponding to the business data requirement based on the source data pre-collected in the data lake, wherein the source data specifically includes structured data, semi-structured data and unstructured data;

[0138] The third judgment module is used to determine whether the industry type covers the source data;

[0139] The third execution module is used to, if not, obtain the data source of the business data requirement according to the industry type, extract the corresponding business data from the data source, convert the business data into the data format preset by the data lake, and periodically update the new data of the business data in the data lake within a preset time period. The data source specifically includes production equipment sensor data, inventory management data, and supply chain data.

[0140] In this embodiment, the system identifies the industry type corresponding to the business data requirement based on the pre-collected source data in the data lake. This source data specifically includes structured, semi-structured, and unstructured data. The system then determines whether these industry types cover the source data and executes corresponding steps accordingly. For example, when the system determines that the industry type corresponding to the business data requirement can cover the source data, it assumes that a comprehensive data source related to that industry can be found in the data lake. The system automatically retrieves the source data related to the business requirement from the data lake, classifies it as structured, semi-structured, or unstructured data, and performs preliminary data integration, including sorting the source data according to the business requirement type. The data is filtered, for example, by selecting financial data, customer profiles, and market dynamics that match the industry type. Simultaneously, using the data lake's layered structure, the extracted source data is categorized into appropriate levels based on data type and content, stored in the data storage layer, data processing layer, and data access layer to support subsequent analysis and access operations. The pre-processed source data is then input into the corresponding analytical model. Combined with the specific objectives of the business data requirements, multi-dimensional industry data is analyzed, including market trend forecasting, customer demand assessment, and financial risk analysis, to provide reliable data support for business decisions. For example, if the system determines that the industry type corresponding to the business data requirement cannot be covered by the source data, then... If the system determines that a comprehensive data source relevant to the industry cannot be found in the data lake, it will acquire data sources based on the industry type to meet business data needs. These data sources specifically include production equipment sensor data, inventory management data, and supply chain data. The system will then extract corresponding business data from these sources, convert this business data into a pre-defined data format for the data lake, and periodically update the data lake with new business data within a pre-defined timeframe. By extracting data from external data sources such as production equipment sensor data, inventory management data, and supply chain data, the system can expand the coverage of the data lake, enabling it to encompass data needs from more industry types. This way, even if there is no direct data source in the data lake... By connecting the data, the system can flexibly respond to various business needs, improving the adaptability and scalability of the data lake. At the same time, after external data is converted into the data lake's preset format, it is managed uniformly with the existing data in the data lake. This standardization of the format allows data from different sources to be accessed and analyzed on a unified platform, ensuring data consistency and ease of access. This lays the foundation for cross-industry analysis and subsequent integration. Furthermore, business data is regularly synchronized to the data lake, keeping the data up-to-date and reflecting changes in the external environment in a timely manner. This avoids analytical errors caused by data lag, ensuring that risk assessment and business decisions are based on the latest information, thereby improving the accuracy of risk assessment and the timeliness of decision-making.

[0141] In this embodiment, the execution module further includes:

[0142] The identification unit is used to identify key fields in the data fields based on the preset data fields of the financial leasing information, wherein the key fields specifically include contract number, lease amount, lease term, customer ID and asset number;

[0143] The judgment unit is used to determine whether the key field matches a preset unified format;

[0144] An execution unit is configured to, if so, associate the financial leasing information with the user using a preset identifier based on the key field, establish index information of the financial leasing information in the data lake, and grant the user access rights to the index information.

[0145] In this embodiment, the system identifies key fields in the pre-defined data fields of the financial leasing information. These key fields include contract number, lease amount, lease term, customer ID, and asset number. The system then determines whether these key fields match a pre-defined uniform format to execute corresponding steps. For example, if the system determines that a key field in the data does not match the pre-defined uniform format, it considers the data format inconsistent or non-compliant, potentially leading to ineffective data processing or analysis. The system provides detailed error feedback, indicating which key fields do not conform to the preset standards. For example, the contract number field may contain non-numeric characters, the lease amount field may contain illegal symbols, and the customer ID field may be empty or formatted incorrectly. The system also records which data fields have undergone format correction and provides corresponding traceability functions to ensure the data processing process is transparent and auditable. This not only aids in subsequent error investigation but also ensures that critical information is not lost during processing. Furthermore, in subsequent data entry and processing stages, the system can automatically detect and prevent any non-compliant data from entering the system by setting stricter format validation rules, improving data quality from the source. For example, if the system determines that a key field in the data matches a pre-defined uniform format... The system adopts a unified data format, recognizing the data as consistent and capable of effective processing and analysis. Based on key fields and pre-defined identifiers, the system associates financial leasing information with user data, establishing an index of financial leasing information within the data lake and granting users access to this index. By creating an index based on key fields, the system enables more efficient data retrieval and management. Each piece of financial leasing information is linked to related user data through the index, significantly improving data search and access speed, especially with large datasets, drastically reducing retrieval time. Furthermore, when granting users access to the index, the system controls access permissions, ensuring only authorized users can access relevant data, enhancing data security and enabling fine-grained access control to prevent unauthorized access or data leakage. The data association and index construction strengthen the mining of relationships between different data sets, allowing the system to extract valuable information from various data sources, optimizing the accuracy and timeliness of business decisions. The establishment of data indexes and user permissions facilitates cross-departmental or cross-industry data sharing and collaboration, enabling different departments to work collaboratively on a unified data platform and quickly locate required information through the index, thereby improving work efficiency.

[0146] In this embodiment, it also includes:

[0147] The segmentation module is used to segment corresponding risk levels based on preset risk factors and construct a risk dimension data model based on the risk levels. Specifically, the risk factors include asset risk, customer credit risk and market risk, and the risk levels include customer layer, contract layer, asset layer and market layer.

[0148] The fourth judgment module is used to determine whether the risk dimension data model can generate risk assessment results;

[0149] The fourth execution module is used to, if not, obtain the error log of the risk dimension data model during runtime, apply a preset outlier filter to identify extreme values ​​of the cross-industry asset data from the error log, and restrict the user from conducting risk assessment based on the extreme values. Specifically, the extreme values ​​include extremely high lease amounts, excessively low customer credit scores, and excessively long lease terms.

[0150] In this embodiment, the system divides risk factors into corresponding risk levels based on pre-defined risk factors. These risk factors specifically include asset risk, customer credit risk, and market risk. The risk levels specifically include customer level, contract level, asset level, and market level. A risk dimension data model is constructed based on these risk levels. The system then determines whether this risk dimension data model can generate risk assessment results to execute corresponding steps. For example, when the system determines that the risk dimension data model can generate risk assessment results, it considers the model to have the ability to generate effective and reliable risk assessment results, accurately assessing various risk factors. The system will then conduct a comprehensive analysis of the business based on the pre-defined risk factors and corresponding risk levels. The risk assessment will integrate these risk factors and levels into the model to form a comprehensive risk assessment report. This report will include assessment results from different risk dimensions. Once the risk assessment results are generated, the system will promote cross-departmental collaboration, particularly among business, risk control, and finance departments. By sharing the risk assessment report, each party can conduct more targeted work based on the assessment results. For example, the finance department can adjust financing amounts based on the risk assessment results, the risk department can adjust risk control strategies, and the business department can optimize customer communication. Furthermore, the system can adjust the parameters of the risk dimension data model in real time according to market changes, asset value fluctuations, or changes in customer behavior, ensuring that the assessment results remain synchronized with the external environment. When the system determines that the risk dimension data model cannot generate a risk assessment result, it considers the user's financing lease information and cross-industry asset data to be abnormal. The system will retrieve the error logs of the risk dimension data model during runtime and apply pre-set outlier filters to identify extreme values ​​in the cross-industry asset data. These extreme values ​​include extremely high lease amounts, excessively low customer credit scores, and excessively long lease terms. Based on these extreme values, the system restricts the user's risk assessment. By retrieving the error logs of the risk dimension data model during runtime, the system can quickly locate and identify anomalies in the data processing process, enabling it to accurately identify which data items or datasets have problems, thereby effectively reducing [the risk]. By assessing erroneous or invalid outputs and introducing an outlier filtering mechanism, the system's fault tolerance is enhanced. When abnormal data occurs, the system can identify and correct it in a timely manner, ensuring that its operation will not fail due to data problems. This guarantees that the system can still output effective risk assessment results stably and efficiently when processing large-scale data. Furthermore, by setting different limits for different extreme values, the system can flexibly adjust risk management strategies. For example, for extremely large lease amounts, the system may automatically set stricter approval processes, while for customers with excessively low credit scores, it may require additional guarantees or other risk mitigation measures. This flexibility enhances the accuracy and timeliness of risk control.

[0151] In this embodiment, the second execution module further includes:

[0152] The synchronization unit is used to synchronize the decision variable information to the preset industry departments based on the decision variable information preset by the data center for the business data requirements, and to establish the user's approval standard value through the industry departments. The decision variable information specifically includes customer credit score, asset valuation and contract terms.

[0153] The second judgment unit is used to determine whether the approval standard value meets the preset review benchmark;

[0154] The second execution unit is used to, if so, share the user and their financial leasing information to the industry department's digital collaboration platform according to the execution permissions preset by the data center, and to construct the user's financial leasing approval information in real time on the digital collaboration platform.

[0155] In this embodiment, the system, based on decision variables pre-set by the data center for business data requirements, including customer credit scores, asset valuations, and contract terms, synchronizes these decision variables to pre-defined industry departments. Different industry departments establish approval standard values ​​for each user. The system then determines whether the user's approval standard value meets the pre-set review benchmark and executes the corresponding steps. For example, if the system determines that the user's approval standard value does not meet the pre-set review benchmark, the system considers the user's financial leasing application to have a certain risk and cannot meet the company's financing requirements or credit standards. The system automatically generates an error or warning message, marking that the user's approval standard value has not met the preset benchmark. The feedback clearly indicates which decision variable caused the approval standard value to fail to meet the benchmark. A buffer period is set, allowing the customer to adjust their financial situation, provide supplementary information, or modify contract terms within a specified time to improve the approval standard value. Furthermore, the system regularly optimizes the approval standards based on historical review results, market changes, and customer feedback. Different approval benchmarks can be set for different industries and different customer types, making the approval process more flexible and adaptable to market demands. For example, if the system determines that the user's approval standard value has met the pre-set review benchmark, then... If the system determines that no risks have been detected in a user's financing lease application, it will share the user and their financing lease information with the industry department's digital collaboration platform according to the pre-set execution permissions in the data center. The platform will then generate the user's financing lease approval information in real time. The system can obtain user financing lease information in real time through different industry departments, collaborating on the digital platform to avoid information silos and improve communication efficiency between departments. All relevant departments can view and update information on the same platform, ensuring transparency and consistency in the financing lease approval process, thereby reducing communication errors and information loss. Simultaneously, data judgment based on preset review criteria reduces bias caused by human factors. Sharing confirmed compliant user information with the digital platform ensures that all decisions and approvals are based on the same, unified data, guaranteeing the accuracy of decisions. Furthermore, after sharing financing lease information and generating approval information in real time, the digital platform can collect data feedback to help companies monitor bottlenecks and potential problems in the approval process. The system can further optimize approval standards and decision-making models based on this feedback. By setting standardized approval criteria and data sharing mechanisms, the approval process for all financing lease applications becomes more standardized and consistent, thereby promoting the standardization and continuous optimization of enterprise business processes.

[0156] In this embodiment, the determination module further includes:

[0157] The acquisition unit is used to acquire the preset time field of the business data requirement;

[0158] The third judgment unit is used to determine whether the time field matches the preset timestamp unit of the data lake platform;

[0159] The third execution unit is used to, if not, unify the time field according to the timestamp unit, and reorder the timestamp partitions of the business data requirements according to the time window of the timestamp unit.

[0160] In this embodiment, the system obtains pre-defined time fields from business data requirements and then determines whether these time fields match the timestamp units pre-defined by the data lake platform to execute corresponding steps. For example, when the system determines that the pre-defined time fields in the business data requirements match the timestamp units pre-defined by the data lake platform, the system considers the time field format in the business data requirements to be consistent with the timestamp units of the data lake platform. This means the system can successfully parse these time data without additional conversion or processing steps. The system will directly input the time field information in the business data requirements into the data lake platform without additional conversion, ensuring the originality and accuracy of the time data. At the same time, a time series index is created based on these time fields. The time series index helps to quickly locate data in a specific time period, improving the efficiency of data query and analysis. Especially when time period comparison, trend analysis, or anomaly detection is required, it can significantly speed up the processing speed. Furthermore, based on the matched timestamps, a synchronization strategy with other data sources is established to ensure that the business data in the data lake is consistent with external data sources. This ensures that all time-related data is kept synchronized and updated in multi-system or cross-departmental collaborative work. For example, when the system determines that the pre-defined time fields in the business data requirements match the timestamp units pre-defined by the data lake platform, the system will execute corresponding steps to ... If a good time field cannot match the timestamp unit pre-defined by the data lake platform, the system will consider the time field format in the business data requirement to be inconsistent with the timestamp unit of the data lake platform, and will be unable to parse this time data. The system will then unify the time field according to the timestamp unit and reorder the timestamp partitions of the business data requirement based on the time windows of these timestamp units. By unifying the time field to the timestamp unit of the data lake platform, the system can ensure that the time field in the business data requirement is consistent with other stored data. This unified format allows data from different data sources to be directly integrated, improving overall data consistency and reducing ambiguity in the analysis and query process. At the same time, after the timestamp unit is consistent and reordered according to the time window, the system can efficiently build a time partition index. This index helps to accelerate data query and retrieval based on the time field, especially in a data lake environment with large data volumes, significantly improving the efficiency of data processing and analysis. Furthermore, the unified timestamp unit makes time series analysis more accurate, avoiding errors caused by inconsistent time fields. For example, in time-based analyses such as trend analysis and anomaly detection, consistent time fields ensure that the data can correctly reflect the correlation between different points in time, enhancing the reliability of the analysis results.

[0161] In this embodiment, the generation module further includes:

[0162] The detection unit is used to detect the items to be cleaned in the business data requirements based on the cleaning measures preset in the data center.

[0163] The fourth judgment unit is used to determine whether the item to be cleaned matches a preset cleanable item;

[0164] The fourth execution unit is used to perform data pre-cleaning on the business data requirements if the conditions are met, generate data to be confirmed, synchronize the data to be confirmed to the user's preset terminal, and obtain the operation permissions for the business data requirements provided by the user through the data center. Specifically, the data pre-cleaning includes removing useless fields, processing missing values, and correcting erroneous data.

[0165] In this embodiment, the system detects items to be cleaned in business data requirements based on pre-established cleaning measures in the data center. The system then determines whether these items match pre-defined cleanable items and executes corresponding steps accordingly. For example, if the system determines that a business data requirement cannot match a pre-defined cleanable item, it considers that some data in the requirement has a format or content type that does not conform to cleaning standards. This data may lead to errors or omissions during cleaning and is difficult to standardize. The system marks the unmatched items and records their specific content and source for further analysis or manual processing. This approach prevents data that doesn't meet cleaning standards from entering subsequent processes, thereby improving overall data quality. Furthermore, based on actual business needs, the system can dynamically adjust cleaning rules. For example, if the formats of items to be cleaned differ slightly (such as slightly different date formats), the system can expand or relax the cleaning rules, allowing more formats of items to match the requirements for cleanable items. This allows for more data to be accommodated during automated cleaning. For data items that cannot be automatically cleaned, the system can notify data maintenance personnel for review to confirm whether manual supplementation or transformation is necessary, ensuring data integrity. Especially when important fields are involved, manual intervention can improve the data cleaning process. Accuracy; for example, when the system determines that the items to be cleaned in the business data requirement match the pre-defined cleanable items, the system assumes that these data can be standardized during the cleaning process. The system will then perform pre-cleaning of the business data requirement, specifically including removing useless fields, handling missing values, and correcting erroneous data. This generates data to be confirmed by the user for secondary confirmation. This data is then synchronized to the user's pre-defined terminal, and the system obtains the user's access permissions for the business data requirement through the data center. By removing useless fields, handling missing values, and correcting erroneous data, the system can eliminate interfering information in the early stages of cleaning, ensuring the accuracy and consistency of the business data, and enabling subsequent... Analysis and decision-making are based on more reliable data sources, thereby improving the accuracy of data applications. At the same time, pre-cleaned data to be confirmed is synchronized to the user terminal for secondary confirmation. This approach not only allows users to control the final data quality but also reduces the risk of misoperation due to inconsistent data formats or data anomalies, ensuring data security and the robustness of business decisions. Furthermore, after synchronizing the data to be confirmed and obtaining the user's operation permissions, the system provides the user with the opportunity to directly review and confirm, enabling the user to control the final data status. In different stages of data processing, the user's visibility and control over the data are improved, which helps to enhance the user's sense of participation and trust in data quality.

[0166] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A cross-industry asset management method for financial leasing, characterized in that, Includes the following steps: Based on pre-input business data requirements, the business data requirements are entered into a preset data lake, and the data lake generates the data type corresponding to the business data requirements. Specifically, the data type includes text, image, video, and log file. Determine whether the data type is compatible with the preset data lake platform; If possible, based on the preset data lake architecture, the business data requirements are organized in the data lake using a preset layered structure, asset data standards corresponding to the data types are constructed, descriptive information of the asset data standards is generated, the business data requirements are migrated from the preset data center to the data lake, and the financial leasing information of the business data requirements is divided. Specifically, the data lake architecture includes a data storage layer, a data processing layer, and a data access layer. The descriptive information specifically includes the source, format, and creation time. The financial leasing information specifically includes financial leasing contracts, asset valuation, and risk management. Determine whether the finance lease information meets the preset completeness requirements; If the conditions are met, the user's cross-industry asset data is identified based on the financial leasing information. The financial leasing information and the cross-industry asset data are then input into a pre-trained risk dimension data model to update the user's cross-industry collaborative decision-making mechanism. The financial leasing approval information corresponding to the business data requirements is then generated through the cross-industry collaborative decision-making mechanism. Specifically, the cross-industry asset data includes asset scope, customer groups, and market popularity. The step of determining whether the data type is compatible with the preset data lake platform further includes: Obtain the preset time field for the business data requirements; Determine whether the time field matches the preset timestamp unit of the data lake platform; If not, then unify the time field according to the timestamp unit, and reorder the timestamp partitions of the business data requirements according to the time window of the timestamp unit.

2. The cross-industry asset management method for financial leasing according to claim 1, characterized in that, Before the step of organizing the business data requirements in the data lake according to the preset data lake architecture and using a preset layered structure, the method further includes: Based on the source data pre-collected by the data lake, the industry type corresponding to the business data requirement is identified, wherein the source data specifically includes structured data, semi-structured data and unstructured data; Determine whether the industry type covers the source data; If not, then based on the industry type, obtain the data source for the business data requirement, extract the corresponding business data from the data source, convert the business data into the preset data format of the data lake, and periodically update the new data of the business data in the data lake within a preset time period. The data source specifically includes production equipment sensor data, inventory management data, and supply chain data.

3. The cross-industry asset management method for financial leasing according to claim 1, characterized in that, The step of migrating the business data requirements from the preset data center to the data lake and classifying the financial leasing information of the business data requirements further includes: Based on the preset data fields of the financial leasing information, key fields in the data fields are identified, wherein the key fields specifically include contract number, lease amount, lease term, customer ID and asset number; Determine whether the key fields match a preset uniform format; If so, then based on the key fields, a preset identifier is applied to associate the financial leasing information with the user, establish index information of the financial leasing information in the data lake, and grant the user access rights to the index information.

4. The cross-industry asset management method for financial leasing according to claim 1, characterized in that, After the step of inputting the financial leasing information and the cross-industry asset data into the pre-trained risk dimension data model, the method further includes: Based on preset risk factors, corresponding risk levels are divided, and a risk dimension data model is constructed according to the risk levels. Specifically, the risk factors include asset risk, customer credit risk, and market risk, and the risk levels include customer layer, contract layer, asset layer, and market layer. Determine whether the risk dimension data model can generate risk assessment results; If not, obtain the error log of the risk dimension data model during runtime, apply the preset outlier filter to identify extreme values ​​of the cross-industry asset data from the error log, and restrict the user from conducting risk assessment based on the extreme values. Specifically, the extreme values ​​include extremely large lease amounts, excessively low customer credit scores, and excessively long lease terms.

5. The cross-industry asset management method for financial leasing according to claim 1, characterized in that, The step of updating the user's cross-industry collaborative decision-making mechanism and generating the financing lease approval information corresponding to the business data requirement through the cross-industry collaborative decision-making mechanism further includes: Based on the decision variable information preset by the data center for the business data requirements, the decision variable information is synchronized to the preset industry departments, and the approval standard value of the user is established through the industry departments. The decision variable information specifically includes customer credit score, asset valuation and contract terms. Determine whether the approval standard value meets the preset review benchmark; If so, the user and their financial leasing information will be shared with the industry department's digital collaboration platform according to the preset execution permissions of the data center, and the user's financial leasing approval information will be constructed in real time on the digital collaboration platform.

6. The cross-industry asset management method for financial leasing according to claim 1, characterized in that, The step of inputting pre-input business data requirements into a preset data lake and generating the data type corresponding to the business data requirements through the data lake further includes: Based on the pre-set cleaning measures of the data center, detect the items to be cleaned in the business data requirements; Determine whether the item to be cleaned matches a preset cleanable item; If so, the business data requirements will be pre-cleaned to generate data to be confirmed. The data to be confirmed will be synchronized to the user's preset terminal. Through the data center, the operation permissions for the business data requirements provided by the user will be obtained. Specifically, the data pre-cleaning includes removing useless fields, processing missing values, and correcting erroneous data.

7. A cross-industry asset management system for financial leasing, characterized in that, include: The generation module is used to input the pre-input business data requirements into a preset data lake based on the pre-input business data requirements, and generate the data type corresponding to the business data requirements through the data lake. The data type specifically includes text, image, video and log file. The judgment module is used to determine whether the data type can be adapted to the preset data lake platform; The execution module is used, if possible, to organize the business data requirements in the data lake according to the preset data lake architecture and the preset layered structure, construct the asset data standard corresponding to the data type, generate the description information of the asset data standard, migrate the business data requirements from the preset data center to the data lake, and divide the financial leasing information of the business data requirements. The data lake architecture specifically includes a data storage layer, a data processing layer, and a data access layer. The description information specifically includes the source, format, and creation time. The financial leasing information specifically includes financial leasing contracts, asset valuation, and risk management. The second judgment module is used to determine whether the financial lease information meets the preset completeness requirements; The second execution module is used to identify the user's cross-industry asset data based on the financial leasing information if the conditions are met, input the financial leasing information and the cross-industry asset data into a pre-trained risk dimension data model, update the user's cross-industry collaborative decision-making mechanism, and generate financial leasing approval information corresponding to the business data requirements through the cross-industry collaborative decision-making mechanism. The cross-industry asset data specifically includes asset scope, customer groups, and market popularity. The judgment module also includes: The acquisition unit is used to acquire the preset time field of the business data requirement; The third judgment unit is used to determine whether the time field matches the preset timestamp unit of the data lake platform; The third execution unit is used to, if not, unify the time field according to the timestamp unit, and reorder the timestamp partitions of the business data requirements according to the time window of the timestamp unit.

8. The cross-industry asset management system for financial leasing according to claim 7, characterized in that, Also includes: The identification module is used to identify the industry type corresponding to the business data requirement based on the source data pre-collected in the data lake, wherein the source data specifically includes structured data, semi-structured data and unstructured data; The third judgment module is used to determine whether the industry type covers the source data; The third execution module is used to, if not, obtain the data source of the business data requirement according to the industry type, extract the corresponding business data from the data source, convert the business data into the data format preset by the data lake, and periodically update the new data of the business data in the data lake within a preset time period. The data source specifically includes production equipment sensor data, inventory management data, and supply chain data.

9. The cross-industry asset management system for financial leasing according to claim 7, characterized in that, The execution module further includes: The identification unit is used to identify key fields in the data fields based on the preset data fields of the financial leasing information, wherein the key fields specifically include contract number, lease amount, lease term, customer ID and asset number; The judgment unit is used to determine whether the key field matches a preset unified format; An execution unit is configured to, if so, associate the financial leasing information with the user using a preset identifier based on the key field, establish index information of the financial leasing information in the data lake, and grant the user access rights to the index information.

Images