A credit profile assessment method for a shipping company

Abstract

The present invention relates to the field of credit assessment technology, and in particular to a method for assessing the credit portrait of a carrier. The method comprises the following steps: obtaining carrier information data; establishing initial enterprise portrait data based on the carrier information data; analyzing the credit assessment demand characteristics of the carrier based on the initial enterprise portrait data to generate credit assessment demand characteristic data; analyzing the entire process of transportation activities based on the initial enterprise portrait data to generate analysis data for the entire process of transportation activities; performing dynamic carrier credit scoring processing based on the credit assessment demand characteristic data and the analysis data for the entire process of transportation activities to generate dynamic carrier credit scoring data; performing image optimization processing on the initial enterprise portrait data based on the dynamic carrier credit scoring data to generate static enterprise credit portrait data. The present invention improves the accuracy of the credit portrait assessment of the carrier by dynamically assessing the credit information of the carrier.

Description

Technical Field

[0001] The present invention relates to the technical field of credit assessment, and in particular to a credit profile assessment method for a carrier enterprise. Background Art

[0002] In modern logistics and supply chain management, carriers play a vital role. They are not only responsible for the transportation and distribution of goods, but also directly affect the efficiency and cost control of the supply chain. By evaluating the credit profile of carriers, management departments can identify those companies with poor credit records. The credit assessment of carriers has become a key factor in ensuring the stable operation of the supply chain. And with the rapid development of information technology, especially the application of big data and cloud computing, new technical means and data sources have been provided for credit assessment. Through these technologies, a more comprehensive credit assessment of carriers can be achieved. However, the existing credit assessment methods of carriers mainly rely on static data. This evaluation method that relies on historical data not only ignores the recent changes in the business performance of carriers, but also fails to effectively reflect the immediate credit status and potential risks of the company. When the credit assessment of a carrier is completed, the result will not reflect the credit information of the carrier for a long period of time. The lag in this assessment leads to uncertainty and increased risk in the credit assessment of carriers. Summary of the invention

[0003] Based on this, the present invention provides a credit profile assessment method for a carrier to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a credit profile assessment method for a carrier includes the following steps:

[0005] Step S1: Acquire information data of the transport enterprise; establish initial enterprise portrait data based on the information data of the transport enterprise;

[0006] Step S2: analyzing the credit assessment demand characteristics of the carrier based on the initial enterprise portrait data to generate credit assessment demand characteristic data;

[0007] Step S3: Analyze the entire process of the transport activities of the carrier based on the initial enterprise portrait data to generate analysis data of the entire process of the transport activities;

[0008] Step S4: Carrier credit scoring is processed based on the credit assessment demand characteristic data and the whole process analysis data of the transportation activity to generate carrier credit scoring data;

[0009] Step S5: collecting carrier spot check data based on the initial enterprise portrait data to generate carrier spot check data; updating the carrier credit score data in real time based on the carrier spot check data to generate dynamic carrier credit score data;

[0010] Step S6: Perform periodic credit score filling optimization processing on the initial enterprise portrait data based on the dynamic carrier enterprise credit score data to generate static enterprise credit portrait data.

[0011] The present invention collects data on various aspects of carriers, such as company size, service scope, historical transportation records, etc., to form a comprehensive carrier information data as the basis for evaluating corporate credit, helping evaluators to quickly understand the company's operating conditions and historical performance. In-depth analysis of key data in corporate portraits identifies core factors that affect credit scores, such as debt repayment ability, transportation efficiency, and customer satisfaction, and determines which features have a greater impact on credit scores, thereby making credit assessments more accurate and targeted, helping to fine-tune the management of credit assessment processes and improve the effectiveness and practicality of assessments. By specifically analyzing the carrier's transportation activity records (such as cargo types, transportation time, transportation route efficiency, etc.), statistical data on the company's transportation activities are generated, revealing the company's performance and efficiency in actual operations, providing a quantitative basis for evaluating the company's operational capabilities and market adaptability, and better understanding the company's operating characteristics and potential risks, which is extremely critical for assessing its credit rating. By integrating key credit assessment features and actual transport activity data to calculate the credit score, a quantitative and objective credit assessment result is provided. This score reflects the credit status and business risk of the carrier, and is an important decision-making tool for credit institutions, partners and regulators. The generation of credit scores helps market participants to more effectively assess and manage the risks of business dealings with carriers. By regularly or randomly checking the actual operating data of the carrier and updating its credit score, the timeliness and accuracy of the credit score are ensured. Real-time updating of the credit score can quickly reflect changes in the operating status or market environment of the enterprise, and provide the latest risk assessment information to the relevant parties. This dynamic assessment mechanism helps to respond to market and operational risks more agilely and maintain the relevance and effectiveness of credit assessment. By feeding the latest credit score data back to the enterprise portrait, the content and accuracy of the enterprise portrait are optimized. This optimization process makes the enterprise portrait more comprehensive and timely, and ensures the dynamic update and improvement of the enterprise portrait data. By continuously updating the enterprise portrait, the latest changes in the operation and credit status of the enterprise can be captured more accurately, and the forward-looking and early warning functions of the credit assessment can be further enhanced, providing valuable information resources for regulators.

[0012] Preferably, step S1 comprises the following steps:

[0013] Step S11: Obtaining carrier information data;

[0014] Step S12: performing information source identification processing on the carrier enterprise information data to generate information source identification data;

[0015] Step S13: performing carrier enterprise data type classification processing on the carrier enterprise information data according to the information source identification data to obtain classified carrier enterprise information data, wherein the classified carrier enterprise information data includes carrier enterprise information sub-data and carrier enterprise information type;

[0016] Step S14: Establish initial enterprise portrait data based on the divided carrier enterprise information data.

[0017] The present invention obtains the basis of the entire credit portrait evaluation process of the carrier information data. By collecting the basic information, financial data, historical transportation records and other information of the carrier, it provides original data for subsequent in-depth analysis and evaluation, ensuring the premise of accuracy and comprehensiveness of the evaluation. By identifying the source of information (such as official records, third-party evaluations, self-reported data, etc.), it helps the evaluator distinguish the reliability and weight of the data. The identification of the source of information is crucial to the accuracy of subsequent analysis, ensuring the quality of data processing and decision-making. Classification of data improves the efficiency and pertinence of data processing. Classification processing enables subsequent analysis to process different types of data more pertinently, improves the processing efficiency and accuracy of information, and facilitates the implementation of more detailed risk assessment and management. Use the classified and verified data to establish a detailed and specific corporate portrait, including multi-dimensional evaluation and analysis. The established corporate portrait not only reflects the real-time status of the enterprise, but also integrates historical trends and industry comparisons, providing a comprehensive corporate credit and risk assessment tool, so that the system can make more intelligent credit assessment decisions based on these portraits.

[0018] Preferably, step S14 comprises the following steps:

[0019] Design carrier information screening decisions and carrier information conversion decisions based on information source identification data;

[0020] Perform data screening on the carrier information sub-data based on the carrier information screening decision to generate screened carrier information sub-data;

[0021] Based on the carrier information conversion decision, the screening carrier information sub-data is coded and converted to generate coded carrier information sub-data;

[0022] A multi-dimensional matrix node of the carrier is established according to the carrier information type, and the coded carrier information sub-data is mapped to the carrier portrait matrix node to establish the initial enterprise portrait data.

[0023] The present invention optimizes the data processing flow by designing information screening and conversion decisions, ensuring that only the most relevant and reliable data are used for portrait establishment. This screening and conversion strategy helps to improve data quality, ensure the accuracy and efficiency of the evaluation model, and reduce the possibility of errors or deviations. The original data is accurately screened, unnecessary or irrelevant data is removed, the burden of subsequent processing is reduced, and the processing speed and the pertinence of data analysis are improved. Such a screening process ensures the accuracy and relevance of the data and provides support for building an accurate corporate portrait. The screened data is formatted or encoded to make the data more suitable for subsequent analysis and application. The encoding conversion improves the operability and compatibility of the data, making the integration of multi-source data more efficient and convenient for processing on a variety of analysis tools or platforms. By constructing a multidimensional matrix and mapping the data, a structured corporate portrait is formed while maintaining the multidimensional characteristics of the data. This structured portrait makes the data easier to analyze and interpret, supports more complex data associations and pattern recognition, and provides a basis for in-depth data mining and insight. In addition, the establishment of a multidimensional matrix provides a powerful analysis framework for corporate risk assessment and decision support.

[0024] Preferably, step S2 comprises the following steps:

[0025] Step S21: extracting the credit assessment demand data of the carrier enterprise from the initial enterprise portrait data to obtain the credit assessment demand data;

[0026] Step S22: performing credit assessment requirement feature analysis on the credit assessment requirement data based on a preset credit assessment requirement calculation logic to generate credit assessment requirement feature data.

[0027] The present invention ensures that the information on which the credit assessment is based is comprehensive and highly relevant by accurately extracting key data related to the credit assessment in the carrier's profile. Effectively extracting data that affects the company's credit is the key to ensuring the accuracy of the credit assessment and helping analysts or credit assessment systems focus on the most important information. Using specially designed algorithms and computational logic to analyze the extracted data, further mining and determining the key features that affect the carrier's credit score not only increases the accuracy of the credit assessment, but also makes the assessment results more detailed and explanatory, revealing the potential driving factors of the company's credit risk, and providing a scientific basis for subsequent risk management and decision-making.

[0028] Preferably, step S21 includes the following steps:

[0029] The credit assessment demand index data is designed according to the information type of the carrier enterprise, the credit assessment demand index data is generated, and the credit assessment demand data of the carrier enterprise is extracted from the initial enterprise portrait data through the credit assessment demand index data to obtain the credit assessment demand data.

[0030] The present invention identifies and classifies key information related to credit assessment in the carrier enterprise portrait by designing credit assessment demand index data. This index design optimizes the data extraction process, so that data that is critical to credit assessment can be quickly and accurately extracted from the huge enterprise portrait data set. Through the pre-designed index data, the data extraction process is faster and more accurate, reducing the time and resources required for manual screening and analysis. The index ensures the relevance and integrity of the extracted data, avoids missing key data, improves the accuracy of credit assessment, supports automated processing, reduces the possibility of human error, and improves the standardization and consistency of data processing.

[0031] Preferably, step S3 comprises the following steps:

[0032] Step S31: parsing the transport data collection interface of the transport enterprise according to the initial enterprise portrait data, and collecting the transport activity data of the transport enterprise through the transport data collection interface of the transport enterprise to generate transport activity data;

[0033] Step S32: Performing full-process analysis on the transport activity data within a preset full-process analysis time to generate full-process analysis data.

[0034] The present invention can automatically collect the transport activity data of the carrier, such as transport routes, transport volumes, transport times, etc., by automatically parsing and utilizing the data collection interface, thereby improving the efficiency and accuracy of data collection. The latest transport activity data can be automatically acquired in real time or periodically to ensure the timeliness of the data, so that credit assessment can be conducted based on the latest business operations. Automated collection reduces the need for manual input and the corresponding risk of errors, and improves the reliability of the data collection process. By statistically analyzing the transport activity data within a specific time period, key indicators such as the transport efficiency, timeliness, and reliability of the carrier can be quantitatively understood. Statistical processing helps identify trends and patterns in transport activities, such as the number of transport activities, which is very important for evaluating the operational capabilities and risk management of the enterprise. The generated statistical data can support higher-level decision-making.

[0035] Preferably, step S4 comprises the following steps:

[0036] Step S41: setting the credit assessment demand characteristic data and the whole process analysis data of the transportation activity as the credit assessment index data of the carrier enterprise;

[0037] Step S42: performing credit assessment index influencing factor analysis on the carrier enterprise credit assessment index data to generate credit assessment index influencing factor data;

[0038] Step S43: Designing the credit assessment index weights according to the credit assessment index influencing factor data to generate credit assessment index weight data;

[0039] Step S44: designing a credit assessment radial basis function kernel according to the credit assessment index weight data;

[0040] Step S45: establishing a mapping relationship between the carrier credit assessment index and the carrier credit score through the credit assessment radial basis function kernel to generate a carrier credit score model;

[0041] Step S46: Use the carrier credit scoring model to process the carrier credit assessment index data to generate carrier credit scoring data.

[0042] The present invention provides a clear quantitative basis for credit evaluation by clearly defining which data are used as evaluation indicators. Formalizing relevant data as evaluation indicators is to ensure the transparency and consistency of the evaluation process. By identifying and analyzing which factors have a significant impact on credit scoring, it helps to manage the credit evaluation process in a refined manner, provides a basis for adjusting the scoring model and optimizing the evaluation strategy, and makes the scoring more accurate and reflects the real credit risk. Assigning weights to each influencing factor ensures that the evaluation model can fully consider the importance of each factor when quantifying the credit of the enterprise. The key to the design of weights is to balance the influence of each factor so that the scoring result is both comprehensive and fair. Using radial basis function kernels in modeling is an effective technique that can handle nonlinear relationships and enable the model to more accurately map the complex relationship between input data and credit scores. The credit scoring model finally generated can comprehensively consider multiple evaluation indicators, capture the relationship between these indicators and credit scores through advanced algorithms, and provide an accurate and reliable credit score for the carrier. By using the final scoring model to score the enterprise, a quantitative evaluation of the credit status of the carrier is achieved, and the generated credit score data provides key information for risk management and partnership evaluation.

[0043] Preferably, step S44 includes the following steps:

[0044] An initial radial basis function kernel is designed based on a preset support vector machine algorithm; and parameters of the initial radial basis function kernel are adjusted according to credit assessment indicator weight data to generate a credit assessment radial basis function kernel.

[0045] The radial basis function kernel of the present invention is an effective means for processing nonlinear data relationships. The radial basis function kernel is introduced by using the support vector machine algorithm to help the model capture and understand complex data patterns, especially in problems such as credit scoring involving multi-dimensional factors. The support vector machine algorithm is a powerful machine learning tool that improves the generalization ability of the model by maximizing the interval of the decision boundary, so that good prediction accuracy can be maintained on unseen data. The radial basis function kernel can map data from the original space to a higher-dimensional feature space, so that the originally linearly inseparable data can be separated in the new space, thereby improving the model's adaptability to complex data structures. By adjusting the parameters of the radial basis function kernel according to the importance of the credit assessment indicators, it is ensured that the model pays more attention to factors that have a greater impact on credit assessment, which helps to enhance the interpretability and accuracy of the model. Parameter adjustment can be optimized according to actual data and business needs, making the model more flexible, able to adapt to different credit assessment scenarios and demand changes, providing more personalized assessment results, and making the model more sensitive to changes in credit assessment scenarios, so that in the assessment process, it can timely reflect the slight changes in the credit status of the enterprise, and improve the accuracy and timeliness of risk warning.

[0046] Preferably, step S1 comprises the following steps:

[0047] Step S51: collecting carrier spot check data based on the initial enterprise portrait data to obtain carrier spot check data, wherein the carrier spot check data includes carrier application physical quantity data and carrier non-implementation license transportation data;

[0048] Step S52: Carrier spot check anomaly analysis is performed on the carrier spot check data to generate carrier spot check anomaly score data;

[0049] Step S53: updating the carrier credit score data in real time based on the carrier spot check abnormal score data to generate dynamic carrier credit score data.

[0050] The present invention obtains the latest data about the carrier through real-time spot checks, such as the data on the physical quantity applied for and the data on unlicensed transportation, thereby ensuring the timeliness and relevance of the data, and helping the regulatory authorities to promptly understand the actual operation of the company. Regular or irregular spot checks enable the regulatory authorities to effectively monitor the compliance of the carrier, promptly detect violations, and enhance the overall regulatory quality of the industry. Anomaly analysis of the spot check data can effectively identify risk points and potential violations in the company's operations, providing a basis for further regulatory decisions. Abnormal scoring data can further quantify the company's credit assessment and help relevant parties more accurately assess the company's credit status and operational quality. Real-time updates to the carrier's credit score can reflect the company's current credit status, making the credit score more accurate and timely. Updated credit score data can be quickly communicated to decision makers, helping them make faster and more reasonable decisions, especially in aspects such as carrier vehicle supervision.

[0051] Preferably, step S52 includes the following steps:

[0052] Conduct critical value analysis on the physical quantity data applied by the carrier to generate critical value data of the physical quantity applied;

[0053] Perform application physical quantity critical value abnormality scoring processing according to the application physical quantity critical value data to generate application physical quantity critical value abnormality scoring data;

[0054] According to the data of the transport enterprises that have not implemented the permitted transport, the critical value abnormal score processing of the non-implementation of permitted transport is carried out to generate the critical value abnormal score data of the applied physical quantity;

[0055] The data of abnormal score of critical value of physical quantity applied for and the data of abnormal score of critical value of physical quantity applied for are integrated to obtain the abnormal score data of spot check of carriers.

[0056] The present invention accurately identifies possible risks or abnormal behaviors, especially inconsistencies or abnormal high and low values ​​in the application data, by setting critical values ​​and analyzing the physical quantity data of the carrier's application. The analysis of critical values ​​helps the regulatory agency or the enterprise itself to ensure that all operations comply with industry standards and promote business transparency and traceability. The critical value analysis results are converted into specific abnormal scores, providing a quantitative risk assessment tool for enterprises or regulators, further risk management and decision-making. The abnormal score enables the enterprise or regulatory agency to quickly identify and respond to potential risks and take necessary measures in time to prevent the carrier from abnormal vehicles. The critical value abnormal score is used for the behavior of not implementing licensed transportation, monitoring whether the enterprise complies with relevant licenses and regulations, improving the compliance of the enterprise, and effectively preventing and reducing unauthorized transportation activities through strict analysis and scoring of these data, as well as identifying abnormal carriers that have licenses but do not run, protecting the interests of enterprises and the public. By integrating abnormal score data from different fields, a comprehensive risk assessment view is obtained, providing an overall risk status report, and the comprehensive abnormal score data provides strong decision support for regulatory agencies, thereby formulating more effective management strategies and intervention measures.

[0057] The beneficial effects of the present application are that the credit profile assessment method of the carrier of the present invention can effectively reflect the company's immediate credit status and potential risks by integrating dynamic data from multiple sources of the carrier for credit assessment, and reflecting in real time the changes in the carrier's recent business performance and other reasons. Moreover, after the carrier's credit assessment is completed, it can also reflect the latest credit information of the carrier, ensure the certainty of the carrier's credit assessment and reduce risks, and dynamically update the carrier's credit assessment information through the carrier's spot check data, enhance the effectiveness of the credit assessment results of the supervision carrier, solve the problem of lagging credit profile assessment, accurately discover and assess the illegal and irregular transportation behaviors of enterprises, and avoid continuous illegal and irregular behaviors affecting the safety of infrastructure such as highways and bridges. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 A schematic diagram of the steps of a credit profile evaluation method for a carrier enterprise according to the present invention;

[0059] Figure 2 for Figure 1 Detailed implementation steps of step S1 in FIG.

[0060] Figure 3 for Figure 1 Detailed implementation steps of step S4 in FIG.

[0061] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0062] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0063] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0064] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0065] To achieve this, please refer to Figures 1 to 3 The present invention provides a credit profile evaluation method for a carrier, comprising the following steps:

[0066] To achieve the above purpose, a credit profile assessment method for a carrier includes the following steps:

[0067] Step S1: Acquire information data of the transport enterprise; establish initial enterprise portrait data based on the information data of the transport enterprise;

[0068] Step S2: analyzing the credit assessment demand characteristics of the carrier based on the initial enterprise portrait data to generate credit assessment demand characteristic data;

[0069] Step S3: Analyze the entire process of the transport activities of the carrier based on the initial enterprise portrait data to generate analysis data of the entire process of the transport activities;

[0070] Step S4: Carrier credit scoring is processed based on the credit assessment demand characteristic data and the whole process analysis data of the transportation activity to generate carrier credit scoring data;

[0071] Step S5: collecting carrier spot check data based on the initial enterprise portrait data to generate carrier spot check data; updating the carrier credit score data in real time based on the carrier spot check data to generate dynamic carrier credit score data;

[0072] Step S6: Perform periodic credit score filling optimization processing on the initial enterprise portrait data based on the dynamic carrier enterprise credit score data to generate static enterprise credit portrait data.

[0073] The present invention collects data on various aspects of carriers, such as company size, service scope, historical transportation records, etc., to form a comprehensive carrier information data as the basis for evaluating corporate credit, helping evaluators to quickly understand the company's operating conditions and historical performance. In-depth analysis of key data in corporate portraits identifies core factors that affect credit scores, such as debt repayment ability, transportation efficiency, and customer satisfaction, and determines which features have a greater impact on credit scores, thereby making credit assessments more accurate and targeted, helping to fine-tune the management of credit assessment processes and improve the effectiveness and practicality of assessments. By specifically analyzing the carrier's transportation activity records (such as cargo types, transportation time, transportation route efficiency, etc.), statistical data on the company's transportation activities are generated, revealing the company's performance and efficiency in actual operations, providing a quantitative basis for evaluating the company's operational capabilities and market adaptability, and better understanding the company's operating characteristics and potential risks, which is extremely critical for assessing its credit rating. By integrating key credit assessment features and actual transport activity data to calculate the credit score, a quantitative and objective credit assessment result is provided. This score reflects the credit status and business risk of the carrier, and is an important decision-making tool for credit institutions, partners and regulators. The generation of credit scores helps market participants to more effectively assess and manage the risks of business dealings with carriers. By regularly or randomly checking the actual operating data of the carrier and updating its credit score, the timeliness and accuracy of the credit score are ensured. Real-time updating of the credit score can quickly reflect changes in the operating status or market environment of the enterprise, and provide the latest risk assessment information to the relevant parties. This dynamic assessment mechanism helps to respond to market and operational risks more agilely and maintain the relevance and effectiveness of credit assessment. By feeding the latest credit score data back to the enterprise portrait, the content and accuracy of the enterprise portrait are optimized. This optimization process makes the enterprise portrait more comprehensive and timely, and ensures the dynamic update and improvement of the enterprise portrait data. By continuously updating the enterprise portrait, the latest changes in the operation and credit status of the enterprise can be captured more accurately, and the forward-looking and early warning functions of the credit assessment can be further enhanced, providing valuable information resources for regulators.

[0074] As an embodiment of the present invention, refer to Figure 1The above is a schematic diagram of the steps of a credit profile evaluation method for a carrier enterprise of the present invention. In this embodiment, the credit profile evaluation method for a carrier enterprise includes the following steps:

[0075] Step S1: Acquire information data of the transport enterprise; establish initial enterprise portrait data based on the information data of the transport enterprise;

[0076] In the embodiment of the present invention, basic information of the enterprise that has been supervised and entered into the database in advance is obtained, including registration information, main transport activities, key transport vehicles, etc., and third-party data services are used to collect the historical transportation records and credit records of the transport enterprise. The collected data is sorted and stored using a database management system (such as MySQL, MongoDB), and data analysis tools (such as Python's Pandas library) are used to clean and pre-process the data, including removing duplicate records, filling missing values, formatting dates and numbers, etc., integrating classified data, and using professional data analysis and management tools (such as ERP systems, CRM systems, or custom database management systems) to build a preliminary corporate portrait to display key indicators such as business activity trends and historical compliance.

[0077] Step S2: analyzing the credit assessment demand characteristics of the carrier based on the initial enterprise portrait data to generate credit assessment demand characteristic data;

[0078] In the embodiment of the present invention, key indicators that affect credit assessment are defined, such as debt repayment ability, financial stability, transportation capacity, historical default records, etc. Statistical software or scripts (such as R or Python's Scikit-Learn library) are used to conduct in-depth analysis of the portrait data to extract features related to credit assessment. Machine learning algorithms (such as decision trees, random forests, or logistic regression) are used to analyze the correlation between each feature and the credit score, adjust the parameters of the analysis model, ensure the accuracy and generalization ability of the model through cross-validation, and generate credit assessment demand feature data, which will be directly used for the calculation of credit scores.

[0079] Step S3: Analyze the entire process of the transport activities of the carrier based on the initial enterprise portrait data to generate analysis data of the entire process of the transport activities;

[0080] In the embodiment of the present invention, specific data of transportation activities are collected, including transportation times, transportation volume, transportation routes, transportation time, cargo types, accident records, etc. Statistical analysis tools (such as SPSS or Excel) are used to quantitatively analyze transportation data, and statistical indicators such as average transportation time, average transportation times, and on-time delivery rate are calculated. Based on the statistical results, an evaluation report of transportation activities is prepared, and the performance of various indicators is recorded in detail. These data will be directly used for the calculation of credit scores, and graphics and charts (such as bar charts, line charts, heat maps, etc.) can be used later to intuitively display the statistical analysis results of transportation activities for easy understanding and comparison.

[0081] Step S4: Carrier credit scoring is processed based on the credit assessment demand characteristic data and the whole process analysis data of the transportation activity to generate carrier credit scoring data;

[0082] In the embodiment of the present invention, the credit assessment demand feature data and the whole process analysis data of the transportation activity are comprehensively used to select a suitable credit scoring model, such as logistic regression, support vector machine or neural network, train the model according to historical data, and optimize the model parameters to achieve the best prediction accuracy. The currently collected feature data is input into the trained model, the credit score is calculated, the generated credit score data is recorded, and it is stored in the database for subsequent query and analysis.

[0083] Step S5: collecting carrier spot check data based on the initial enterprise portrait data to generate carrier spot check data; updating the carrier credit score data in real time based on the carrier spot check data to generate dynamic carrier credit score data;

[0084] In the embodiment of the present invention, the time, location and frequency of spot checks are planned based on key information in the enterprise profile, such as historical violations, risk indicators, etc. Spot checks are implemented to collect real-time operational data of the carrier, such as GPS tracking data, cargo loading data, transportation license status, etc. The spot check data is analyzed to determine whether there are any violations or abnormal behaviors, the credit score is adjusted according to the spot check results, and the credit score data in the database is updated.

[0085] Step S6: Perform periodic credit score filling optimization processing on the initial enterprise portrait data based on the dynamic carrier enterprise credit score data to generate static enterprise credit portrait data.

[0086] In the embodiment of the present invention, the relevant data fields in the carrier profile, such as credit rating and risk level, are updated according to the dynamic carrier credit score data, and the dynamic carrier credit score data is filled by filling the credit assessment node of the initial enterprise profile data to generate static enterprise credit profile data, which includes the basic information of the carrier and the corresponding credit score. To ensure the consistency and accuracy of the profile data, adjust other relevant data in the profile, such as adjusting the credit warning signal, etc., and use data analysis and visualization tools to re-analyze the updated enterprise profile to ensure that all information is up to date and reflects the current credit status of the enterprise for internal review or external supervision.

[0087] Preferably, step S1 comprises the following steps:

[0088] Step S11: Obtaining carrier information data;

[0089] Step S12: performing information source identification processing on the carrier enterprise information data to generate information source identification data;

[0090] Step S13: performing carrier enterprise data type classification processing on the carrier enterprise information data according to the information source identification data to obtain classified carrier enterprise information data, wherein the classified carrier enterprise information data includes carrier enterprise information sub-data and carrier enterprise information type;

[0091] Step S14: Establish initial enterprise portrait data based on the divided carrier enterprise information data.

[0092] The present invention obtains the basis of the entire credit portrait evaluation process of the carrier information data. By collecting the basic information, financial data, historical transportation records and other information of the carrier, it provides original data for subsequent in-depth analysis and evaluation, ensuring the premise of accuracy and comprehensiveness of the evaluation. By identifying the source of information (such as official records, third-party evaluations, self-reported data, etc.), it helps the evaluator distinguish the reliability and weight of the data. The identification of the source of information is crucial to the accuracy of subsequent analysis, ensuring the quality of data processing and decision-making. Classification of data improves the efficiency and pertinence of data processing. Classification processing enables subsequent analysis to process different types of data more pertinently, improves the processing efficiency and accuracy of information, and facilitates the implementation of more detailed risk assessment and management. Use the classified and verified data to establish a detailed and specific corporate portrait, including multi-dimensional evaluation and analysis. The established corporate portrait not only reflects the real-time status of the enterprise, but also integrates historical trends and industry comparisons, providing a comprehensive corporate credit and risk assessment tool, so that the system can make more intelligent credit assessment decisions based on these portraits.

[0093] As an embodiment of the present invention, refer to Figure 2 As shown, Figure 1 Detailed implementation step flow chart of step S1 in the embodiment, step S1 in this embodiment includes:

[0094] Step S11: Obtaining carrier information data;

[0095] In the embodiment of the present invention, data requirements are determined and the types of data to be collected are listed, such as basic enterprise information, financial data, transportation records, compliance records, etc. Data is collected through multiple channels, including obtaining carrier information data through regulatory enterprise information or third parties, and the collected data is preliminarily sorted, such as unifying the data format, marking the data source, etc., in preparation for subsequent processing.

[0096] Step S12: performing information source identification processing on the carrier enterprise information data to generate information source identification data;

[0097] In an embodiment of the present invention, the source of each data is marked, such as official data, third-party data, self-reported data, etc., so as to evaluate the credibility and weight of the data in subsequent analysis, and use automated tools (such as data labeling software) to assist in identifying and recording the source of data, improve processing efficiency and accuracy, generate an information source identification report, and record the detailed information of the source of each data element in detail to provide a basis for subsequent data processing.

[0098] Step S13: performing carrier enterprise data type classification processing on the carrier enterprise information data according to the information source identification data to obtain classified carrier enterprise information data, wherein the classified carrier enterprise information data includes carrier enterprise information sub-data and carrier enterprise information type;

[0099] In the embodiment of the present invention, data is classified into different types according to the source and nature of the data, such as financial data, carrier status data, carrier compliance data, etc. Data classification algorithms (which may involve machine learning techniques) are applied to automatically classify the data to improve efficiency and consistency, and a classification report is generated. Each type of data is marked to clarify its application and importance in the enterprise portrait.

[0100] Step S14: Establish initial enterprise portrait data based on the divided carrier enterprise information data.

[0101] In an embodiment of the present invention, the classified data is integrated, and a preliminary corporate portrait is constructed using professional data analysis and management tools (such as an ERP system, a CRM system, or a customized database management system). Data analysis techniques (such as statistical analysis and predictive models) are used to further process the data, extract key information and insights, and strengthen the content of the portrait. After the corporate portrait is completed, multiple rounds of verification and adjustment are carried out to ensure the accuracy and practicality of the portrait, ultimately forming a comprehensive and dynamically updated carrier portrait.

[0102] Preferably, step S14 comprises the following steps:

[0103] Design carrier information screening decisions and carrier information conversion decisions based on information source identification data;

[0104] Perform data screening on the carrier information sub-data based on the carrier information screening decision to generate screened carrier information sub-data;

[0105] Based on the carrier information conversion decision, the screening carrier information sub-data is coded and converted to generate coded carrier information sub-data;

[0106] A multi-dimensional matrix node of the carrier is established according to the carrier information type, and the coded carrier information sub-data is mapped to the carrier portrait matrix node to establish the initial enterprise portrait data.

[0107] The present invention optimizes the data processing flow by designing information screening and conversion decisions, ensuring that only the most relevant and reliable data are used for portrait establishment. This screening and conversion strategy helps to improve data quality, ensure the accuracy and efficiency of the evaluation model, and reduce the possibility of errors or deviations. The original data is accurately screened, unnecessary or irrelevant data is removed, the burden of subsequent processing is reduced, and the processing speed and the pertinence of data analysis are improved. Such a screening process ensures the accuracy and relevance of the data and provides support for building an accurate corporate portrait. The screened data is formatted or encoded to make the data more suitable for subsequent analysis and application. The encoding conversion improves the operability and compatibility of the data, making the integration of multi-source data more efficient and convenient for processing on a variety of analysis tools or platforms. By constructing a multidimensional matrix and mapping the data, a structured corporate portrait is formed while maintaining the multidimensional characteristics of the data. This structured portrait makes the data easier to analyze and interpret, supports more complex data associations and pattern recognition, and provides a basis for in-depth data mining and insight. In addition, the establishment of a multidimensional matrix provides a powerful analysis framework for corporate risk assessment and decision support.

[0108] In an embodiment of the present invention, the reliability and relevance of different data sources are analyzed and evaluated according to the information source identification data, which data is most critical to building a corporate portrait, and the carrier information screening decision is designed, such as excluding incomplete, outdated or irrelevant data. According to the information source identification data, the data type that needs to be formatted or encoded is determined, such as converting the text date to a standard date format, or encoding the unstructured text data into an analyzable format, so as to set the carrier information conversion decision, such as using regular expressions to process text, or applying specific data conversion tools. The collected data is screened through automated scripts and carrier information screening decisions, and the data items are checked to meet the preset quality standards and relevance requirements, and the data that does not meet the conditions is removed. The screened data is saved as the screening carrier information sub-data, and the screening process and results are recorded in detail for subsequent review and adjustment. According to the carrier information conversion decision, the screened data is subjected to necessary formatting and encoding processing, such as converting the numerical format, standardizing the text data, etc., and the data conversion tool or programming language (such as Python) is used to perform automated conversion operations. Check that the converted data maintains the integrity and accuracy of the data to ensure that no conversion errors are introduced. The converted data is stored as coded carrier information sub-data for further data processing and analysis. According to the type and structure of the carrier information, a multidimensional data model (such as a star schema or a snowflake schema) is designed to organize the data, and the physical structure of the data model is created using a database management system or a professional data modeling tool. The coded carrier information sub-data is mapped to the corresponding matrix nodes according to the data model to ensure the relevance and consistency of the data during the mapping process, and the mapping rules are adjusted to adapt to the optimization of the data structure. All mapped data nodes are integrated to generate complete initial enterprise portrait data, and the final data review and optimization are performed to ensure the accuracy and practicality of the portrait.

[0109] Preferably, step S2 comprises the following steps:

[0110] Step S21: extracting the credit assessment demand data of the carrier enterprise from the initial enterprise portrait data to obtain the credit assessment demand data;

[0111] Step S22: performing credit assessment requirement feature analysis on the credit assessment requirement data based on a preset credit assessment requirement calculation logic to generate credit assessment requirement feature data.

[0112] The present invention ensures that the information on which the credit assessment is based is comprehensive and highly relevant by accurately extracting key data related to the credit assessment in the carrier's profile. Effectively extracting data that affects the company's credit is the key to ensuring the accuracy of the credit assessment and helping analysts or credit assessment systems focus on the most important information. Using specially designed algorithms and computational logic to analyze the extracted data, further mining and determining the key features that affect the carrier's credit score not only increases the accuracy of the credit assessment, but also makes the assessment results more detailed and explanatory, revealing the potential driving factors of the company's credit risk, and providing a scientific basis for subsequent risk management and decision-making.

[0113] In the embodiment of the present invention, the existing initial enterprise portrait data is reviewed to determine which data fields are necessary for credit assessment, including financial data (such as income, profit, liabilities), operational data (such as transportation volume, default rate), compliance data (such as historical compliance records, legal disputes), etc., and specific standards and conditions for extracting these data are set to ensure the relevance and accuracy of the data. Data that meets the standards are extracted from the data warehouse or database using data query tools (such as SQL queries), data extraction software or programming languages ​​(such as Python scripts), and the extracted data is checked to ensure its completeness and accuracy, and possible data missing or errors are processed. The extracted data is preliminarily processed, such as data cleaning and formatting, for subsequent analysis and use, and the cleaned data is saved as credit assessment demand data, and the source and extraction date of the data are recorded to support future audits and reviews. Based on the goals and industry standards of credit assessment, the calculation logic required for credit assessment is defined, such as carrier demand conversion rate, calculation of debt service ratio, etc., including specifying which indicators have the greatest impact on credit scores, such as debt-to-asset ratio, profit margin, customer satisfaction, etc. Apply the designed credit assessment demand calculation logic, conduct in-depth analysis of the credit assessment demand data, use statistical software or customized analysis tools to perform these calculations, extract or construct key features that affect credit scores from the data, such as financial stability indicators, carrier effectiveness indicators, etc., format and save the analyzed feature data, and generate credit assessment demand feature data.

[0114] Preferably, step S21 includes the following steps:

[0115] The credit assessment demand index data is designed according to the information type of the carrier enterprise, the credit assessment demand index data is generated, and the credit assessment demand data of the carrier enterprise is extracted from the initial enterprise portrait data through the credit assessment demand index data to obtain the credit assessment demand data.

[0116] The present invention identifies and classifies key information related to credit assessment in the carrier enterprise portrait by designing credit assessment demand index data. This index design optimizes the data extraction process, so that data that is critical to credit assessment can be quickly and accurately extracted from the huge enterprise portrait data set. Through the pre-designed index data, the data extraction process is faster and more accurate, reducing the time and resources required for manual screening and analysis. The index ensures the relevance and integrity of the extracted data, avoids missing key data, improves the accuracy of credit assessment, supports automated processing, reduces the possibility of human error, and improves the standardization and consistency of data processing.

[0117] In an embodiment of the present invention, the key information types required for evaluating the credit of a carrier are determined, such as financial data, operational data, compliance history, etc. The potential impact of each data type on the credit score is analyzed for priority processing. A data index structure is designed that can map the data type and the storage location in the database to quickly and accurately access and query the data, and a suitable data modeling tool or software (such as ERD tool, UML modeling) is used to design the logical architecture of the index. Specific rules and algorithms for creating indexes are defined, including data matching, keyword extraction, data classification, etc., to ensure that these rules can support dynamic updates and maintenance to adapt to changes in data structures. Apply the designed rules and algorithms to build a credit assessment demand index in the database system, and use SQL or a specific scripting language (such as Python) to test the index building process to ensure that it can correctly index all key data types. Verify the accuracy and completeness of the index data to ensure that it reflects the latest and most relevant data sets, adjust the structure and rules of the index based on the test results, and optimize performance and response time. Use the designed index data to extract the required credit assessment data from the initial corporate portrait database through database query or specialized data extraction tools, perform necessary processing on the extracted data, such as format conversion, unit unification, etc., to meet the needs of subsequent analysis, store the processed data in an appropriate format and location to ensure its easy access and use, and record detailed logs of data extraction and processing.

[0118] Preferably, step S3 comprises the following steps:

[0119] Step S31: parsing the transport data collection interface of the transport enterprise according to the initial enterprise portrait data, and collecting the transport activity data of the transport enterprise through the transport data collection interface of the transport enterprise to generate transport activity data;

[0120] Step S32: Performing full-process analysis on the transport activity data within a preset full-process analysis time to generate full-process analysis data.

[0121] The present invention can automatically collect the transport activity data of the carrier, such as transport routes, transport volumes, transport times, etc., by automatically parsing and utilizing the data collection interface, thereby improving the efficiency and accuracy of data collection. The latest transport activity data can be automatically acquired in real time or periodically to ensure the timeliness of the data, so that credit assessment can be conducted based on the latest business operations. Automated collection reduces the need for manual input and the corresponding risk of errors, and improves the reliability of the data collection process. By statistically analyzing the transport activity data within a specific time period, key indicators such as the transport efficiency, timeliness, and reliability of the carrier can be quantitatively understood. Statistical processing helps identify trends and patterns in transport activities, such as the number of transport activities, which is very important for evaluating the operational capabilities and risk management of the enterprise. The generated statistical data can support higher-level decision-making.

[0122] In an embodiment of the present invention, the information in the initial enterprise portrait data is analyzed to identify relevant parts of the transportation data, such as information related to the GPS tracking system and freight management system related to the carrier. An interface parsing plan is designed and implemented to determine the data interfaces for accessing these data interfaces related to the transportation activities of the carrier. According to the interface information obtained by parsing, a data collection tool is developed or configured, such as an API call, a database query, or a direct data interface application programming interface (API), to perform data collection operations and collect transportation activity data from various systems and platforms, such as vehicle location, transportation time, cargo type and quantity, etc. The collected transportation activity data is verified to ensure the integrity and accuracy of the data, and necessary data preprocessing is performed, such as formatting, noise removal, data synchronization, etc., to facilitate subsequent statistical analysis. Specific parameters for the analysis of the entire transportation activity process are determined, such as the statistical time range (e.g., daily, weekly, monthly), and key statistical indicators (e.g., average transportation time, transportation volume, delay rate, etc.). Statistical models and algorithms are designed to ensure that valuable information can be effectively extracted from large amounts of data. Apply statistical software or write specific scripts, such as using Python's Pandas library or R language for data analysis, perform statistical analysis, process data according to preset parameters, and generate required statistical reports. Organize the results of statistical analysis and make them into tables, graphs, etc. for easy understanding and reporting, save statistical results in an appropriate format and location, ensure data security and accessibility, and provide a data basis for subsequent business analysis or decision support.

[0123] Preferably, step S4 comprises the following steps:

[0124] Step S41: setting the credit assessment demand characteristic data and the whole process analysis data of the transportation activity as the credit assessment index data of the carrier enterprise;

[0125] Step S42: performing credit assessment index influencing factor analysis on the carrier enterprise credit assessment index data to generate credit assessment index influencing factor data;

[0126] Step S43: Designing the credit assessment index weights according to the credit assessment index influencing factor data to generate credit assessment index weight data;

[0127] Step S44: designing a credit assessment radial basis function kernel according to the credit assessment index weight data;

[0128] Step S45: establishing a mapping relationship between the carrier credit assessment index and the carrier credit score through the credit assessment radial basis function kernel to generate a carrier credit score model;

[0129] Step S46: Use the carrier credit scoring model to process the carrier credit assessment index data to generate carrier credit scoring data.

[0130] The present invention provides a clear quantitative basis for credit evaluation by clearly defining which data are used as evaluation indicators. Formalizing relevant data as evaluation indicators is to ensure the transparency and consistency of the evaluation process. By identifying and analyzing which factors have a significant impact on credit scoring, it helps to manage the credit evaluation process in a refined manner, provides a basis for adjusting the scoring model and optimizing the evaluation strategy, and makes the scoring more accurate and reflects the real credit risk. Assigning weights to each influencing factor ensures that the evaluation model can fully consider the importance of each factor when quantifying the credit of the enterprise. The key to the design of weights is to balance the influence of each factor so that the scoring result is both comprehensive and fair. Using radial basis function kernels in modeling is an effective technique that can handle nonlinear relationships and enable the model to more accurately map the complex relationship between input data and credit scores. The credit scoring model finally generated can comprehensively consider multiple evaluation indicators, capture the relationship between these indicators and credit scores through advanced algorithms, and provide an accurate and reliable credit score for the carrier. By using the final scoring model to score the enterprise, a quantitative evaluation of the credit status of the carrier is achieved, and the generated credit score data provides key information for risk management and partnership evaluation.

[0131] As an embodiment of the present invention, refer to Figure 3 As shown, Figure 1 Detailed implementation step flow chart of step S4 in the embodiment, step S4 in this embodiment includes:

[0132] Step S41: setting the credit assessment demand characteristic data and the whole process analysis data of the transportation activity as the credit assessment index data of the carrier enterprise;

[0133] In the embodiment of the present invention, key indicators such as debt repayment ability, transportation efficiency, default record, etc. are selected from the credit assessment demand feature data and the whole process analysis data of the transportation activities, and the calculation method and data source of each indicator are defined to ensure that these indicators can effectively reflect the credit status of the carrier. The selected indicator data is standardized, and the data format and measurement unit are unified for subsequent analysis, and the unified indicator data is stored in a dedicated database or data warehouse.

[0134] Step S42: performing credit assessment index influencing factor analysis on the carrier enterprise credit assessment index data to generate credit assessment index influencing factor data;

[0135] In the embodiment of the present invention, statistical analysis and data mining techniques (such as regression analysis and principal component analysis) and historical data related to the carrier's credit assessment index data are used to determine the influence of each credit assessment index. The contribution and sensitivity of each index to the credit score are analyzed, the most influential factors are identified, and the analysis results are organized into credit assessment index influencing factor data, and the influence weight and relevance of each factor are clearly listed.

[0136] Step S43: Designing the credit assessment index weights according to the credit assessment index influencing factor data to generate credit assessment index weight data;

[0137] In the embodiment of the present invention, based on the results of the influencing factor analysis, a weight is assigned to each credit assessment indicator to reflect its importance to the credit score, and an appropriate mathematical model and algorithm are used to optimize the weight assignment to ensure the accuracy and fairness of the score. These weights are implemented in the credit scoring system to ensure that each score is based on the latest weight data, and the weight data is recorded and saved in detail to support the transparency and traceability of the system.

[0138] Step S44: designing a credit assessment radial basis function kernel according to the credit assessment index weight data;

[0139] In the embodiment of the present invention, an applicable radial basis function kernel type (such as Gaussian radial basis function) is selected, the parameters of the kernel function are configured according to the credit assessment indicator weight data, and the kernel function is designed to optimize the mapping of data in the feature space to improve the classification accuracy and generalization ability of the model. The designed radial basis function kernel is tested in a controlled environment to evaluate its performance to ensure that it can effectively process credit assessment data, and the parameters of the kernel function are adjusted according to the test results to optimize its processing effect.

[0140] Step S45: establishing a mapping relationship between the carrier credit assessment index and the carrier credit score through the credit assessment radial basis function kernel to generate a carrier credit score model;

[0141] In the embodiment of the present invention, the radial basis function kernel is used to map the credit assessment index to the high-dimensional feature space, and a complex relationship model between the index and the credit score is established. The machine learning training method (such as support vector machine) is used to train and verify the model to ensure the accuracy and stability of its prediction. The credit scoring model is established and verified on the actual data set to ensure the effectiveness of the model in practical applications, and the model parameters and verification results are recorded to provide data support for the subsequent adjustment and optimization of the model.

[0142] Step S46: Use the carrier credit scoring model to process the carrier credit assessment index data to generate carrier credit scoring data.

[0143] In the embodiment of the present invention, the collected and pre-processed carrier credit assessment index data is input into the credit scoring model, the model is used to calculate the carrier's credit score, and the detailed data in the scoring process is recorded. The generated credit score data is stored in a secure database to ensure the security and integrity of the data, and the credit score results are released to relevant departments or management as needed for credit management and decision support.

[0144] Preferably, step S44 includes the following steps:

[0145] An initial radial basis function kernel is designed based on a preset support vector machine algorithm; and parameters of the initial radial basis function kernel are adjusted according to credit assessment indicator weight data to generate a credit assessment radial basis function kernel.

[0146] The radial basis function kernel of the present invention is an effective means for processing nonlinear data relationships. The radial basis function kernel is introduced by using the support vector machine algorithm to help the model capture and understand complex data patterns, especially in problems such as credit scoring involving multi-dimensional factors. The support vector machine algorithm is a powerful machine learning tool that improves the generalization ability of the model by maximizing the interval of the decision boundary, so that good prediction accuracy can be maintained on unseen data. The radial basis function kernel can map data from the original space to a higher-dimensional feature space, so that the originally linearly inseparable data can be separated in the new space, thereby improving the model's adaptability to complex data structures. By adjusting the parameters of the radial basis function kernel according to the importance of the credit assessment indicators, it is ensured that the model pays more attention to factors that have a greater impact on credit assessment, which helps to enhance the interpretability and accuracy of the model. Parameter adjustment can be optimized according to actual data and business needs, making the model more flexible, able to adapt to different credit assessment scenarios and demand changes, providing more personalized assessment results, and making the model more sensitive to changes in credit assessment scenarios, so that in the assessment process, it can timely reflect the slight changes in the credit status of the enterprise, and improve the accuracy and timeliness of risk warning.

[0147] In the embodiment of the present invention, the Gaussian kernel in the support vector machine algorithm is used to evaluate the credit of the carrier, and the initial parameters are pre-set, including setting the carrier credit evaluation threshold and the kernel width parameter γ based on the historical carrier credit evaluation data. The kernel width parameter γ controls the width of the kernel function, affects the sensitivity and generalization ability of the model, and obtains the initial radial basis function kernel. The credit evaluation index weight data is analyzed to determine the contribution of each index to the credit score to guide the adjustment of the kernel parameter γ. Indicators with higher weights should have a greater impact in the model, which may require adjustment of γ to increase the sensitivity of the model to these indicators. According to the index weights and preliminary test results, the γ value is adjusted. Reducing the γ value will increase the kernel width, making the model less sensitive to small changes in the input data, which is suitable for avoiding overfitting and improving the generalization ability of the model. Multiple rounds of parameter adjustment and verification are performed, and the performance of the adjusted model is evaluated using cross-validation and other techniques to ensure that the optimal γ setting is found. Once the system is tested and verified, the final γ setting is confirmed and solidified as part of the credit evaluation radial basis function kernel to generate a credit evaluation radial basis function kernel for evaluating the credit information of the carrier.

[0148] Preferably, step S1 comprises the following steps:

[0149] Step S51: collecting carrier spot check data based on the initial enterprise portrait data to obtain carrier spot check data, wherein the carrier spot check data includes carrier application physical quantity data and carrier non-implementation license transportation data;

[0150] Step S52: Carrier spot check anomaly analysis is performed on the carrier spot check data to generate carrier spot check anomaly score data;

[0151] Step S53: updating the carrier credit score data in real time based on the carrier spot check abnormal score data to generate dynamic carrier credit score data.

[0152] The present invention obtains the latest data about the carrier through real-time spot checks, such as the data on the physical quantity applied for and the data on unlicensed transportation, thereby ensuring the timeliness and relevance of the data, and helping the regulatory authorities to promptly understand the actual operation of the company. Regular or irregular spot checks enable the regulatory authorities to effectively monitor the compliance of the carrier, promptly detect violations, and enhance the overall regulatory quality of the industry. Anomaly analysis of the spot check data can effectively identify risk points and potential violations in the company's operations, providing a basis for further regulatory decisions. Abnormal scoring data can further quantify the company's credit assessment and help relevant parties more accurately assess the company's credit status and operational quality. Real-time updates to the carrier's credit score can reflect the company's current credit status, making the credit score more accurate and timely. Updated credit score data can be quickly communicated to decision makers, helping them make faster and more reasonable decisions, especially in aspects such as carrier vehicle supervision.

[0153] In an embodiment of the present invention, the existing initial enterprise portrait data and external data interface are used to activate the data collection process, ensure the correctness and real-time nature of the data source, and obtain real-time data through field visits, electronic data access, or cooperation with carriers, including access to enterprise databases, API capture, or field monitoring, etc., to collect carrier spot check data. Identify and list the carrier's application physical quantity data (such as the applied transportation volume, number of applications, etc.) and unlicensed transportation data (such as records of transportation activities that violate the licensing regulations), set specific collection standards and parameters for each type of data, such as the time range of the data, the required level of detail, and the source of the data, perform quality checks on the collected data, verify the integrity and accuracy of the data, and ensure that there are no errors and omissions. Based on the pre-designed application physical quantity abnormality threshold and the carrier's unlicensed abnormal interval, define abnormal critical decisions for spot check data, such as transportation activities that exceed the normal range of the application physical quantity or are not licensed or licensed but not run. Use statistical analysis and machine learning techniques (such as outlier detection algorithms) to identify abnormal patterns in the data. Based on the degree of anomaly detected, generate an anomaly score for each carrier, which reflects the risk level that the company may have. Record the anomaly score in the database and classify and annotate it appropriately. Determine the potential impact of abnormal behavior found in the spot check on the overall credit score of the carrier. Based on the anomaly score results, adjust the carrier's credit score to ensure that the score reflects the latest company status. Use the adjusted carrier credit score to instantly adjust the carrier's credit profile, and promptly notify the internal department or external regulator that needs this information of the updated credit score.

[0154] Preferably, step S52 includes the following steps:

[0155] Conduct critical value analysis on the physical quantity data applied by the carrier to generate critical value data of the physical quantity applied;

[0156] Perform application physical quantity critical value abnormality scoring processing according to the application physical quantity critical value data to generate application physical quantity critical value abnormality scoring data;

[0157] According to the data of the transport enterprises that have not implemented the permitted transport, the critical value abnormal score processing of the non-implementation of permitted transport is carried out to generate the critical value abnormal score data of the applied physical quantity;

[0158] The data of abnormal score of critical value of physical quantity applied for and the data of abnormal score of critical value of physical quantity applied for are integrated to obtain the abnormal score data of spot check of carriers.

[0159] The present invention accurately identifies possible risks or abnormal behaviors, especially inconsistencies or abnormal high and low values ​​in the application data, by setting critical values ​​and analyzing the physical quantity data of the carrier's application. The analysis of critical values ​​helps the regulatory agency or the enterprise itself to ensure that all operations comply with industry standards and promote business transparency and traceability. The critical value analysis results are converted into specific abnormal scores, providing a quantitative risk assessment tool for enterprises or regulators, further risk management and decision-making. The abnormal score enables the enterprise or regulatory agency to quickly identify and respond to potential risks and take necessary measures in time to prevent the carrier from abnormal vehicles. The critical value abnormal score is used for the behavior of not implementing licensed transportation, monitoring whether the enterprise complies with relevant licenses and regulations, improving the compliance of the enterprise, and effectively preventing and reducing unauthorized transportation activities through strict analysis and scoring of these data, as well as identifying abnormal carriers that have licenses but do not run, protecting the interests of enterprises and the public. By integrating abnormal score data from different fields, a comprehensive risk assessment view is obtained, providing an overall risk status report, and the comprehensive abnormal score data provides strong decision support for regulatory agencies, thereby formulating more effective management strategies and intervention measures.

[0160] In an embodiment of the present invention, past application physical quantity data, including applied transportation volume, frequency, etc., are analyzed to determine the statistical range and distribution of normal operation. Based on the distribution of historical data, a critical value is set to identify abnormally high or abnormally low application physical quantities. This may involve calculating statistical indicators such as mean, median and standard deviation. The critical value defined above is applied to the currently collected application physical quantity data to identify data points that exceed the normal range, and the identified abnormal data points are marked and recorded as application physical quantity critical value data. Scoring rules are formulated for application physical quantity critical value data. For example, the farther the data point is from the critical value, the higher the abnormal score is assigned. The conversion rules are converted into an operational scoring system, and a numerical method is usually used to quantify the degree of abnormality. The scoring rules are applied to all marked abnormal application physical quantity data to generate specific scoring values, and the scoring results are saved as application physical quantity critical value abnormal scoring data. An analysis similar to the application physical quantity critical value abnormal scoring process is performed on the data of unlicensed transportation, and violations of regulations are identified. Each violation of the license is assigned an abnormal score according to its severity, and saved as unlicensed transportation critical value abnormal scoring data. The abnormal score data of applied physical quantities and unlicensed transportation are aggregated and integrated, and a comprehensive abnormal score is calculated based on the integrated data, which reflects the overall risk or degree of abnormality. The integrated spot check abnormal score data is stored in the database to ensure the security and queryability of the data, and to back up the data to prevent data loss or damage and ensure the long-term availability of the data.

[0161] The beneficial effects of the present application are that the credit profile assessment method of the carrier of the present invention can effectively reflect the company's immediate credit status and potential risks by integrating dynamic data from multiple sources of the carrier for credit assessment, and reflecting in real time the changes in the carrier's recent business performance and other reasons. Moreover, after the carrier's credit assessment is completed, it can also reflect the latest credit information of the carrier, ensure the certainty of the carrier's credit assessment and reduce risks, and dynamically update the carrier's credit assessment information through the carrier's spot check data, enhance the effectiveness of the credit assessment results of the supervision carrier, solve the problem of lagging credit profile assessment, accurately discover and assess the illegal and irregular transportation behaviors of enterprises, and avoid continuous illegal and irregular behaviors affecting the safety of infrastructure such as highways and bridges.

[0162] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0163] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A credit profile assessment method for a carrier, characterized in that: The following steps are involved: Step S1: Obtaining carrier information data; Establish initial enterprise portrait data based on the carrier information data; Step S2: analyzing the credit assessment demand characteristics of the carrier based on the initial enterprise portrait data to generate credit assessment demand characteristic data; Step S3: Analyze the entire process of the transport activities of the carrier based on the initial enterprise portrait data to generate analysis data of the entire process of the transport activities; Step S4: Carrier credit scoring is processed based on the credit assessment demand characteristic data and the whole process analysis data of the transportation activity to generate carrier credit scoring data; Wherein, step S4 comprises: Step S41: setting the credit assessment demand characteristic data and the whole process analysis data of the transportation activity as the credit assessment index data of the carrier enterprise; Step S42: performing credit assessment index influencing factor analysis on the carrier enterprise credit assessment index data to generate credit assessment index influencing factor data; Step S43: Designing the credit assessment index weights according to the credit assessment index influencing factor data to generate credit assessment index weight data; Step S44: designing a credit assessment radial basis function kernel according to the credit assessment index weight data, including: designing an initial radial basis function kernel based on a preset support vector machine algorithm; and adjusting parameters of the initial radial basis function kernel according to the credit assessment index weight data to generate a credit assessment radial basis function kernel; Step S45: establishing a mapping relationship between the carrier credit assessment index and the carrier credit score through the credit assessment radial basis function kernel to generate a carrier credit score model; Step S46: using the carrier credit scoring model to process the carrier credit assessment index data to generate carrier credit scoring data; Step S5: collecting carrier spot check data based on the initial enterprise portrait data to generate carrier spot check data; updating the carrier credit score data in real time based on the carrier spot check data to generate dynamic carrier credit score data; Step S6: Perform periodic credit score filling optimization processing on the initial enterprise portrait data based on the dynamic carrier enterprise credit score data to generate static enterprise credit portrait data.

2. The credit profile evaluation method for a carrier according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: Obtaining carrier information data; Step S12: performing information source identification processing on the carrier enterprise information data to generate information source identification data; Step S13: performing carrier enterprise data type classification processing on the carrier enterprise information data according to the information source identification data to obtain classified carrier enterprise information data, wherein the classified carrier enterprise information data includes carrier enterprise information sub-data and carrier enterprise information type; Step S14: Establish initial enterprise portrait data based on the divided carrier enterprise information data.

3. The credit profile evaluation method for a carrier according to claim 2, characterized in that: Step S14 includes the following steps: Design carrier information screening decisions and carrier information conversion decisions based on information source identification data; Perform data screening on the carrier information sub-data based on the carrier information screening decision to generate screened carrier information sub-data; Based on the carrier information conversion decision, the screening carrier information sub-data is coded and converted to generate coded carrier information sub-data; A multi-dimensional matrix node of the carrier is established according to the carrier information type, and the coded carrier information sub-data is mapped to the carrier portrait matrix node to establish the initial enterprise portrait data.

4. The credit profile evaluation method for a carrier according to claim 2, characterized in that: Step S2 includes the following steps: Step S21: extracting the credit assessment demand data of the carrier enterprise from the initial enterprise portrait data to obtain the credit assessment demand data; Step S22: performing credit assessment requirement feature analysis on the credit assessment requirement data based on a preset credit assessment requirement calculation logic to generate credit assessment requirement feature data.

5. The credit profile evaluation method for a carrier according to claim 4, characterized in that: Step S21 includes the following steps: The credit assessment demand index data is designed according to the information type of the carrier enterprise, the credit assessment demand index data is generated, and the credit assessment demand data of the carrier enterprise is extracted from the initial enterprise portrait data through the credit assessment demand index data to obtain the credit assessment demand data.

6. The credit profile evaluation method for a carrier according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: parsing the transport data collection interface of the transport enterprise according to the initial enterprise portrait data, and collecting the transport activity data of the transport enterprise through the transport data collection interface of the transport enterprise to generate transport activity data; Step S32: Performing full-process analysis on the transport activity data within a preset full-process analysis time to generate full-process analysis data.

7. The credit profile evaluation method for a carrier according to claim 1, characterized in that: Step S5 includes the following steps: Step S51: collecting carrier spot check data based on the initial enterprise portrait data to obtain carrier spot check data, wherein the carrier spot check data includes carrier application physical quantity data and carrier non-implementation license transportation data; Step S52: Carrier spot check anomaly analysis is performed on the carrier spot check data to generate carrier spot check anomaly score data; Step S53: updating the carrier credit score data in real time based on the carrier spot check abnormal score data to generate dynamic carrier credit score data.

8. The credit profile evaluation method for a carrier according to claim 7, characterized in that: Step S52 includes the following steps: Conduct critical value analysis on the physical quantity data applied by the carrier to generate critical value data of the physical quantity applied; Perform application physical quantity critical value abnormality scoring processing according to the application physical quantity critical value data to generate application physical quantity critical value abnormality scoring data; According to the data of the transport enterprises that have not implemented the permitted transport, the critical value abnormal score processing of the non-implementation of permitted transport is carried out to generate the critical value abnormal score data of the applied physical quantity; The data of abnormal score of critical value of physical quantity applied for and the data of abnormal score of critical value of physical quantity applied for are integrated to obtain the abnormal score data of spot check of carriers.

Images