A credit evaluation method for construction industry enterprises

Abstract

A credit evaluation method for construction industry enterprises relates to the field of informatization, characterized in that the method comprises the following steps: standard modeling: data analysis in index extraction, forming a standard model library; data aggregation: real-time detection of the qualification of the evaluation object and the evaluation user through unified authentication, forming the final evaluation data; data aggregation of the current evaluation period; special evaluation: executing the final evaluation data, index model and other data sources as parameters into the calculation engine for process log recording, obtaining the operation result and persistent storage. Advantage: by accessing different construction industry enterprise qualification credit evaluation standards, a unified standard index analysis is provided, mainly from the aspects of type, traceability length, data source, aggregation mode and the like. Through index data analysis, a data model of index extraction data is formed, and a standard index information is formed taking one type of information as an example.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a credit evaluation method for construction enterprises that provides unified standard indicator analysis by accessing different qualification and credit evaluation standards for construction enterprises. The analysis focuses on aspects such as type, traceability duration, data source, and aggregation mode, forming a data model for extracting indicator data. Taking one type of information as an example, standard indicator information is generated. The extracted indicator data is used to extract logical calculations to form general calculation rules. Grouping, splitting, and restriction logic modularizes and subdivides the entire indicator calculation process, forming a configurable and modular calculation rule system. Background Technology

[0002] With social development, information technology is gradually entering all walks of life. Currently, a relatively complete management method is needed to form a credit evaluation system for various enterprises in the construction industry. This method aims to achieve this by accessing different credit evaluation standards for construction enterprises and providing unified standard indicator analysis. The analysis should focus on aspects such as type, traceability duration, data source, and collection mode to form a data model for extracting indicator data. Taking one type of information as an example, standard indicator information is formed. The logic of calculation is extracted from the extracted indicator data to form general calculation rules. The calculation logic of grouping, splitting, and limiting modularizes and subdivides the entire indicator calculation, forming a configurable and modular calculation rule system. Summary of the Invention

[0003] This invention provides a credit evaluation method for construction enterprises. This method provides unified standard indicator analysis by accessing different credit evaluation standards for construction enterprises, mainly analyzing aspects such as type, traceability duration, data source, and aggregation mode. Through indicator data analysis, a data model for extracting indicator data is formed. Taking one type of information as an example, standard indicator information is formed. The extracted indicator data is used to extract computational logic to form general calculation rules. Through grouping, splitting, and restriction calculation logic, the entire indicator calculation is modularized and subdivided, forming a configurable and modular calculation rule system. Based on the indicator model formed by the indicator rule analysis, a set of indicator model configurations based on structured data is created. Based on modular programming technology, the indicator model configuration is divided into scoring details, data collection forms, duplicate verification, and scoring algorithms, achieving configurable programming. Subsequent standard modifications only require adding the corresponding modules. Using deep learning technology, by modeling existing indicator analysis and training its self-learning ability, the module of the indicator model is gradually optimized automatically. Based on the modular concept and the single responsibility principle, functions such as dynamic generation of input forms, global verification of input data, general configuration of scoring rules, and fine-grained scoring algorithms are implemented. The system offers two data collection methods: data extraction and self-gathering. Self-gathering verifies data validity through human-machine peer review. Human review provides three levels of approval (initial review, secondary review, and confirmation), requiring human intervention for initial or complex information. Machine review incorporates a machine learning module, continuously learning from human review data and data models, and accumulating sample data to gradually implement a combination of machine review and human verification. The system utilizes a pre-developed, self-developed Java-based scoring calculation engine. Multi-threading technology enables real-time calculation of tens of millions of data points; distributed computing significantly improves scoring efficiency; the calculation process is fully asynchronous, avoiding thread blocking, and includes asynchronous tasks, asynchronous messages, and asynchronous exception handling; a breakpoint resume function is implemented, allowing for resume scoring even when interrupted due to unforeseen circumstances, similar to resuming interrupted downloads; the entire scoring process is fully traceable, including real-time monitoring of scoring data and logging of the entire evaluation process. To ensure fast data retrieval, the system employs a distributed big data storage system, using unstructured data to significantly improve query speed; some configuration data is stored in structured data, improving readability.

[0004] This invention provides a credit evaluation method for construction companies, which includes the following steps:

[0005] Standard modeling: Extract the data type, duration, source, and aggregation pattern characteristics of the indicators, complete the indicator data extraction and analysis, refine the calculation rules of specific indicators based on the indicator characteristics, and establish the indicator nodes, observation nodes, scoring details, aggregation forms, duplicate checks, and scoring algorithm configurations in the indicator model to form a standard model library;

[0006] Data aggregation: Real-time detection of aggregation eligibility for evaluation objects and users that have passed unified authentication. Once the detection is passed, they automatically acquire aggregation eligibility and automatically extract and interactively aggregate evaluation indicator models to complete the construction of the evaluation data source and form the final evaluation data; data aggregation for this evaluation cycle is carried out.

[0007] Specialized evaluation: The evaluation process involves retrieving final evaluation data and using data sources such as indicator models as parameters from the evaluation objects and evaluation users as defined by the evaluation criteria. This data is then fed into the calculation engine for process logging, calculation results acquisition, and persistent storage.

[0008] A credit evaluation method for construction companies, wherein the standard modeling includes the following steps:

[0009] Standard indicator extraction and analysis: The evaluation standards are sorted and grouped to form data types. The validity period keywords in the data types are traced for a duration analysis. The accuracy of the data sources is judged and analyzed to form rating data sources. The rating data sources are collected according to the enterprise declaration data, shared data, and data crawling type to form a preliminary reserve for data aggregation.

[0010] Indicator data extraction and analysis: Analyze the types in the indicator data to obtain the specified indicator data items, and then analyze the specified indicator data according to real-time data and specified time frame through traceability analysis; determine the data source of the specified indicator data items through data source analysis, and complete the collection of the specified indicator data items through automatic docking, data sharing and human-machine combination; form indicator data extraction rules;

[0011] Indicator calculation rule analysis: By extracting the specified indicator data through calculation formulas to obtain the logical extraction to form general calculation rules, and then through calculation logic such as grouping, splitting, and restriction, the entire indicator calculation is modularized and subdivided to form a set of configurable scoring calculation models.

[0012] Indicator Model Configuration: Based on the calculation formula and final calculation result in the scoring calculation model, indicator nodes are created. Each created indicator node is configured with one or more observation nodes. Information items in existing observation node nodes are defined to complete the scoring details configuration, forming the name of the scoring details and the specific scoring details up to the lowest score. A data plotting form is defined to form scoring data items. Data content and duplication rate verification are performed on the same observation point submitted by the same evaluation object. The scoring details and data content of the observation node are integrated and calculated using an association algorithm group to obtain the evaluation result for a single observation node.

[0013] Standard model library management: The scoring details configuration, generated forms, evaluation execution rules, and duplicate verification rules are used to form indicator models and stored in the database; when a new indicator model is added, the data of the new indicator model is automatically expanded and integrated with the indicator models in the database.

[0014] A credit evaluation method for construction companies, wherein the standard indicator extraction and analysis includes the following steps:

[0015] Data analysis to obtain data types: The evaluation content and scoring criteria in the evaluation criteria are sorted out to obtain the sorted information types. The relevant information is then grouped according to type to obtain four data types: basic information, business performance, good information, and bad behavior information.

[0016] Duration analysis: Analyze the validity period of time-related data in various data types based on keywords;

[0017] Data source analysis: Accuracy assessment of the sources of various data in each evaluation indicator to obtain the data sources for evaluation and to prepare for data collection.

[0018] Data collection pattern analysis: The evaluation data sources formed by the analysis of data sources are classified, collected and formatted through enterprise declaration data, shared data and data crawling, and cached to prepare for the early stage of data collection.

[0019] A credit evaluation method for construction companies, wherein the analysis of indicator calculation rules includes the following steps:

[0020] Group Calculation Logic: This logic, based on the rules for extracting indicator data, involves dividing the specified indicator data into groups and then performing calculations on the data within each group. The formula is as follows:

[0021]

[0022] Where GROUP(i) represents the grouping calculation rule for the evaluation data. Indicate the specific calculation rules between groups

[0023] Indicator data extraction rules: Based on actual needs, assign values, split, classify, and time out one or more data extraction rules within the same group of calculation logic to obtain the indicator data extraction rules and calculation results that meet the requirements;

[0024] Boundary calculation logic: By comparing and analyzing the calculation results within the group with the specified boundary data, the boundary calculation logic and final calculation results that meet the requirements are obtained; the formula is:

[0025]

[0026] Where MIN represents the boundary minimum value, MAX represents the boundary maximum value, and ∩ indicates that...

[0027] Construct a scoring calculation model: By physically matching the final calculation result with the general calculation logic of the same group, the index data extraction rules, and the boundary calculation logic, a scoring calculation model that meets the requirements is formed. This model contains one or more scoring calculation formulas.

[0028] The upper limit of the interval is determined through a formula, which is as follows:

[0029]

[0030] Where a, b, and c represent the values ​​taken after comparison, and x, y, and z represent the values ​​compared.

[0031] The lower limit is obtained through formula calculation when performing interval determination. The formula is as follows:

[0032]

[0033] Where a, b, and c represent the values ​​taken after comparison, and x, y, and z represent the values ​​compared.

[0034] The comparison calculation is performed using a formula, which is as follows:

[0035] B(i)=(x=b?x=n:x=m)

[0036] Where x is the value to be compared, b is the value to be compared, n is the value of a successful comparison, and m is the value of a failed comparison.

[0037] The question mark (?) represents a comparison operator, and a colon (:) represents another case.

[0038] When performing comparison and exceeding / insufficiency calculations, the result is obtained through a calculation formula, which is as follows:

[0039] B(i)=(x=b?x=n:x=m)±(x÷c*|(xb)|)

[0040] Where x is the value to be compared, b is the value to be compared, n is the value of a successful comparison, and m is the value of a failed comparison.

[0041] The question mark (?) represents a comparison operator, the colon (:) represents another case, the plus or minus signifies addition or subtraction, the 'c' represents the base of the score, and the absolute value (||) represents taking the absolute value.

[0042] A credit evaluation method for construction companies, wherein the indicator model configuration includes the following steps:

[0043] Create indicator nodes: Create indicator nodes based on the calculation formula and final calculation result in the scoring calculation model. The created indicator nodes include: name, code, score, and associated indicator.

[0044] Observation point node: Configure one or more observation nodes for the created indicator node, define the information items in the existing observation point node to complete the scoring rules configuration, form the name of the scoring rules, the specific scoring rules configuration up to the lowest score, and define the data drawing form to form the scoring data items. Verify the data content and duplication rate of the same observation point submitted by the same evaluation object; integrate the scoring rules and data content of the observation point node through the association algorithm group to obtain the evaluation execution rules of a single observation node.

[0045] A credit evaluation method for construction companies, wherein the observation point nodes include the following steps:

[0046] Scoring details configuration: Configure one or more observation nodes for the created indicator nodes, define the information items in the existing observation node nodes to complete the scoring details configuration, and form the name of the scoring details and the specific scoring details configuration up to the lowest score.

[0047] Data aggregation form: Select whether to automatically generate a form; if yes, automatically generate the form; if no, connect to a customized form and generate a customized form based on specified parameters; select information items for evaluation based on indicator data, standardize and configure them to generate a form and configure the data;

[0048] Duplicate verification: For data content submitted by the same evaluation object under the same observation point, set one or more sets of keywords, and obtain the duplicate rate to support the data collection and evaluation results by semantic association matching of keywords and their similar words and synonyms;

[0049] Scoring: The scoring rules and data content of the observation point nodes are integrated by associating algorithm groups to obtain the evaluation execution rules for a single observation node, which is the scoring algorithm group.

[0050] A credit evaluation method for construction companies, wherein the data aggregation includes the following steps:

[0051] Unified Authentication: Provides authentication modes based on the type of the evaluation object, performs qualification verification and authentication based on built-in data and the identity of the evaluation object, and performs qualification verification and authentication through third-party authentication data sources and relevant identity information of the evaluation users;

[0052] Data aggregation: Evaluation objects and users who have passed unified certification are organized and aggregated into the evaluation qualification queue, and indicator data is automatically extracted and interactively aggregated; historical information is saved for evaluation objects and users who have not passed unified certification.

[0053] Constructing the evaluation data source: Retrieve each information item from the data aggregation form in the indicator model configuration, process it with the information model stored in the big data warehouse to remove interference items, merge the information models, and perform calculations on the information models to obtain the final evaluation data.

[0054] A credit evaluation method for construction companies, wherein the specific steps of data aggregation are as follows: evaluation objects and evaluation users who have passed unified certification are organized and aggregated into the evaluation qualification queue, and indicator data are automatically extracted and interactively aggregated; historical information is saved for evaluation objects and evaluation users who have not passed unified certification.

[0055] Data aggregation eligibility: Evaluation objects and users who have passed unified certification will be aggregated into the evaluation eligibility queue, which will be monitored in real time. If the evaluation object's aggregation eligibility is valid at the current time, the subsequent data aggregation work will continue; otherwise, the evaluation object or user's eligibility will be suspended until the evaluation object or user regains the relevant eligibility.

[0056] Automatic extraction: Evaluation objects and users meeting the aggregation qualifications are automatically included in the data extraction queue for extraction of evaluation indicator model data. Data interface authentication and connection are performed through a publicly available database. The evaluation objects and users are compared with the model data. When discrepancies are found, they are analyzed and compared to extract the relevant information. If the comparison results are consistent, the next step is performed. Each evaluation object is individually matched against the data model in the public database, and the matching results are synchronized to the local database. The local database records historical versions and is updated periodically. The evaluation indicator model data includes: enterprise data, personnel data, qualification data, and performance data.

[0057] Interactive Aggregation: Evaluation subjects and users who meet the aggregation eligibility criteria submit their self-reported positive information models along with relevant supporting materials according to the approval process. Reviewers verify the relevant information through a human-computer interaction mode. Positive information models that meet the requirements are stored in the big data warehouse, while those that do not meet the requirements are returned for modification and re-review. Reviewers then search for negative data models of evaluation subjects and users through regulatory records. When a relevant negative data model is found, the negative data of the relevant evaluation subject and user is extracted from the regulatory records, populated into the negative data model, and published. Evaluation subjects and users review and confirm this. The process ends when no relevant written data model is found.

[0058] A credit evaluation method for construction companies, wherein the specific evaluation includes the following steps:

[0059] Evaluation Restrictions: Evaluation prohibition clauses are set according to the evaluation rules. The characteristic data of the evaluation objects and users are compared one by one according to these prohibition clauses. If a prohibition clause is met, the evaluation object or user is not included in the evaluation object or user list. If a prohibition clause is not met, the evaluation object or user passes the restriction clause review process. The characteristic data of the evaluation objects and users are compared one by one according to the evaluation criteria of the restriction clauses. If all restriction clauses are met, a specified score is assigned, and the evaluation object or user is included in the evaluation object or user list. If the restriction clauses for newly included enterprises are met, a specified score is assigned, and the enterprise is included in the evaluation object or user list. If some restriction clauses are partially met, the evaluation is pending.

[0060] Evaluation Calculation: Retrieve the user list of each evaluation object and the evaluation data and indicator model that match it, and use them as parameters for the calculation engine to perform multi-threaded parallel calculations, obtain the calculation results and process logs, and push abnormal log messages. The entire process logs and calculation results are stored in unstructured distributed storage.

[0061] Results push: Define one or more push tasks and configure the information items included in the task, the push time period, the network environment, and the agreed encryption / decryption rules; push the configured push tasks.

[0062] The result delivery includes the following steps:

[0063] Information disclosure: Retrieve good information models from the big data warehouse and incorporate them into the automatic disclosure process. If no objections are received during the specified disclosure period, the data will be included in the evaluation data source.

[0064] Results sharing: Evaluation benchmark information and evaluation result information are shared to a public database through a data docking and authentication interface for data sharing.

[0065] A credit rating method for construction companies, wherein the rating calculation includes the following steps:

[0066] Obtaining data sources: Retrieve the final evaluation data, indicator model, and list of evaluation objects and users to form a parameter set for calculation;

[0067] The computing engine groups the evaluation objects and user lists sequentially and deploys one or more data processors according to actual needs. The data processors operate in a multi-threaded computing mode, performing asynchronous calculations on the data within the sequence based on the evaluation data parameters and indicator model parameters. It monitors the progress, messages, status, and anomalies during the asynchronous process in real time, generating a full-process log and calculation results. When abnormal tasks or results are displayed, the engine performs a cyclical calculation on the abnormal task or result after the calculation is completed. If the calculation result is determined to be a normal result, a full-process log is generated; if the calculation result is determined to be an abnormal result, the log is recorded and the log message is simultaneously pushed to the administrator.

[0068] formula:

[0069]

[0070]

[0071]

[0072] Where score represents the final calculated score, sources(i) is the data source, company is the enterprise data, scoring(i) is the calculation engine, the arrow indicates the data flow, base(i) is the basic information, good(i) is the good information, bad(i) is the bad information, and management(i) is the management information.

[0073] Persistent storage: An unstructured distributed storage model is used to store the entire process log and calculation results.

[0074] Therefore, it can be seen that:

[0075] This invention provides a credit evaluation method for construction enterprises by accessing different credit evaluation standards for construction enterprise qualifications, offering unified standard indicator analysis. The analysis focuses on aspects such as type, traceability duration, data source, and aggregation mode. Through indicator data analysis, a data model for extracting indicator data is formed. Taking one type of information as an example, standard indicator information is created. The extracted indicator data undergoes logical extraction to form general calculation rules. Through grouping, splitting, and restriction calculation logic, the entire indicator calculation is modularized and subdivided, forming a configurable and modular calculation rule system. Based on the indicator model formed by the indicator rule analysis, a set of indicator model configurations based on structured data is created. Using modular programming technology, the indicator model configuration is divided into scoring details, data collection forms, duplicate verification, and scoring algorithms, achieving configurable programming. Subsequent standard modifications only require adding the corresponding modules. Using deep learning technology, by modeling existing indicator analyses and training its self-learning ability, the module for automatically optimizing the indicator model is gradually realized. Based on the modular approach and the single responsibility principle, functions such as dynamic generation of input forms, global verification of input data, general configuration of scoring rules, and fine-grained scoring algorithms are implemented. The system offers two data collection methods: data extraction and self-gathering. Self-gathering verifies data validity through human-machine peer review. Human review provides three levels of approval (initial review, secondary review, and confirmation), requiring human intervention for initial or complex information. Machine review incorporates a machine learning module, continuously learning from human review data and data models, and accumulating sample data to gradually implement a combination of machine review and human verification. The system utilizes a pre-developed, self-developed Java-based scoring calculation engine. Multi-threading technology enables real-time calculation of tens of millions of data points; distributed computing significantly improves scoring efficiency; the calculation process is fully asynchronous, avoiding thread blocking, and includes asynchronous tasks, asynchronous messages, and asynchronous exception handling; a breakpoint resume function is implemented, allowing for resume scoring even when interrupted due to unforeseen circumstances, similar to resuming interrupted downloads; the entire scoring process is fully traceable, including real-time monitoring of scoring data and logging of the entire evaluation process. To ensure fast data retrieval, the system employs a distributed big data storage system, using unstructured data to significantly improve query speed; some configuration data is stored in structured data, improving readability. Attached Figure Description

[0076] Figure 1 A schematic diagram of the overall process of a credit evaluation method for construction enterprises provided as an embodiment of the present invention;

[0077] Figure 2 A flowchart illustrating the standard modeling steps in a credit evaluation method for construction enterprises, provided as an embodiment of the present invention;

[0078] Figure 3 A flowchart illustrating the standard indicator extraction and analysis steps in a credit evaluation method for construction enterprises, provided as an embodiment of the present invention;

[0079] Figure 4 A flowchart illustrating the indicator calculation rule analysis steps in a credit evaluation method for construction enterprises, provided as an embodiment of the present invention;

[0080] Figure 5 A flowchart illustrating the indicator model configuration steps in a credit evaluation method for construction enterprises, provided as an embodiment of the present invention;

[0081] Figure 6 A flowchart illustrating the observation point node steps in a credit evaluation method for construction enterprises, provided as an embodiment of the present invention;

[0082] Figure 7 A flowchart illustrating the data aggregation step in a credit evaluation method for construction enterprises, provided as an embodiment of the present invention;

[0083] Figure 8 A flowchart illustrating the data aggregation step in a credit evaluation method for construction enterprises, provided as an embodiment of the present invention;

[0084] Figure 9 A flowchart illustrating the specific evaluation steps in a credit evaluation method for construction enterprises, provided as an embodiment of the present invention;

[0085] Figure 10 This is a flowchart illustrating the evaluation calculation steps in a credit evaluation method for construction enterprises, provided as an embodiment of the present invention. Detailed Implementation

[0086] To enable those skilled in the art to better understand the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. The illustrative embodiments and descriptions of the present invention are used to explain the present invention, but are not intended to limit the present invention.

[0087] Example 1:

[0088] Figure 1 As a credit rating method for construction companies, such as Figure 1 As shown, the method includes the following steps:

[0089] Standard modeling: Extract the data type, duration, source, and aggregation pattern characteristics of the indicators, complete the indicator data extraction and analysis, refine the calculation rules of specific indicators based on the indicator characteristics, and establish the indicator nodes, observation nodes, scoring details, aggregation forms, duplicate checks, and scoring algorithm configurations in the indicator model to form a standard model library;

[0090] Data aggregation: Real-time detection of aggregation eligibility for evaluation objects and users that have passed unified authentication. Once the detection is passed, they automatically acquire aggregation eligibility and automatically extract and interactively aggregate evaluation indicator models to complete the construction of the evaluation data source and form the final evaluation data; data aggregation for this evaluation cycle is carried out.

[0091] Specialized evaluation: The evaluation process involves retrieving final evaluation data and using data sources such as indicator models as parameters from the evaluation objects and evaluation users as defined by the evaluation criteria. This data is then fed into the calculation engine for process logging, calculation results acquisition, and persistent storage.

[0092] like Figure 2 The present invention discloses a credit evaluation method for construction enterprises, wherein the standard modeling includes the following steps:

[0093] Standard indicator extraction and analysis: The evaluation standards are sorted and grouped to form data types. The validity period keywords in the data types are traced for a duration analysis. The accuracy of the data sources is judged and analyzed to form rating data sources. The rating data sources are collected according to the enterprise declaration data, shared data, and data crawling type to form a preliminary reserve for data aggregation.

[0094] Indicator data extraction and analysis: Analyze the types in the indicator data to obtain the specified indicator data items, and then analyze the specified indicator data according to real-time data and specified time frame through traceability analysis; determine the data source of the specified indicator data items through data source analysis, and complete the collection of the specified indicator data items through automatic docking, data sharing and human-machine combination; form indicator data extraction rules;

[0095] Indicator calculation rule analysis: By extracting the specified indicator data through calculation formulas to obtain the logical extraction to form general calculation rules, and then through calculation logic such as grouping, splitting, and restriction, the entire indicator calculation is modularized and subdivided to form a set of configurable scoring calculation models.

[0096] Indicator Model Configuration: Based on the calculation formula and final calculation result in the scoring calculation model, indicator nodes are created. Each created indicator node is configured with one or more observation nodes. Information items in existing observation node nodes are defined to complete the scoring details configuration, forming the name of the scoring details and the specific scoring details up to the lowest score. A data plotting form is defined to form scoring data items. Data content and duplication rate verification are performed on the same observation point submitted by the same evaluation object. The scoring details and data content of the observation node are integrated and calculated using an association algorithm group to obtain the evaluation result for a single observation node.

[0097] Standard model library management: The scoring details configuration, generated forms, evaluation execution rules, and duplicate verification rules are used to form indicator models and stored in the database; when a new indicator model is added, the data of the new indicator model is automatically expanded and integrated with the indicator models in the database.

[0098] like Figure 3 The present invention discloses a credit evaluation method for construction enterprises, wherein the standard indicator extraction and analysis includes the following steps:

[0099] Data analysis to obtain data types: The evaluation content and scoring criteria in the evaluation criteria are sorted out to obtain the sorted information types. The relevant information is then grouped according to type to obtain four data types: basic information, business performance, good information, and bad behavior information.

[0100] Duration analysis: Analyze the validity period of time-related data in various data types based on keywords;

[0101] Data source analysis: Accuracy assessment of the sources of various data in each evaluation indicator to obtain the data sources for evaluation and to prepare for data collection.

[0102] Data collection pattern analysis: The evaluation data sources formed by the analysis of data sources are classified, collected and formatted through enterprise declaration data, shared data and data crawling, and cached to prepare for the early stage of data collection.

[0103] like Figure 4 This paper presents a credit evaluation method for construction companies, in which the analysis of indicator calculation rules includes the following steps:

[0104] Group Calculation Logic: This logic, based on the rules for extracting indicator data, involves dividing the specified indicator data into groups and then performing calculations on the data within each group. The formula is as follows:

[0105]

[0106] Where GROUP(i) represents the grouping calculation rule for the evaluation data. Indicate the specific calculation rules between groups

[0107] Indicator data extraction rules: Based on actual needs, assign values, split, classify, and time out one or more data extraction rules within the same group of calculation logic to obtain the indicator data extraction rules and calculation results that meet the requirements;

[0108] Boundary calculation logic: By comparing and analyzing the calculation results within the group with the specified boundary data, the boundary calculation logic and final calculation results that meet the requirements are obtained; the formula is:

[0109]

[0110] Where MIN represents the boundary minimum value, MAX represents the boundary maximum value, and ∩ indicates that...

[0111] Construct a scoring calculation model: By physically matching the final calculation result with the general calculation logic of the same group, the index data extraction rules, and the boundary calculation logic, a scoring calculation model that meets the requirements is formed. This model contains one or more scoring calculation formulas.

[0112] The upper limit of the interval is determined through a formula, which is as follows:

[0113]

[0114] Where a, b, and c represent the values ​​taken after comparison, and x, y, and z represent the values ​​compared.

[0115] The lower limit is obtained through formula calculation when performing interval determination. The formula is as follows:

[0116]

[0117] Where a, b, and c represent the values ​​taken after comparison, and x, y, and z represent the values ​​compared.

[0118] The comparison calculation is performed using a formula, which is as follows:

[0119] B(i)=(x=b?x=n:x=m)

[0120] Where x is the value to be compared, b is the value to be compared, n is the value of a successful comparison, and m is the value of a failed comparison.

[0121] The question mark (?) represents a comparison operator, and a colon (:) represents another case.

[0122] When performing comparison and exceeding / insufficiency calculations, the result is obtained through a calculation formula, which is as follows:

[0123] B(i)=(x=b?x=n:x=m)±(x÷c*|(xb)|)

[0124] Where x is the value to be compared, b is the value to be compared, n is the value of a successful comparison, and m is the value of a failed comparison.

[0125] The question mark (?) represents a comparison operator, the colon (:) represents another case, the plus or minus signifies addition or subtraction, the 'c' represents the base of the score, and the absolute value (||) represents taking the absolute value.

[0126] like Figure 5 The present invention discloses a credit evaluation method for construction enterprises, wherein the configuration of the indicator model includes the following steps:

[0127] Create indicator nodes: Create indicator nodes based on the calculation formula and final calculation result in the scoring calculation model. The created indicator nodes include: name, code, score, and associated indicator.

[0128] Observation point node: Configure one or more observation nodes for the created indicator node, define the information items in the existing observation point node to complete the scoring rules configuration, form the name of the scoring rules, the specific scoring rules configuration up to the lowest score, and define the data drawing form to form the scoring data items. Verify the data content and duplication rate of the same observation point submitted by the same evaluation object; integrate the scoring rules and data content of the observation point node through the association algorithm group to obtain the evaluation execution rules of a single observation node.

[0129] like Figure 6 The present invention discloses a credit evaluation method for construction enterprises, wherein the observation point node includes the following steps:

[0130] Scoring details configuration: Configure one or more observation nodes for the created indicator nodes, define the information items in the existing observation node nodes to complete the scoring details configuration, and form the name of the scoring details and the specific scoring details configuration up to the lowest score.

[0131] Data aggregation form: Select whether to automatically generate a form; if yes, automatically generate the form; if no, connect to a customized form and generate a customized form based on specified parameters; select information items for evaluation based on indicator data, standardize and configure them to generate a form and configure the data;

[0132] Duplicate verification: For data content submitted by the same evaluation object under the same observation point, set one or more sets of keywords, and obtain the duplicate rate to support the data collection and evaluation results by semantic association matching of keywords and their similar words and synonyms;

[0133] Scoring: The scoring rules and data content of the observation point nodes are integrated by associating algorithm groups to obtain the evaluation execution rules for a single observation node, which is the scoring algorithm group.

[0134] like Figure 7 The present invention discloses a credit evaluation method for construction enterprises, wherein the data aggregation includes the following steps:

[0135] Unified Authentication: Provides authentication modes based on the type of the evaluation object, performs qualification verification and authentication based on built-in data and the identity of the evaluation object, and performs qualification verification and authentication through third-party authentication data sources and relevant identity information of the evaluation users;

[0136] Data aggregation: Evaluation objects and users who have passed unified certification are organized and aggregated into the evaluation qualification queue, and indicator data is automatically extracted and interactively aggregated; historical information is saved for evaluation objects and users who have not passed unified certification.

[0137] Constructing the evaluation data source: Retrieve each information item from the data aggregation form in the indicator model configuration, process it with the information model stored in the big data warehouse to remove interference items, merge the information models, and perform calculations on the information models to obtain the final evaluation data.

[0138] like Figure 8 The present invention provides a credit evaluation method for construction enterprises, wherein the specific steps of data aggregation are as follows: evaluation objects and evaluation users that have passed unified certification are organized and aggregated into the evaluation qualification queue, and indicator data are automatically extracted and interactively aggregated; historical information is saved for evaluation objects and evaluation users that have not passed unified certification.

[0139] Data aggregation eligibility: Evaluation objects and users who have passed unified certification will be aggregated into the evaluation eligibility queue, which will be monitored in real time. If the evaluation object's aggregation eligibility is valid at the current time, the subsequent data aggregation work will continue; otherwise, the evaluation object or user's eligibility will be suspended until the evaluation object or user regains the relevant eligibility.

[0140] Automatic extraction: Evaluation objects and users meeting the aggregation qualifications are automatically included in the data extraction queue for extraction of evaluation indicator model data. Data interface authentication and connection are performed through a publicly available database. The evaluation objects and users are compared with the model data. When discrepancies are found, they are analyzed and compared to extract the relevant information. If the comparison results are consistent, the next step is performed. Each evaluation object is individually matched against the data model in the public database, and the matching results are synchronized to the local database. The local database records historical versions and is updated periodically. The evaluation indicator model data includes: enterprise data, personnel data, qualification data, and performance data.

[0141] Interactive Aggregation: Evaluation subjects and users who meet the aggregation eligibility criteria submit their self-reported positive information models along with relevant supporting materials according to the approval process. Reviewers verify the relevant information through a human-computer interaction mode. Positive information models that meet the requirements are stored in the big data warehouse, while those that do not meet the requirements are returned for modification and re-review. Reviewers then search for negative data models of evaluation subjects and users through regulatory records. When a relevant negative data model is found, the negative data of the relevant evaluation subject and user is extracted from the regulatory records, populated into the negative data model, and published. Evaluation subjects and users review and confirm this. The process ends when no relevant written data model is found.

[0142] like Figure 9 The present invention discloses a credit evaluation method for construction enterprises, wherein the specific evaluation includes the following steps:

[0143] Evaluation Restrictions: Evaluation prohibition clauses are set according to the evaluation rules. The characteristic data of the evaluation objects and users are compared one by one according to these prohibition clauses. If a prohibition clause is met, the evaluation object or user is not included in the evaluation object or user list. If a prohibition clause is not met, the evaluation object or user passes the restriction clause review process. The characteristic data of the evaluation objects and users are compared one by one according to the evaluation criteria of the restriction clauses. If all restriction clauses are met, a specified score is assigned, and the evaluation object or user is included in the evaluation object or user list. If the restriction clauses for newly included enterprises are met, a specified score is assigned, and the enterprise is included in the evaluation object or user list. If some restriction clauses are partially met, the evaluation is pending.

[0144] Evaluation Calculation: Retrieve the user list of each evaluation object and the evaluation data and indicator model that match it, and use them as parameters for the calculation engine to perform multi-threaded parallel calculations, obtain the calculation results and process logs, and push abnormal log messages. The entire process logs and calculation results are stored in unstructured distributed storage.

[0145] Results push: Define one or more push tasks and configure the information items included in the task, the push time period, the network environment, and the agreed encryption / decryption rules; push the configured push tasks.

[0146] The result delivery includes the following steps:

[0147] Information disclosure: Retrieve good information models from the big data warehouse and incorporate them into the automatic disclosure process. If no objections are received during the specified disclosure period, the data will be included in the evaluation data source.

[0148] Results sharing: Evaluation benchmark information and evaluation result information are shared to a public database through a data docking and authentication interface for data sharing.

[0149] like Figure 10 The present invention discloses a credit evaluation method for construction enterprises, wherein the evaluation calculation includes the following steps:

[0150] Obtaining data sources: Retrieve the final evaluation data, indicator model, and list of evaluation objects and users to form a parameter set for calculation;

[0151] The computing engine groups the evaluation objects and user lists sequentially and deploys one or more data processors according to actual needs. The data processors operate in a multi-threaded computing mode, performing asynchronous calculations on the data within the sequence based on the evaluation data parameters and indicator model parameters. It monitors the progress, messages, status, and anomalies during the asynchronous process in real time, generating a full-process log and calculation results. When abnormal tasks or results are displayed, the engine performs a cyclical calculation on the abnormal task or result after the calculation is completed. If the calculation result is determined to be a normal result, a full-process log is generated; if the calculation result is determined to be an abnormal result, the log is recorded and the log message is simultaneously pushed to the administrator.

[0152] formula:

[0153]

[0154]

[0155]

[0156] Where score represents the final calculated score, sources(i) is the data source, company is the enterprise data, scoring(i) is the calculation engine, the arrow indicates the data flow, base(i) is the basic information, good(i) is the good information, bad(i) is the bad information, and management(i) is the management information.

[0157] Persistent storage: An unstructured distributed storage model is used to store the entire process log and calculation results.

[0158] Therefore, the credit evaluation method for construction enterprises in this embodiment of the invention provides a unified standard indicator analysis by accessing different credit evaluation standards for construction enterprises, mainly analyzing aspects such as type, traceability time, data source, and aggregation mode. Through indicator data analysis, a data model for extracting indicator data is formed, using one type of information as an example to form standard indicator information. The extracted indicator data is used to extract computational logic to form general calculation rules; through grouping, splitting, and restriction calculation logic, the entire indicator calculation is modularized and subdivided, forming a configurable and modular calculation rule system. Based on the indicator model formed by the indicator rule analysis, a set of indicator model configurations based on structured data is created; based on modular programming technology, the indicator model configuration is divided into scoring details, data collection forms, duplicate verification, and scoring algorithms, achieving configurable programming, and subsequent standard modifications only require adding the corresponding modules; using deep learning technology, through modeling existing indicator analysis, and by training its self-learning ability, the module of the indicator model is gradually automated for optimization; based on the modular idea and the single responsibility principle, functions such as dynamic generation of input forms, global verification of input data, general configuration of scoring rules, and fine-grained scoring algorithms are realized. The system offers two data collection methods: data extraction and self-gathering. Self-gathering verifies data validity through human-machine peer review. Human review provides three levels of approval (initial review, secondary review, and confirmation), requiring human intervention for initial or complex information. Machine review incorporates a machine learning module, continuously learning from human review data and data models, and accumulating sample data to gradually implement a combination of machine review and human verification. The system utilizes a pre-developed, self-developed Java-based scoring calculation engine. Multi-threading technology enables real-time calculation of tens of millions of data points; distributed computing significantly improves scoring efficiency; the calculation process is fully asynchronous, avoiding thread blocking, and includes asynchronous tasks, asynchronous messages, and asynchronous exception handling; a breakpoint resume function is implemented, allowing for resume scoring even when interrupted due to unforeseen circumstances, similar to resuming interrupted downloads; the entire scoring process is fully traceable, including real-time monitoring of scoring data and logging of the entire evaluation process. To ensure fast data retrieval, the system employs a distributed big data storage system, using unstructured data to significantly improve query speed; some configuration data is stored in structured data, improving readability.

[0159] Although embodiments of the invention have been described by way of examples, those skilled in the art will recognize that the invention has many variations and modifications without departing from its spirit, and it is intended that the appended claims cover such variations and modifications without departing from the spirit of the invention.

Claims

1. A credit evaluation method for construction enterprises, characterized in that, The method includes the following steps: Standard modeling: Extract the data type, duration, source, and aggregation pattern characteristics of the indicators, complete the indicator data extraction and analysis, refine the calculation rules of specific indicators based on the indicator characteristics, and establish the indicator nodes, observation nodes, scoring details, aggregation forms, duplicate checks, and scoring algorithm configurations in the indicator model to form a standard model library; The standard modeling includes the following steps: Standard indicator extraction and analysis: The evaluation standards are sorted and grouped to form data types. The validity period keywords in the data types are traced for a duration analysis. The accuracy of the data sources is judged and analyzed to form rating data sources. The rating data sources are collected according to the enterprise declaration data, shared data, and data crawling type to form a preliminary reserve for data aggregation. Indicator data extraction and analysis: Analyze the types in the indicator data to obtain the specified indicator data items, and then analyze the specified indicator data according to real-time data and specified time frame through traceability analysis; determine the data source of the specified indicator data items through data source analysis, and complete the collection of the specified indicator data items through automatic docking, data sharing and human-machine combination; form indicator data extraction rules; Indicator calculation rule analysis: By extracting the specified indicator data through calculation formulas to obtain the logical extraction to form general calculation rules, and then by grouping, splitting and restricting the calculation logic, the calculation of the entire indicator is modularized and subdivided to form a set of configurable scoring calculation models. Indicator Model Configuration: Based on the calculation formula and final calculation result in the scoring calculation model, indicator nodes are created. Each created indicator node is configured with one or more observation nodes. Information items in existing observation node nodes are defined to complete the scoring details configuration, forming the name of the scoring details and the specific scoring details up to the lowest score. A data plotting form is defined to form scoring data items. Data content and duplication rate verification are performed on the same observation point submitted by the same evaluation object. The scoring details and data content of the observation node are integrated and calculated using an association algorithm group to obtain the evaluation result for a single observation node. Standard model library management: The scoring details configuration, generated forms, evaluation execution rules, and duplicate verification rules are combined into an indicator model and stored in the database; when a new indicator model is added, the data of the new indicator model is automatically expanded and integrated with the indicator models in the database. Data aggregation: Real-time detection of aggregation eligibility for evaluation objects and users that have passed unified authentication. Once the detection is passed, they automatically acquire aggregation eligibility and automatically extract and interactively aggregate evaluation indicator models to complete the construction of the evaluation data source and form the final evaluation data; data aggregation for this evaluation cycle is carried out. The data aggregation includes the following steps: Unified Authentication: Provides authentication modes based on the type of the evaluation object, performs qualification verification and authentication based on built-in data and the identity of the evaluation object, and performs qualification verification and authentication through third-party authentication data sources and relevant identity information of the evaluation users; Data aggregation: Evaluation objects and users who have passed unified certification are organized and aggregated into the evaluation qualification queue, and indicator data is automatically extracted and interactively aggregated; historical information is saved for evaluation objects and users who have not passed unified certification. Constructing the evaluation data source: Retrieve each information item from the data aggregation form from the indicator model configuration, process it with the information model stored in the big data warehouse to remove interference items, merge the information models, and calculate the information model to obtain the final evaluation data; Specialized evaluation: For evaluation objects and evaluation users who have passed the evaluation criteria, the final evaluation data is retrieved and the indicator model data source is used as parameters to input into the calculation engine for process log recording, acquisition of calculation results and persistent storage; The specific evaluation includes the following steps: Evaluation restrictions: Evaluation prohibition clauses are set according to the evaluation rules. The characteristic data of the evaluation object and the evaluation user are compared one by one according to the prohibition clauses. If it is determined that the prohibition clauses are met, the evaluation object or evaluation user will not be included in the list. If it is determined that the prohibition clauses are not met, the evaluation will pass the restriction clause review process. According to the evaluation criteria of the restriction clauses, the characteristic data of the evaluation objects and evaluation users are compared one by one. When it is determined that all the restriction clauses are met, a specified score is assigned and the evaluation object or evaluation user is added to the list of evaluation objects or evaluation users. When it is determined that the restriction clauses of newly included enterprises are met, a specified score is assigned to the evaluation object or evaluation user and the evaluation object or evaluation user is added to the list of evaluation objects or evaluation users. When it is determined that some of the restriction clauses are met, the evaluation is pending. Evaluation Calculation: Retrieve the user list of each evaluation object and the evaluation data and indicator model that match it, and use them as parameters for the calculation engine to perform multi-threaded parallel calculations, obtain the calculation results and process logs, and push abnormal log messages. The entire process logs and calculation results are stored in unstructured distributed storage. Results push: Define one or more push tasks and configure the information items included in the task, the push time period, the network environment, and the agreed encryption / decryption rules; push the configured push tasks. The result delivery includes the following steps: Information disclosure: Retrieve good information models from the big data warehouse and incorporate them into the automatic disclosure process. If no objections are received during the specified disclosure period, the data will be included in the evaluation data source. Results sharing: Evaluation benchmark information and evaluation result information are shared to a public database through a data docking and authentication interface.

2. The credit evaluation method for construction enterprises according to claim 1, characterized in that, The standard indicator extraction and analysis includes the following steps: Data analysis to obtain data types: The evaluation content and scoring criteria in the evaluation criteria are sorted out to obtain the sorted information types. The relevant information is then grouped according to type to obtain four data types: basic information, business performance, good information, and bad behavior information. Duration analysis: Analyze the validity period of time-related data in various data types based on keywords; Data source analysis: Accuracy assessment of the sources of various data in each evaluation indicator to obtain the data sources for evaluation and to prepare for data collection. Data collection pattern analysis: The evaluation data sources formed by the analysis of data sources are classified, collected and formatted through enterprise declaration data, shared data and data crawling, and cached to prepare for the early stage of data collection.

3. The credit evaluation method for construction enterprises according to claim 1, characterized in that: The analysis of indicator calculation rules includes the following steps: Group Calculation Logic: This logic, based on the rules for extracting indicator data, involves dividing the specified indicator data into groups and then performing calculations on the data within each group. The formula is as follows: ,in This indicates the grouping and calculation rules for the evaluation data. Indicates the specific calculation rules between groups; Indicator data extraction rules: Based on actual needs, assign values, split, classify, and time out one or more data extraction rules within the same group of calculation logic to obtain the indicator data extraction rules and calculation results that meet the requirements; Boundary calculation logic: By comparing and analyzing the calculation results within the group with the specified boundary data, the boundary calculation logic and final calculation results that meet the requirements are obtained; the formula is: Where MIN represents the boundary minimum value and MAX represents the boundary maximum value. Indicates "and"; Construct a scoring calculation model: By physically matching the final calculation result with the general calculation logic of the same group, the index data extraction rules, and the boundary calculation logic, a scoring calculation model that meets the requirements is formed. This model contains one or more scoring calculation formulas. The upper limit of the interval is determined through a formula, which is as follows: Where a, b, and c represent the values ​​taken after comparison, and x, y, and z represent the values ​​compared. The lower limit is obtained through formula calculation when performing interval determination. The formula is as follows: Where a, b, and c represent the values ​​taken after comparison, and x, y, and z represent the values ​​compared. The comparison calculation is performed using a formula, which is as follows: , where x is the value to be compared, b is the value to be compared, n is the value that was successfully compared, m is the value that was not successfully compared, ? represents the comparison operator, and : represents another case; When performing comparison and exceeding / insufficiency calculations, the result is obtained through a calculation formula, which is as follows: Where x is the value to be compared, b is the value to be compared, n is the value of a successful comparison, and m is the value of a failed comparison. The question mark (?) indicates a comparison operator, and a colon (:) indicates another case. This indicates addition or subtraction, where c represents the base of the score, and || represents taking the absolute value.

4. The credit evaluation method for construction enterprises according to claim 1, characterized in that: The configuration of the indicator model includes the following steps: Create indicator nodes: Create indicator nodes based on the calculation formula and final calculation result in the scoring calculation model. The created indicator nodes include: name, code, score, and associated indicator. Observation point node: Configure one or more observation nodes for the created indicator node, define the information items in the existing observation point node to complete the scoring rules configuration, form the name of the scoring rules, the specific scoring rules configuration up to the lowest score, and define the data drawing form to form the scoring data items. Verify the data content and duplication rate of the same observation point submitted by the same evaluation object; integrate the scoring rules and data content of the observation point node through the association algorithm group to obtain the evaluation execution rules of a single observation node.

5. A credit evaluation method for construction enterprises according to claim 4, characterized in that, The observation point node includes the following steps: Scoring details configuration: Configure one or more observation nodes for the created indicator nodes, define the information items in the existing observation node nodes to complete the scoring details configuration, and form the name of the scoring details and the specific scoring details configuration up to the lowest score. Data aggregation form: Select whether to automatically generate a form; if yes, automatically generate the form; if no, connect to a customized form and generate a customized form based on specified parameters; select information items for evaluation based on indicator data, standardize and configure them to generate a form and configure the data; Duplicate verification: For data content submitted by the same evaluation object under the same observation point, set one or more sets of keywords, and obtain the duplicate rate to support the data collection and evaluation results by semantic association matching of keywords and their similar words and synonyms; Scoring: The scoring rules and data content of the observation point nodes are integrated by associating algorithm groups to obtain the evaluation execution rules for a single observation node, which is the scoring algorithm group.

6. The credit evaluation method for construction enterprises according to claim 1, characterized in that, The specific steps of the data aggregation are as follows: the evaluation objects and evaluation users that have passed unified authentication are organized and aggregated into the evaluation qualification queue, and the indicator data is automatically extracted and interactively aggregated; for the evaluation objects and evaluation users that have not passed unified authentication, historical information is saved. Data aggregation eligibility: Evaluation objects and users who have passed unified certification will be aggregated into the evaluation eligibility queue, which will be monitored in real time. If the evaluation object's aggregation eligibility is valid at the current time, the subsequent data aggregation work will continue; otherwise, the evaluation object or user's eligibility will be suspended until the evaluation object or user regains the relevant eligibility. Automatic extraction: Evaluation objects and users that meet the aggregation qualifications are automatically included in the data extraction queue. Evaluation indicator model data is extracted, and data interface authentication and connection are performed through a public database. The evaluation objects and users are compared with the model data. When discrepancies are found, they are analyzed and compared to obtain the discrepancy information. When the data comparison results are consistent, the next step is performed. The evaluation object information is searched and matched one by one with the data model in the public database. The matching results are synchronized to the local database, and historical versions are recorded and updated regularly in the local database. The evaluation indicator model data includes: enterprise data, personnel data, qualification data, and performance data; Interactive Aggregation: Evaluation subjects and users who meet the aggregation eligibility criteria submit their self-reported positive information models along with relevant supporting materials according to the approval process. Reviewers verify the relevant information through a human-computer interaction mode. Positive information models that meet the requirements are stored in the big data warehouse, while those that do not meet the requirements are returned for modification and re-review. Reviewers then search for negative data models of evaluation subjects and users through regulatory records. When a relevant negative data model is found, the negative data of the relevant evaluation subject and user is extracted from the regulatory records, populated into the negative data model, and published. Evaluation subjects and users review and confirm this. The process ends when no relevant written data model is found.

7. The credit evaluation method for construction enterprises according to claim 1, characterized in that, The evaluation calculation includes the following steps: Obtaining data sources: Retrieve the final evaluation data, indicator model, and list of evaluation objects and users to form a parameter set for calculation; The computing engine groups the evaluation objects and user lists sequentially and deploys one or more data processors according to actual needs. The data processors operate in a multi-threaded computing mode, performing asynchronous calculations on the data within the sequence based on the evaluation data parameters and indicator model parameters. It monitors the progress, messages, status, and anomalies during the asynchronous process in real time, generating a full-process log and calculation results. When abnormal tasks or results are displayed, the engine performs a cyclical calculation on the abnormal task or result after the calculation is completed. If the calculation result is determined to be a normal result, a full-process log is generated; if the calculation result is determined to be an abnormal result, the log is recorded and the log message is simultaneously pushed to the administrator. official: ; ; ; Where score represents the final calculated score, sources(i) is the data source, company is the enterprise data, scoring(i) is the calculation engine, the arrow indicates the data flow, base(i) is the basic information, good(i) is the good information, bad(i) is the bad information, and management(i) is the management information. Persistent storage: An unstructured distributed storage model is used to store the entire process log and calculation results.

Images