A decision-making system and method based on enterprise credit

Abstract

The present invention provides a decision-making system and method based on corporate credit, wherein the system includes: an acquisition module for acquiring multiple cooperation items, wherein the cooperation items include: a first cooperating enterprise and a corresponding first cooperation project; an evaluation module for acquiring the credit information of the first enterprise, performing a credit evaluation based on the credit information, and obtaining an evaluation result; and a decision module for deciding whether to execute the corresponding first cooperation project based on the evaluation result, and outputting the decision result. In the decision-making system and method based on corporate credit of the present invention, a user only needs to input the cooperation item, and the system comprehensively acquires the credit information of the first enterprise in the cooperation item, performs a credit evaluation based on the credit information, and decides whether to execute the corresponding first cooperation project in the cooperation item based on the evaluation result, and outputs the decision result. Manual collection of credit information and evaluation is unnecessary, thereby reducing labor costs and avoiding the problem of incomplete manual collection.

Description

Technical Field

[0001] The present invention relates to the field of enterprise credit evaluation, and in particular to a decision-making system and method based on enterprise credit. Background Art

[0002] Currently, before cooperating with other companies, companies need to manually collect the credit information of the companies they are about to cooperate with, and then decide whether to cooperate with them based on the credit information.

[0003] However, manual collection of credit information has the disadvantages of incomplete collection and high labor costs;

[0004] Therefore, a solution is urgently needed. Summary of the Invention

[0005] One of the purposes of the present invention is to provide a decision-making system and method based on corporate credit. The user only needs to input the cooperation item, and the system will comprehensively obtain the credit information of the first enterprise in the cooperation item, conduct a credit evaluation based on the credit information, and decide whether to execute the corresponding first cooperation project in the cooperation item based on the evaluation result, and output the decision result. There is no need to manually collect credit information and evaluation, which reduces labor costs and avoids the problem of incomplete manual collection.

[0006] An embodiment of the present invention provides a decision-making system based on enterprise credit, including:

[0007] An acquisition module is used to acquire multiple cooperation items, where the cooperation items include: a first cooperating enterprise and a corresponding first cooperation project;

[0008] An evaluation module, configured to obtain credit information of the first enterprise, perform a credit evaluation based on the credit information, and obtain an evaluation result;

[0009] The decision module is used to decide whether to execute the corresponding first cooperation project based on the evaluation result and output the decision result.

[0010] Preferably, the evaluation module performs the following operations:

[0011] Obtaining a preset crawling strategy set, the crawling strategy set including: multiple first crawling strategies;

[0012] Perform strategy analysis and split the first crawling strategy to obtain multiple first strategies;

[0013] Sort the first strategy in order of strategy to obtain a strategy sequence;

[0014] Traverse the first strategy from the starting point to the end point of the strategy sequence;

[0015] Obtain a preset risk assessment model, input the traversed first strategy into the risk assessment model, and obtain a first assessment result, the first assessment result including: a first risk value and / or a risk direction, the risk direction including: forward and / or backward;

[0016] When the evaluation result includes a first risk value, if the first risk value is greater than or equal to a preset first threshold, obtain the first crawling object corresponding to the traversed first strategy, and at the same time, obtain the first credibility of the first crawling object. If the first credibility is less than or equal to a preset second threshold, stop traversing and eliminate the corresponding first crawling strategy;

[0017] When the evaluation result includes a risk direction, the first strategy in the risk direction of the first strategy traversed in the strategy sequence is selected as the second strategy;

[0018] Inputting the second strategy into the risk assessment model to obtain a second assessment result, the second assessment result including: a second risk value;

[0019] If the second risk value is greater than or equal to a preset third threshold, obtain the second crawling object corresponding to the second strategy, and at the same time, obtain the second credibility of the second crawling object. If the credibility is less than or equal to a preset fourth threshold, stop traversal and eliminate the corresponding first crawling strategy;

[0020] When all the first crawling strategies that need to be eliminated in the first crawling strategy are eliminated, the remaining first crawling strategies are used as the second crawling strategies;

[0021] crawling at least one identity corresponding to the first enterprise based on the second crawling strategy;

[0022] Acquiring credit record information, and determining at least one first information item corresponding to the identity in the credit record information;

[0023] Acquire provided information of the first information item, the provided information including: at least one first provider;

[0024] When the number of first providers is 1, obtaining a third credibility of the first provider, and if the third credibility is less than or equal to a preset fifth threshold, discarding the corresponding first information item;

[0025] When the number of first providers is greater than 1, obtaining the amount provided by the first providers corresponding to the first information item;

[0026] The first provider corresponding to the largest supply amount is designated as the second provider, and the remaining first providers are designated as the third providers;

[0027] Obtain the guarantee method for the second provider to guarantee the third provider;

[0028] Based on the preset guarantee method-guarantee value database, determine the guarantee value corresponding to the guarantee method;

[0029] obtaining a fourth credibility of the second provider;

[0030] If the fourth credibility is less than or equal to a preset sixth threshold and / or the guarantee value is less than or equal to a preset seventh threshold, the corresponding first information item is discarded;

[0031] After all first information items that need to be eliminated from the first information items are eliminated, the remaining first information items are used as second information items;

[0032] The second information items are integrated to obtain the credit information of the first enterprise, thereby completing the acquisition.

[0033] Preferably, the evaluation module performs the following operations:

[0034] performing basic feature extraction on the second information item to obtain at least one first feature;

[0035] Obtain a preset suspicious feature library, perform feature matching on the first feature with the second feature in the suspicious feature library, and if a match is found, use the corresponding second information item as the third information item, and at the same time, use the matched second feature as the third feature;

[0036] Acquire a first occurrence object corresponding to the third information item, and at the same time, acquire a second occurrence object corresponding to a second information item other than the third information item;

[0037] If the first occurrence object and the second occurrence object are the same, the corresponding second information item is used as the fourth information item;

[0038] Establishing a time axis, and arranging the third information item and the fourth information item on the time axis in a corresponding time sequence;

[0039] Selecting a fourth information item within a preset first range before and / or after the third information item on the time axis as the fifth information item;

[0040] performing deep feature extraction on the third information item and the fifth information item to obtain a plurality of fourth features;

[0041] Determining at least one first confirmed feature corresponding to the third feature based on a preset feature-confirmation feature library;

[0042] Perform feature matching on the fourth feature and the first verified feature. If a match is found, use the matched first verified feature as the second verified feature, and simultaneously obtain the source of the second verified feature, which includes: the third information item or the fifth information item;

[0043] Selecting the fourth information item within the preset second range after the source on the time axis as the sixth information item;

[0044] Determine the compensation analysis model corresponding to the second confirmation feature based on the preset confirmation feature-compensation analysis model library;

[0045] Inputting the sixth information item into the compensation analysis model to obtain a first analysis result, the first analysis result including: compensated and uncompensated;

[0046] When the first analysis result indicates that compensation has not been made, determining at least one credit evaluation item corresponding to the second substantiation feature based on a preset substantiation feature-credit evaluation item library;

[0047] Integrate various credit evaluation items, obtain evaluation results, and complete the credit evaluation.

[0048] Preferably, the decision module performs the following operations:

[0049] When the evaluation result includes a credit evaluation item, the corresponding first enterprise will be regarded as the second enterprise;

[0050] Select the first cooperation project corresponding to the second enterprise and use it as the second cooperation project;

[0051] Analyze and split the process of the second cooperation project to obtain multiple first processes;

[0052] Sort the first process according to the process sequence to obtain a process sequence;

[0053] Determining, based on a preset credit evaluation item-influence feature library, at least one first influencing feature corresponding to the credit evaluation item included in the evaluation result;

[0054] Traverse the first process from the starting point to the end point of the process sequence;

[0055] Performing feature analysis on the traversed first process to obtain at least one fifth feature;

[0056] Perform feature matching on the fifth feature and the first influencing feature. If the match is correct, use the traversed first process as the second process. At the same time, use the first influencing feature that matches as the second influencing feature.

[0057] Selecting the first process within a preset third range before and / or after the second process in the process sequence as the third process;

[0058] Combine the third process with the second process respectively to obtain a group to be analyzed;

[0059] Obtaining a preset correlation analysis model, inputting the group to be analyzed into the correlation analysis model, and obtaining a second analysis result, the second analysis result including: correlation degree;

[0060] When the correlation degree is greater than or equal to the preset eighth threshold, the corresponding correlation degree is valid;

[0061] Query the preset impact feature-impact value library to determine the impact value corresponding to the second impact feature;

[0062] Summarize the correlation and influence values to obtain the target value;

[0063] Continue to traverse the first process, and after each traversal, summarize the target values obtained previously to obtain the target value sum;

[0064] When the target value is greater than or equal to the preset ninth threshold, the second cooperation project cannot be executed and the decision is completed.

[0065] Preferably, the decision module performs the following operations:

[0066] When it is determined that the second cooperation project cannot be executed, outputting a decision result that the second cooperation project cannot be executed;

[0067] Otherwise, the output is a decision result that the second cooperation project can be executed.

[0068] An embodiment of the present invention provides a decision-making method based on corporate credit, including:

[0069] Step S1: Acquire multiple cooperation items, where the cooperation items include: a first cooperating enterprise and a corresponding first cooperation project;

[0070] Step S2: Obtaining the credit information of the first enterprise, performing a credit evaluation based on the credit information, and obtaining an evaluation result;

[0071] Step S3: Based on the evaluation result, decide whether to execute the corresponding first cooperation project and output the decision result.

[0072] Preferably, in step S2, obtaining the credit information of the first enterprise includes:

[0073] Obtaining a preset crawling strategy set, the crawling strategy set including: multiple first crawling strategies;

[0074] Perform strategy analysis and split the first crawling strategy to obtain multiple first strategies;

[0075] Sort the first strategy in order of strategy to obtain a strategy sequence;

[0076] Traverse the first strategy from the starting point to the end point of the strategy sequence;

[0077] Obtain a preset risk assessment model, input the traversed first strategy into the risk assessment model, and obtain a first assessment result, the first assessment result including: a first risk value and / or a risk direction, the risk direction including: forward and / or backward;

[0078] When the evaluation result includes a first risk value, if the first risk value is greater than or equal to a preset first threshold, obtain the first crawling object corresponding to the traversed first strategy, and at the same time, obtain the first credibility of the first crawling object. If the first credibility is less than or equal to a preset second threshold, stop traversing and eliminate the corresponding first crawling strategy;

[0079] When the evaluation result includes a risk direction, the first strategy in the risk direction of the first strategy traversed in the strategy sequence is selected as the second strategy;

[0080] Inputting the second strategy into the risk assessment model to obtain a second assessment result, the second assessment result including: a second risk value;

[0081] If the second risk value is greater than or equal to a preset third threshold, obtain the second crawling object corresponding to the second strategy, and at the same time, obtain the second credibility of the second crawling object. If the credibility is less than or equal to a preset fourth threshold, stop traversal and eliminate the corresponding first crawling strategy;

[0082] When all the first crawling strategies that need to be eliminated in the first crawling strategy are eliminated, the remaining first crawling strategies are used as the second crawling strategies;

[0083] crawling at least one identity corresponding to the first enterprise based on the second crawling strategy;

[0084] Acquiring credit record information, and determining at least one first information item corresponding to the identity in the credit record information;

[0085] Acquire provided information of the first information item, the provided information including: at least one first provider;

[0086] When the number of first providers is 1, obtaining a third credibility of the first provider, and if the third credibility is less than or equal to a preset fifth threshold, discarding the corresponding first information item;

[0087] When the number of first providers is greater than 1, obtaining the amount provided by the first providers corresponding to the first information item;

[0088] The first provider corresponding to the largest supply amount is designated as the second provider, and the remaining first providers are designated as the third providers;

[0089] Obtain the guarantee method for the second provider to guarantee the third provider;

[0090] Based on the preset guarantee method-guarantee value database, determine the guarantee value corresponding to the guarantee method;

[0091] obtaining a fourth credibility of the second provider;

[0092] If the fourth credibility is less than or equal to a preset sixth threshold and / or the guarantee value is less than or equal to a preset seventh threshold, the corresponding first information item is discarded;

[0093] After all first information items that need to be eliminated from the first information items are eliminated, the remaining first information items are used as second information items;

[0094] The second information items are integrated to obtain the credit information of the first enterprise, thereby completing the acquisition.

[0095] Preferably, in step S2, performing a credit evaluation based on the credit information to obtain an evaluation result includes:

[0096] performing basic feature extraction on the second information item to obtain at least one first feature;

[0097] Obtain a preset suspicious feature library, perform feature matching on the first feature with the second feature in the suspicious feature library, and if a match is found, use the corresponding second information item as the third information item, and at the same time, use the matched second feature as the third feature;

[0098] Acquire a first occurrence object corresponding to the third information item, and at the same time, acquire a second occurrence object corresponding to a second information item other than the third information item;

[0099] If the first occurrence object and the second occurrence object are the same, the corresponding second information item is used as the fourth information item;

[0100] Establishing a time axis, and arranging the third information item and the fourth information item on the time axis in a corresponding time sequence;

[0101] Selecting a fourth information item within a preset first range before and / or after the third information item on the time axis as the fifth information item;

[0102] performing deep feature extraction on the third information item and the fifth information item to obtain a plurality of fourth features;

[0103] Determining at least one first confirmed feature corresponding to the third feature based on a preset feature-confirmation feature library;

[0104] Perform feature matching on the fourth feature and the first verified feature. If a match is found, use the matched first verified feature as the second verified feature, and simultaneously obtain the source of the second verified feature, which includes: the third information item or the fifth information item;

[0105] Selecting the fourth information item within the preset second range after the source on the time axis as the sixth information item;

[0106] Determine the compensation analysis model corresponding to the second confirmation feature based on the preset confirmation feature-compensation analysis model library;

[0107] Inputting the sixth information item into the compensation analysis model to obtain a first analysis result, the first analysis result including: compensated and uncompensated;

[0108] When the first analysis result indicates that compensation has not been made, determining at least one credit evaluation item corresponding to the second substantiation feature based on a preset substantiation feature-credit evaluation item library;

[0109] Integrate various credit evaluation items, obtain evaluation results, and complete the credit evaluation.

[0110] Preferably, in step S3, based on the evaluation result, deciding whether to execute the corresponding first cooperation project includes:

[0111] When the evaluation result includes a credit evaluation item, the corresponding first enterprise will be regarded as the second enterprise;

[0112] Select the first cooperation project corresponding to the second enterprise and use it as the second cooperation project;

[0113] Analyze and split the process of the second cooperation project to obtain multiple first processes;

[0114] Sort the first process according to the process sequence to obtain a process sequence;

[0115] Determining, based on a preset credit evaluation item-influence feature library, at least one first influencing feature corresponding to the credit evaluation item included in the evaluation result;

[0116] Traverse the first process from the starting point to the end point of the process sequence;

[0117] Performing feature analysis on the traversed first process to obtain at least one fifth feature;

[0118] Perform feature matching on the fifth feature and the first influencing feature. If the match is correct, use the traversed first process as the second process. At the same time, use the first influencing feature that matches as the second influencing feature.

[0119] Selecting the first process within a preset third range before and / or after the second process in the process sequence as the third process;

[0120] Combine the third process with the second process respectively to obtain a group to be analyzed;

[0121] Obtaining a preset correlation analysis model, inputting the group to be analyzed into the correlation analysis model, and obtaining a second analysis result, the second analysis result including: correlation degree;

[0122] When the correlation degree is greater than or equal to the preset eighth threshold, the corresponding correlation degree is valid;

[0123] Query the preset impact feature-impact value library to determine the impact value corresponding to the second impact feature;

[0124] Summarize the correlation and influence values to obtain the target value;

[0125] Continue to traverse the first process, and after each traversal, summarize the target values obtained previously to obtain the target value sum;

[0126] When the target value is greater than or equal to the preset ninth threshold, the second cooperation project cannot be executed and the decision is completed.

[0127] Preferably, in step S3, outputting the decision result includes:

[0128] When it is determined that the second cooperation project cannot be executed, outputting a decision result that the second cooperation project cannot be executed;

[0129] Otherwise, the output is a decision result that the second cooperation project can be executed.

[0130] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purposes and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0131] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0132] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0133] Figure 1 Schematic diagram of a decision-making system based on enterprise credit in an embodiment of the present invention;

[0134] Figure 2 Flowchart of a decision-making method based on enterprise credit in an embodiment of the present invention;

[0135] Figure 3 This is a flowchart of another decision-making method based on corporate credit in an embodiment of the present invention. DETAILED DESCRIPTION

[0136] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0137] The embodiment of the present invention provides a decision-making system based on enterprise credit, such as Figure 1 Shown, including:

[0138] An acquisition module 1 is used to acquire multiple cooperation items, where the cooperation items include: a first cooperating enterprise and a corresponding first cooperation project;

[0139] Evaluation module 2, configured to obtain credit information of the first enterprise, perform credit evaluation based on the credit information, and obtain an evaluation result;

[0140] The decision module 3 is used to decide whether to execute the corresponding first cooperation project based on the evaluation result and output the decision result.

[0141] The working principle and beneficial effects of the above technical solution are:

[0142] Acquire multiple cooperation items, including the first enterprise to be cooperated with and the corresponding first cooperation project; obtain the credit information of the first enterprise (e.g., guarantee, pledge, and mortgage records, etc.), conduct a credit evaluation (evaluate whether the credit is good) based on the credit information, and obtain the evaluation result; based on the evaluation result, decide whether to execute the corresponding first cooperation project and output the decision result (sent to the user);

[0143] In the embodiment of the present invention, the user only needs to input the cooperation item, and the system will comprehensively obtain the credit information of the first enterprise in the cooperation item, conduct a credit evaluation based on the credit information, and decide whether to execute the corresponding first cooperation project in the cooperation item based on the evaluation result, and output the decision result. There is no need to manually collect credit information and evaluation, which reduces labor costs and avoids the problem of incomplete manual collection.

[0144] The embodiment of the present invention provides a decision-making system based on enterprise credit, wherein the evaluation module 2 performs the following operations:

[0145] Obtaining a preset crawling strategy set, the crawling strategy set including: multiple first crawling strategies;

[0146] Perform strategy analysis and split the first crawling strategy to obtain multiple first strategies;

[0147] Sort the first strategy in order of strategy to obtain a strategy sequence;

[0148] Traverse the first strategy from the starting point to the end point of the strategy sequence;

[0149] Obtain a preset risk assessment model, input the traversed first strategy into the risk assessment model, and obtain a first assessment result, the first assessment result including: a first risk value and / or a risk direction, the risk direction including: forward and / or backward;

[0150] When the evaluation result includes a first risk value, if the first risk value is greater than or equal to a preset first threshold, obtain the first crawling object corresponding to the traversed first strategy, and at the same time, obtain the first credibility of the first crawling object. If the first credibility is less than or equal to a preset second threshold, stop traversing and eliminate the corresponding first crawling strategy;

[0151] When the evaluation result includes a risk direction, the first strategy in the risk direction of the first strategy traversed in the strategy sequence is selected as the second strategy;

[0152] Inputting the second strategy into the risk assessment model to obtain a second assessment result, the second assessment result including: a second risk value;

[0153] If the second risk value is greater than or equal to a preset third threshold, obtain the second crawling object corresponding to the second strategy, and at the same time, obtain the second credibility of the second crawling object. If the credibility is less than or equal to a preset fourth threshold, stop traversal and eliminate the corresponding first crawling strategy;

[0154] When all the first crawling strategies that need to be eliminated in the first crawling strategy are eliminated, the remaining first crawling strategies are used as the second crawling strategies;

[0155] crawling at least one identity corresponding to the first enterprise based on the second crawling strategy;

[0156] Acquiring credit record information, and determining at least one first information item corresponding to the identity in the credit record information;

[0157] Acquire provided information of the first information item, the provided information including: at least one first provider;

[0158] When the number of first providers is 1, obtaining a third credibility of the first provider, and if the third credibility is less than or equal to a preset fifth threshold, discarding the corresponding first information item;

[0159] When the number of first providers is greater than 1, obtaining the amount provided by the first providers corresponding to the first information item;

[0160] The first provider corresponding to the largest supply amount is designated as the second provider, and the remaining first providers are designated as the third providers;

[0161] Obtain the guarantee method for the second provider to guarantee the third provider;

[0162] Based on the preset guarantee method-guarantee value database, determine the guarantee value corresponding to the guarantee method;

[0163] obtaining a fourth credibility of the second provider;

[0164] If the fourth credibility is less than or equal to a preset sixth threshold and / or the guarantee value is less than or equal to a preset seventh threshold, the corresponding first information item is discarded;

[0165] After all first information items that need to be eliminated from the first information items are eliminated, the remaining first information items are used as second information items;

[0166] The second information items are integrated to obtain the credit information of the first enterprise, thereby completing the acquisition.

[0167] The working principle and beneficial effects of the above technical solution are:

[0168] An enterprise may use different identities to cooperate with other parties, etc. Therefore, a first crawling strategy is set to crawl the identity of the enterprise to ensure the comprehensiveness and accuracy of credit information acquisition; the first crawling strategy (for example: crawling the website, etc.) is subjected to strategy analysis (analyzing which sub-strategies are there) and split to obtain the first strategy (for example: performing web page security detection, starting crawling, detecting whether there is a hyperlink, performing hyperlink security detection, crawling the web page data corresponding to the hyperlink), and the first strategy is sorted according to the strategy sequence (time sequence) to obtain a strategy sequence; the first strategy is input into a preset risk assessment model (a model generated by learning a large number of manual strategy risk assessment records using a machine learning algorithm) for risk assessment to obtain a first assessment result; if the first strategy has risks, the first assessment result will include a first risk value, and the larger the first risk value, the higher the risk level; if the first strategy is preceded by a If there is a risk in the first strategy and / or the next strategy, the first assessment result will include the risk direction; when the first risk value is greater than or equal to a preset first threshold value (for example, 25), obtain the first crawling object (for example, a web page) corresponding to the first strategy, obtain the first credibility of the first crawling object (for example, web page credibility); if the first credibility is less than or equal to a preset second threshold value (for example, 95), stop traversing, the first crawling strategy is not desirable and is eliminated; select the second strategy in the risk direction, input the second strategy into the risk assessment model, obtain the second risk value, the larger the second risk value, the higher the risk level, when the second risk value is greater than or equal to a preset third threshold value (for example, 10), determine the corresponding second crawling object (for example, a web page), obtain the second credibility, when the second credibility is less than or equal to a preset fourth threshold value (for example, 98), the first crawling strategy is not desirable and is eliminated;

[0169] Using the risk assessment model to determine the risk value, quickly determine the risk level, and at the same time, determine the risk direction, quickly identify strategies that may have risks, and conduct further risk assessments, thereby improving the screening efficiency of the first crawling strategy and ensuring the feasibility of the screened second crawling strategy;

[0170] Based on the second crawling strategy, the identity of the first enterprise (e.g., identity tag, etc.) is crawled to obtain credit record information (enterprise credit big data), and the first information item corresponding to the identity is determined. The first information item corresponds to a first provider (e.g., a credit investigation agency); if the number of first providers is 1, it means that it alone provides the first information item, and the third credibility of the first provider is obtained (which can be determined based on the authenticity of the information it has provided historically). If the third credibility is less than or equal to a preset fifth threshold (e.g., 98), the first information item is untrustworthy and can be eliminated; if the number of first providers is greater than 1, it means that multiple first providers provide the first information item in combination, and the provision amount (how much of the provided data accounts for the first information item) is determined. , the first provider corresponding to the largest amount provided is designated as the second provider, and the rest are designated as third providers; the first information item must be primarily provided by the second provider, and the second provider provides a guarantee for the third provider; obtain the guarantee method, and determine the guarantee value corresponding to the guarantee method (e.g., amount guarantee) based on a preset guarantee method-guarantee value database (a database containing guarantee values corresponding to different guarantee methods, where the larger the guarantee value, the stronger the guarantee); obtain the fourth credibility of the second provider; when the fourth credibility is less than or equal to a preset sixth threshold (e.g., 95) and / or the guarantee value is less than or equal to a preset seventh threshold (e.g., 85), the first information item is untrustworthy and is eliminated; and the remaining second information items are used as credit information;

[0171] When determining the first information items, verification is performed based on the number of first providers of the first information items, and first information items that fail the verification are eliminated to complete the screening and ensure the credibility of the remaining second information items.

[0172] The embodiment of the present invention provides a decision-making system based on enterprise credit, wherein the evaluation module 2 performs the following operations:

[0173] performing basic feature extraction on the second information item to obtain at least one first feature;

[0174] Obtain a preset suspicious feature library, perform feature matching on the first feature with the second feature in the suspicious feature library, and if a match is found, use the corresponding second information item as the third information item, and at the same time, use the matched second feature as the third feature;

[0175] Acquire a first occurrence object corresponding to the third information item, and at the same time, acquire a second occurrence object corresponding to a second information item other than the third information item;

[0176] If the first occurrence object and the second occurrence object are the same, the corresponding second information item is used as the fourth information item;

[0177] Establishing a time axis, and arranging the third information item and the fourth information item on the time axis in a corresponding time sequence;

[0178] Selecting a fourth information item within a preset first range before and / or after the third information item on the time axis as the fifth information item;

[0179] performing deep feature extraction on the third information item and the fifth information item to obtain a plurality of fourth features;

[0180] Determining at least one first confirmed feature corresponding to the third feature based on a preset feature-confirmation feature library;

[0181] Perform feature matching on the fourth feature and the first verified feature. If a match is found, use the matched first verified feature as the second verified feature, and simultaneously obtain the source of the second verified feature, which includes: the third information item or the fifth information item;

[0182] Selecting the fourth information item within the preset second range after the source on the time axis as the sixth information item;

[0183] Determine the compensation analysis model corresponding to the second confirmation feature based on the preset confirmation feature-compensation analysis model library;

[0184] Inputting the sixth information item into the compensation analysis model to obtain a first analysis result, the first analysis result including: compensated and uncompensated;

[0185] When the first analysis result indicates that compensation has not been made, determining at least one credit evaluation item corresponding to the second substantiation feature based on a preset substantiation feature-credit evaluation item library;

[0186] Integrate various credit evaluation items, obtain evaluation results, and complete the credit evaluation.

[0187] The working principle and beneficial effects of the above technical solution are:

[0188] Perform basic feature extraction on the second information item, and perform feature matching between the extracted first feature and the second feature in a preset suspicious feature library (a database containing multiple suspicious features, the specific suspicious features are, for example, suspicious features of bad credit); obtain the first occurrence object (for example, cooperation with enterprise B generates bad credit features, and the occurrence object is enterprise B), filter out the fourth information item with the same occurrence object, and set the third information item and the fourth information item on the time axis (corresponding the generation time node of the information item to the time node on the time axis); if there is a suspicious feature in the third information item, there may be more bad credit records near the corresponding time node, and select the fifth information item within a preset first range (for example, within one week) before and / or after it; based on the preset feature-confirmation feature library (containing confirmation features corresponding to different suspicious features of bad credit, the confirmation features are, for example, : a certain feature does exist), determine the first confirmed feature corresponding to the third feature; perform deep feature extraction on the third information item and the fifth information item, perform feature matching on the extracted fourth feature and the first confirmed feature, if the match is consistent, determine the source, after determining the source, whether the enterprise supplements the occurrence object, select the sixth information item within the preset second range (for example: one month), input the preset confirmed feature-compensation analysis model library (a database containing compensation analysis models corresponding to different confirmed features) to determine the compensation analysis model corresponding to the second confirmed feature (a model generated by learning a large number of manual compensation analysis records using a machine learning algorithm), and obtain the first analysis result; when no compensation is made, determine the credit evaluation item based on the preset confirmed feature-credit evaluation item library (a database containing credit evaluation items corresponding to different confirmed features), and integrate to obtain the evaluation result;

[0189] When evaluating credit information, basic feature extraction is first performed to determine whether there are suspicious features. If there are suspicious features, further in-depth feature extraction is performed. The settings are reasonable and the evaluation efficiency is improved.

[0190] The embodiment of the present invention provides a decision-making system based on enterprise credit, wherein the decision-making module 3 performs the following operations:

[0191] When the evaluation result includes a credit evaluation item, the corresponding first enterprise will be regarded as the second enterprise;

[0192] Select the first cooperation project corresponding to the second enterprise and use it as the second cooperation project;

[0193] Analyze and split the process of the second cooperation project to obtain multiple first processes;

[0194] Sort the first process according to the process sequence to obtain a process sequence;

[0195] Determining, based on a preset credit evaluation item-influence feature library, at least one first influencing feature corresponding to the credit evaluation item included in the evaluation result;

[0196] Traverse the first process from the starting point to the end point of the process sequence;

[0197] Performing feature analysis on the traversed first process to obtain at least one fifth feature;

[0198] Perform feature matching on the fifth feature and the first influencing feature. If the match is correct, use the traversed first process as the second process. At the same time, use the first influencing feature that matches as the second influencing feature.

[0199] Selecting the first process within a preset third range before and / or after the second process in the process sequence as the third process;

[0200] Combine the third process with the second process respectively to obtain a group to be analyzed;

[0201] Obtaining a preset correlation analysis model, inputting the group to be analyzed into the correlation analysis model, and obtaining a second analysis result, the second analysis result including: correlation degree;

[0202] When the correlation degree is greater than or equal to the preset eighth threshold, the corresponding correlation degree is valid;

[0203] Query the preset impact feature-impact value library to determine the impact value corresponding to the second impact feature;

[0204] Summarize the correlation and influence values to obtain the target value;

[0205] Continue to traverse the first process, and after each traversal, summarize the target values obtained previously to obtain the target value sum;

[0206] When the target value is greater than or equal to the preset ninth threshold, the second cooperation project cannot be executed and the decision is completed.

[0207] The working principle and beneficial effects of the above technical solution are:

[0208] Perform process analysis on the second cooperation project (analyze which processes there are) and split it to obtain the first process; sort the first process according to the process sequence (time sequence) to obtain the process sequence; based on the preset credit evaluation item-impact feature library (a database containing impact features corresponding to different credit evaluation items, for example: the credit evaluation item is delaying supply and settlement during cooperation, and the impact features are: supply and settlement, etc.), determine the first impact feature corresponding to the credit evaluation item; match the fifth feature with the first impact feature, and if the match is consistent, it means that it has an impact on it; there may be a correlation between the first processes of the second cooperation project, so select the third process within a preset third range (for example: 3) before and / or after the second process process; combining the third process and the second process, inputting a preset association analysis model (a model generated by learning a large number of records of manual association analysis using a machine learning algorithm) to obtain a correlation degree; when the correlation degree is greater than or equal to a preset eighth threshold (for example, 8), the corresponding correlation degree is prioritized; based on a preset impact feature-impact value library (a database containing impact values corresponding to different impact features, the larger the impact value, the greater the impact), determining the impact value; summing up (calculating the sum) the correlation degree and the impact value to obtain a target value, the larger the target value, the greater the impact; summing up (calculating the sum) the target values, when the target value sum is greater than or equal to a preset ninth threshold (for example, 100), the impact is large, and a decision that cannot be executed is made;

[0209] When deciding whether to execute the second cooperation project, it is necessary to determine whether the credit evaluation item has an impact on the first process in the second cooperation project and the extent of the impact. When analyzing the extent of the impact, the degree of correlation between the previous and / or subsequent processes is determined, and the overall impact (target value and) is analyzed. When the overall impact is large, it cannot be executed, which improves the rationality of the decision.

[0210] The embodiment of the present invention provides a decision-making system based on enterprise credit, wherein the decision-making module 3 performs the following operations:

[0211] When it is determined that the second cooperation project cannot be executed, outputting a decision result that the second cooperation project cannot be executed;

[0212] Otherwise, the output is a decision result that the second cooperation project can be executed.

[0213] The working principle and beneficial effects of the above technical solution are:

[0214] Based on the decision output, the corresponding decision results are convenient for users to view in time.

[0215] The embodiment of the present invention provides a decision-making system based on enterprise credit, wherein the evaluation module 2 performs the following operations:

[0216] Acquire a preset interface set, the interface set comprising: a plurality of first interfaces;

[0217] Obtaining a first credit value of the first interface, and if the first credit value is less than or equal to a preset tenth threshold, eliminating the corresponding first interface;

[0218] Otherwise, determining a plurality of first record items corresponding to the first interface based on a preset cooperation record library;

[0219] Acquire a second interface that cooperates with the first interface corresponding to the first record item, where the second interface includes: other first interfaces;

[0220] Summarize the first record items corresponding to the cooperation between the first interface and the same second interface and use them as the second record items;

[0221] Obtaining a preset cooperation evaluation model, inputting the second record item into the cooperation evaluation model, obtaining a cooperation value, and simultaneously obtaining a second credit value of the second interface;

[0222] The determination index is calculated based on the first credit value, cooperation value and second credit value, and the calculation formula is as follows:

[0223]

[0224]

[0225] Wherein, γ is the judgment index, l is the intermediate variable, σ1 and σ2 are preset weight values, α1 is the first credit value, β is the cooperation value, α2 is the second credit value, min is the minimum value function, {…,…} represents a set, and d is a preset constant;

[0226] If the determination index is less than or equal to a preset eleventh threshold, the corresponding first interface is eliminated;

[0227] After all the first interfaces that need to be removed from the first interfaces are removed, the remaining first interfaces are used as the third interface;

[0228] Acquire new target data through the third interface;

[0229] Integrate the target data, obtain credit record information, and complete the acquisition.

[0230] The working principle and beneficial effects of the above technical solution are:

[0231] The first interface is a network interface, corresponding to an information providing organization (for example, a credit investigation website); a first credit value of the first interface is obtained (which can be determined based on the authenticity of the data provided in the past); if the first credit value is less than or equal to the preset tenth threshold value (for example, 90), it is directly eliminated; otherwise, based on a preset cooperation record library (a database containing cooperation records between different interfaces, the cooperation record is specifically, for example, multiple interfaces jointly provide a record); the second record item is input into a preset cooperation evaluation model (a model generated by learning the records of manual evaluation interface cooperation using a machine learning algorithm) to obtain a cooperation value, the larger the cooperation value, the closer the cooperation; a second credit value of the second interface is obtained (which can also be determined based on the authenticity of the data provided in the past), and a judgment index is calculated based on the first credit value, the cooperation value, and the second credit value. If the judgment index is less than or equal to the preset eleventh threshold value (for example, 60), the corresponding first interface is eliminated, and the target data (for example, a credit record) is obtained through the remaining third interface for integration to ensure the security and reliability of the acquisition;

[0232] In the formula, the first credit value, the second credit value, and the cooperation value are positively correlated with the judgment index. |α1-α2|≥d indicates that the difference between the first credit value and the second credit value is large, that is, the second credit value is smaller, and the smaller value should be taken. Otherwise, it means that they are similar, and the sum of the two is taken.

[0233] The embodiment of the present invention provides a decision-making method based on corporate credit, such as Figure 2 Shown, including:

[0234] Step S1: Acquire multiple cooperation items, where the cooperation items include: a first cooperating enterprise and a corresponding first cooperation project;

[0235] Step S2: Obtaining the credit information of the first enterprise, performing a credit evaluation based on the credit information, and obtaining an evaluation result;

[0236] Step S3: Based on the evaluation result, decide whether to execute the corresponding first cooperation project and output the decision result.

[0237] The working principle and beneficial effects of the above technical solution are:

[0238] Acquire multiple cooperation items, including the first enterprise to be cooperated with and the corresponding first cooperation project; obtain the credit information of the first enterprise (e.g., guarantee, pledge, and mortgage records, etc.), conduct a credit evaluation (evaluate whether the credit is good) based on the credit information, and obtain the evaluation result; based on the evaluation result, decide whether to execute the corresponding first cooperation project and output the decision result (sent to the user);

[0239] In the embodiment of the present invention, the user only needs to input the cooperation item, and the system will comprehensively obtain the credit information of the first enterprise in the cooperation item, conduct a credit evaluation based on the credit information, and decide whether to execute the corresponding first cooperation project in the cooperation item based on the evaluation result, and output the decision result. There is no need to manually collect credit information and evaluation, which reduces labor costs and avoids the problem of incomplete manual collection.

[0240] An embodiment of the present invention provides a decision-making method based on enterprise credit. In step S2, the credit information of a first enterprise is obtained, including:

[0241] Obtaining a preset crawling strategy set, the crawling strategy set including: multiple first crawling strategies;

[0242] Perform strategy analysis and split the first crawling strategy to obtain multiple first strategies;

[0243] Sort the first strategy in order of strategy to obtain a strategy sequence;

[0244] Traverse the first strategy from the starting point to the end point of the strategy sequence;

[0245] Obtain a preset risk assessment model, input the traversed first strategy into the risk assessment model, and obtain a first assessment result, the first assessment result including: a first risk value and / or a risk direction, the risk direction including: forward and / or backward;

[0246] When the evaluation result includes a first risk value, if the first risk value is greater than or equal to a preset first threshold, obtain the first crawling object corresponding to the traversed first strategy, and at the same time, obtain the first credibility of the first crawling object. If the first credibility is less than or equal to a preset second threshold, stop traversing and eliminate the corresponding first crawling strategy;

[0247] When the evaluation result includes a risk direction, the first strategy in the risk direction of the first strategy traversed in the strategy sequence is selected as the second strategy;

[0248] Inputting the second strategy into the risk assessment model to obtain a second assessment result, the second assessment result including: a second risk value;

[0249] If the second risk value is greater than or equal to a preset third threshold, obtain the second crawling object corresponding to the second strategy, and at the same time, obtain the second credibility of the second crawling object. If the credibility is less than or equal to a preset fourth threshold, stop traversal and eliminate the corresponding first crawling strategy;

[0250] When all the first crawling strategies that need to be eliminated in the first crawling strategy are eliminated, the remaining first crawling strategies are used as the second crawling strategies;

[0251] crawling at least one identity corresponding to the first enterprise based on the second crawling strategy;

[0252] Acquiring credit record information, and determining at least one first information item corresponding to the identity in the credit record information;

[0253] Acquire provided information of the first information item, the provided information including: at least one first provider;

[0254] When the number of first providers is 1, obtaining a third credibility of the first provider, and if the third credibility is less than or equal to a preset fifth threshold, discarding the corresponding first information item;

[0255] When the number of first providers is greater than 1, obtaining the amount provided by the first providers corresponding to the first information item;

[0256] The first provider corresponding to the largest supply amount is designated as the second provider, and the remaining first providers are designated as the third providers;

[0257] Obtain the guarantee method for the second provider to guarantee the third provider;

[0258] Based on the preset guarantee method-guarantee value database, determine the guarantee value corresponding to the guarantee method;

[0259] obtaining a fourth credibility of the second provider;

[0260] If the fourth credibility is less than or equal to a preset sixth threshold and / or the guarantee value is less than or equal to a preset seventh threshold, the corresponding first information item is discarded;

[0261] After all first information items that need to be eliminated from the first information items are eliminated, the remaining first information items are used as second information items;

[0262] The second information items are integrated to obtain the credit information of the first enterprise, thereby completing the acquisition.

[0263] The working principle and beneficial effects of the above technical solution are:

[0264] An enterprise may use different identities to cooperate with other parties, etc. Therefore, a first crawling strategy is set to crawl the identity of the enterprise to ensure the comprehensiveness and accuracy of credit information acquisition; the first crawling strategy (for example: crawling the website, etc.) is subjected to strategy analysis (analyzing which sub-strategies are there) and split to obtain the first strategy (for example: performing web page security detection, starting crawling, detecting whether there is a hyperlink, performing hyperlink security detection, crawling the web page data corresponding to the hyperlink), and the first strategy is sorted according to the strategy sequence (time sequence) to obtain a strategy sequence; the first strategy is input into a preset risk assessment model (a model generated by learning a large number of manual strategy risk assessment records using a machine learning algorithm) for risk assessment to obtain a first assessment result; if the first strategy has risks, the first assessment result will include a first risk value, and the larger the first risk value, the higher the risk level; if the first strategy is preceded by a If there is a risk in the first strategy and / or the next strategy, the first assessment result will include the risk direction; when the first risk value is greater than or equal to a preset first threshold value (for example, 25), obtain the first crawling object (for example, a web page) corresponding to the first strategy, obtain the first credibility of the first crawling object (for example, web page credibility); if the first credibility is less than or equal to a preset second threshold value (for example, 95), stop traversing, the first crawling strategy is not desirable and is eliminated; select the second strategy in the risk direction, input the second strategy into the risk assessment model, obtain the second risk value, the larger the second risk value, the higher the risk level, when the second risk value is greater than or equal to a preset third threshold value (for example, 10), determine the corresponding second crawling object (for example, a web page), obtain the second credibility, when the second credibility is less than or equal to a preset fourth threshold value (for example, 98), the first crawling strategy is not desirable and is eliminated;

[0265] Using the risk assessment model to determine the risk value, quickly determine the risk level, and at the same time, determine the risk direction, quickly identify strategies that may have risks, and conduct further risk assessments, thereby improving the screening efficiency of the first crawling strategy and ensuring the feasibility of the screened second crawling strategy;

[0266] Based on the second crawling strategy, the identity of the first enterprise (e.g., identity tag, etc.) is crawled to obtain credit record information (enterprise credit big data), and the first information item corresponding to the identity is determined. The first information item corresponds to a first provider (e.g., a credit investigation agency); if the number of first providers is 1, it means that it alone provides the first information item, and the third credibility of the first provider is obtained (which can be determined based on the authenticity of the information it has provided historically). If the third credibility is less than or equal to a preset fifth threshold (e.g., 98), the first information item is untrustworthy and can be eliminated; if the number of first providers is greater than 1, it means that multiple first providers provide the first information item in combination, and the provision amount (how much of the provided data accounts for the first information item) is determined. , the first provider corresponding to the largest amount provided is designated as the second provider, and the rest are designated as third providers; the first information item must be primarily provided by the second provider, and the second provider provides a guarantee for the third provider; obtain the guarantee method, and determine the guarantee value corresponding to the guarantee method (e.g., amount guarantee) based on a preset guarantee method-guarantee value database (a database containing guarantee values corresponding to different guarantee methods, where the larger the guarantee value, the stronger the guarantee); obtain the fourth credibility of the second provider; when the fourth credibility is less than or equal to a preset sixth threshold (e.g., 95) and / or the guarantee value is less than or equal to a preset seventh threshold (e.g., 85), the first information item is untrustworthy and is eliminated; and the remaining second information items are used as credit information;

[0267] When determining the first information items, verification is performed based on the number of first providers of the first information items, and first information items that fail the verification are eliminated to complete the screening and ensure the credibility of the remaining second information items.

[0268] The embodiment of the present invention provides a decision-making method based on enterprise credit. In step S2, a credit evaluation is performed based on credit information to obtain an evaluation result, including:

[0269] performing basic feature extraction on the second information item to obtain at least one first feature;

[0270] Obtain a preset suspicious feature library, perform feature matching on the first feature with the second feature in the suspicious feature library, and if a match is found, use the corresponding second information item as the third information item, and at the same time, use the matched second feature as the third feature;

[0271] Acquire a first occurrence object corresponding to the third information item, and at the same time, acquire a second occurrence object corresponding to a second information item other than the third information item;

[0272] If the first occurrence object and the second occurrence object are the same, the corresponding second information item is used as the fourth information item;

[0273] Establishing a time axis, and arranging the third information item and the fourth information item on the time axis in a corresponding time sequence;

[0274] Selecting a fourth information item within a preset first range before and / or after the third information item on the time axis as the fifth information item;

[0275] performing deep feature extraction on the third information item and the fifth information item to obtain a plurality of fourth features;

[0276] Determining at least one first confirmed feature corresponding to the third feature based on a preset feature-confirmation feature library;

[0277] Perform feature matching on the fourth feature and the first verified feature. If a match is found, use the matched first verified feature as the second verified feature, and simultaneously obtain the source of the second verified feature, which includes: the third information item or the fifth information item;

[0278] Selecting the fourth information item within the preset second range after the source on the time axis as the sixth information item;

[0279] Determine the compensation analysis model corresponding to the second confirmation feature based on the preset confirmation feature-compensation analysis model library;

[0280] Inputting the sixth information item into the compensation analysis model to obtain a first analysis result, the first analysis result including: compensated and uncompensated;

[0281] When the first analysis result indicates that compensation has not been made, determining at least one credit evaluation item corresponding to the second substantiation feature based on a preset substantiation feature-credit evaluation item library;

[0282] Integrate various credit evaluation items, obtain evaluation results, and complete the credit evaluation.

[0283] The working principle and beneficial effects of the above technical solution are:

[0284] Perform basic feature extraction on the second information item, and perform feature matching between the extracted first feature and the second feature in a preset suspicious feature library (a database containing multiple suspicious features, the specific suspicious features are, for example, suspicious features of bad credit); obtain the first occurrence object (for example, cooperation with enterprise B generates bad credit features, and the occurrence object is enterprise B), filter out the fourth information item with the same occurrence object, and set the third information item and the fourth information item on the time axis (corresponding the generation time node of the information item to the time node on the time axis); if there is a suspicious feature in the third information item, there may be more bad credit records near the corresponding time node, and select the fifth information item within a preset first range (for example, within one week) before and / or after it; based on the preset feature-confirmation feature library (containing confirmation features corresponding to different suspicious features of bad credit, the confirmation features are, for example, : a certain feature does exist), determine the first confirmed feature corresponding to the third feature; perform deep feature extraction on the third information item and the fifth information item, perform feature matching on the extracted fourth feature and the first confirmed feature, if the match is consistent, determine the source, after determining the source, whether the enterprise supplements the occurrence object, select the sixth information item within the preset second range (for example: one month), input the preset confirmed feature-compensation analysis model library (a database containing compensation analysis models corresponding to different confirmed features) to determine the compensation analysis model corresponding to the second confirmed feature (a model generated by learning a large number of manual compensation analysis records using a machine learning algorithm), and obtain the first analysis result; when no compensation is made, determine the credit evaluation item based on the preset confirmed feature-credit evaluation item library (a database containing credit evaluation items corresponding to different confirmed features), and integrate to obtain the evaluation result;

[0285] When evaluating credit information, basic feature extraction is first performed to determine whether there are suspicious features. If there are suspicious features, further in-depth feature extraction is performed. The settings are reasonable and the evaluation efficiency is improved.

[0286] An embodiment of the present invention provides a decision-making method based on enterprise credit. In step S3, based on the evaluation result, a decision is made as to whether to execute the corresponding first cooperation project, including:

[0287] When the evaluation result includes a credit evaluation item, the corresponding first enterprise will be regarded as the second enterprise;

[0288] Select the first cooperation project corresponding to the second enterprise and use it as the second cooperation project;

[0289] Analyze and split the process of the second cooperation project to obtain multiple first processes;

[0290] Sort the first process according to the process sequence to obtain a process sequence;

[0291] Determining, based on a preset credit evaluation item-influence feature library, at least one first influencing feature corresponding to the credit evaluation item included in the evaluation result;

[0292] Traverse the first process from the starting point to the end point of the process sequence;

[0293] Performing feature analysis on the traversed first process to obtain at least one fifth feature;

[0294] Perform feature matching on the fifth feature and the first influencing feature. If the match is correct, use the traversed first process as the second process. At the same time, use the first influencing feature that matches as the second influencing feature.

[0295] Selecting the first process within a preset third range before and / or after the second process in the process sequence as the third process;

[0296] Combine the third process with the second process respectively to obtain a group to be analyzed;

[0297] Obtaining a preset correlation analysis model, inputting the group to be analyzed into the correlation analysis model, and obtaining a second analysis result, the second analysis result including: correlation degree;

[0298] When the correlation degree is greater than or equal to the preset eighth threshold, the corresponding correlation degree is valid;

[0299] Query the preset impact feature-impact value library to determine the impact value corresponding to the second impact feature;

[0300] Summarize the correlation and influence values to obtain the target value;

[0301] Continue to traverse the first process, and after each traversal, summarize the target values obtained previously to obtain the target value sum;

[0302] When the target value is greater than or equal to the preset ninth threshold, the second cooperation project cannot be executed and the decision is completed.

[0303] The working principle and beneficial effects of the above technical solution are:

[0304] Perform process analysis on the second cooperation project (analyze which processes there are) and split it to obtain the first process; sort the first process according to the process sequence (time sequence) to obtain the process sequence; based on the preset credit evaluation item-impact feature library (a database containing impact features corresponding to different credit evaluation items, for example: the credit evaluation item is delaying supply and settlement during cooperation, and the impact features are: supply and settlement, etc.), determine the first impact feature corresponding to the credit evaluation item; match the fifth feature with the first impact feature, and if the match is consistent, it means that it has an impact on it; there may be a correlation between the first processes of the second cooperation project, so select the third process within a preset third range (for example: 3) before and / or after the second process process; combining the third process and the second process, inputting a preset association analysis model (a model generated by learning a large number of records of manual association analysis using a machine learning algorithm) to obtain a correlation degree; when the correlation degree is greater than or equal to a preset eighth threshold (for example, 8), the corresponding correlation degree is prioritized; based on a preset impact feature-impact value library (a database containing impact values corresponding to different impact features, the larger the impact value, the greater the impact), determining the impact value; summing up (calculating the sum) the correlation degree and the impact value to obtain a target value, the larger the target value, the greater the impact; summing up (calculating the sum) the target values, when the target value sum is greater than or equal to a preset ninth threshold (for example, 100), the impact is large, and a decision that cannot be executed is made;

[0305] When deciding whether to execute the second cooperation project, it is necessary to determine whether the credit evaluation item has an impact on the first process in the second cooperation project and the extent of the impact. When analyzing the extent of the impact, the degree of correlation between the previous and / or subsequent processes is determined, and the overall impact (target value and) is analyzed. When the overall impact is large, it cannot be executed, which improves the rationality of the decision.

[0306] The embodiment of the present invention provides a decision-making method based on corporate credit, such as Figure 3 As shown, in step S3, the decision result is output, including:

[0307] Step S301: when it is determined that the second cooperation project cannot be executed, outputting a decision result that the second cooperation project cannot be executed;

[0308] Step S302: Otherwise, the output is a decision result that the second cooperation project can be executed.

[0309] The working principle and beneficial effects of the above technical solution are:

[0310] Based on the decision output, the corresponding decision results are convenient for users to view in time.

[0311] An embodiment of the present invention provides a decision-making method based on enterprise credit, which obtains credit record information, including:

[0312] Acquire a preset interface set, the interface set comprising: a plurality of first interfaces;

[0313] Obtaining a first credit value of the first interface, and if the first credit value is less than or equal to a preset tenth threshold, eliminating the corresponding first interface;

[0314] Otherwise, determining a plurality of first record items corresponding to the first interface based on a preset cooperation record library;

[0315] Acquire a second interface that cooperates with the first interface corresponding to the first record item, where the second interface includes: other first interfaces;

[0316] Summarize the first record items corresponding to the cooperation between the first interface and the same second interface and use them as the second record items;

[0317] Obtaining a preset cooperation evaluation model, inputting the second record item into the cooperation evaluation model, obtaining a cooperation value, and simultaneously obtaining a second credit value of the second interface;

[0318] The determination index is calculated based on the first credit value, cooperation value and second credit value, and the calculation formula is as follows:

[0319]

[0320]

[0321] Wherein, γ is the judgment index, l is the intermediate variable, σ1 and σ2 are preset weight values, α1 is the first credit value, β is the cooperation value, α2 is the second credit value, min is the minimum value function, {…,…} represents a set, and d is a preset constant;

[0322] If the determination index is less than or equal to a preset eleventh threshold, the corresponding first interface is eliminated;

[0323] After all the first interfaces that need to be removed from the first interfaces are removed, the remaining first interfaces are used as the third interface;

[0324] Acquire new target data through the third interface;

[0325] Integrate the target data, obtain credit record information, and complete the acquisition.

[0326] The working principle and beneficial effects of the above technical solution are:

[0327] The first interface is a network interface, corresponding to an information providing organization (for example, a credit investigation website); a first credit value of the first interface is obtained (which can be determined based on the authenticity of the data provided in the past); if the first credit value is less than or equal to the preset tenth threshold value (for example, 90), it is directly eliminated; otherwise, based on a preset cooperation record library (a database containing cooperation records between different interfaces, the cooperation record is specifically, for example, multiple interfaces jointly provide a record); the second record item is input into a preset cooperation evaluation model (a model generated by learning the records of manual evaluation interface cooperation using a machine learning algorithm) to obtain a cooperation value, the larger the cooperation value, the closer the cooperation; a second credit value of the second interface is obtained (which can also be determined based on the authenticity of the data provided in the past), and a judgment index is calculated based on the first credit value, the cooperation value, and the second credit value. If the judgment index is less than or equal to the preset eleventh threshold value (for example, 60), the corresponding first interface is eliminated, and the target data (for example, a credit record) is obtained through the remaining third interface for integration to ensure the security and reliability of the acquisition;

[0328] In the formula, the first credit value, the second credit value, and the cooperation value are positively correlated with the judgment index. |α1-α2|≥d indicates that the difference between the first credit value and the second credit value is large, that is, the second credit value is smaller, and the smaller value should be taken. Otherwise, it means that they are similar, and the sum of the two is taken.

[0329] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A decision-making system based on corporate credit, characterized by: include: An acquisition module is used to acquire multiple cooperation items, where the cooperation items include: a first cooperating enterprise and a corresponding first cooperation project; An evaluation module, configured to obtain credit information of the first enterprise, perform a credit evaluation based on the credit information, and obtain an evaluation result; A decision module, configured to decide whether to execute the corresponding first cooperation project based on the evaluation result and output the decision result; The decision module performs the following operations: When the evaluation result includes a credit evaluation item, the corresponding first enterprise will be used as the second enterprise; Select the first cooperation project corresponding to the second enterprise and use it as the second cooperation project; Analyze and split the process of the second cooperation project to obtain multiple first processes; Sort the first process according to the process sequence to obtain a process sequence; Determining, based on a preset credit evaluation item-influence feature library, at least one first influencing feature corresponding to the credit evaluation item included in the evaluation result; Traverse the first process from the starting point to the end point of the process sequence; Performing feature analysis on the traversed first process to obtain at least one fifth feature; Perform feature matching on the fifth feature and the first influencing feature. If the match is correct, use the traversed first process as the second process. At the same time, use the first influencing feature that matches as the second influencing feature. Selecting the first process within a preset third range before and / or after the second process in the process sequence as the third process; Combine the third process with the second process respectively to obtain a group to be analyzed; Obtaining a preset correlation analysis model, inputting the group to be analyzed into the correlation analysis model, and obtaining a second analysis result, the second analysis result including: correlation degree; When the correlation degree is greater than or equal to the preset eighth threshold, the corresponding correlation degree is valid; Query the preset impact feature-impact value library to determine the impact value corresponding to the second impact feature; Summarize the correlation and influence values to obtain the target value; Continue to traverse the first process, and after each traversal, summarize the target values obtained previously to obtain the target value sum; When the target value is greater than or equal to the preset ninth threshold, the second cooperation project cannot be executed and the decision is completed.

2. A decision-making system based on corporate credit as claimed in claim 1, characterized in that: The evaluation module performs the following operations: Obtaining a preset crawling strategy set, wherein the crawling strategy set includes: a plurality of first crawling strategies; Performing strategy analysis and splitting on the first crawling strategy to obtain multiple first strategies; Arrange the first strategies in order of strategy to obtain a strategy sequence; Traversing the first strategy in sequence from the starting point to the end point of the strategy sequence; Obtain a preset risk assessment model, input the traversed first strategy into the risk assessment model, and obtain a first assessment result, wherein the first assessment result includes: a first risk value and / or a risk direction, wherein the risk direction includes: forward and / or backward; When the assessment result includes a first risk value, if the first risk value is greater than or equal to a preset first threshold, obtaining a first crawling object corresponding to the traversed first strategy, and at the same time, obtaining a first credibility of the first crawling object; if the first credibility is less than or equal to a preset second threshold, stopping the traversal and eliminating the corresponding first crawling strategy; When the evaluation result includes a risk direction, selecting the first strategy in the risk direction of the first strategy traversed in the strategy sequence as the second strategy; Inputting the second strategy into the risk assessment model to obtain a second assessment result, wherein the second assessment result includes: a second risk value; If the second risk value is greater than or equal to a preset third threshold, obtain a second crawling object corresponding to the second strategy, and at the same time, obtain a second credibility of the second crawling object. If the credibility is less than or equal to a preset fourth threshold, stop traversal and eliminate the corresponding first crawling strategy; After all first crawling strategies that need to be eliminated in the first crawling strategies are eliminated, the remaining first crawling strategies are used as second crawling strategies; crawling at least one identity corresponding to the first enterprise based on the second crawling strategy; Acquiring credit record information, and determining at least one first information item corresponding to the identity in the credit record information; Acquire provided information of the first information item, the provided information including: at least one first provider; When the number of the first providers is 1, obtaining a third credibility of the first providers, and if the third credibility is less than or equal to a preset fifth threshold, eliminating the corresponding first information item; When the number of the first providers is greater than 1, obtaining the amount provided by the first providers corresponding to the first information item; The first provider corresponding to the largest provided amount is used as the second provider, and the remaining first providers are used as third providers; Obtaining the guarantee method used by the second provider to guarantee the third provider; Based on the preset guarantee method-guarantee value library, determine the guarantee value corresponding to the guarantee method; obtaining a fourth credibility of the second provider; If the fourth credibility is less than or equal to a preset sixth threshold and / or the guarantee value is less than or equal to a preset seventh threshold, discard the corresponding first information item; After all first information items that need to be eliminated from the first information items are eliminated, the remaining first information items are used as second information items; Integrate each of the second information items to obtain the credit information of the first enterprise, thereby completing the acquisition.

3. A decision-making system based on corporate credit as claimed in claim 2, characterized in that: The evaluation module performs the following operations: performing basic feature extraction on the second information item to obtain at least one first feature; Obtaining a preset suspicious feature library, performing feature matching on the first feature with a second feature in the suspicious feature library, and if a match is found, using the corresponding second information item as a third information item, and simultaneously using the matched second feature as a third feature; Acquire a first occurrence object corresponding to the third information item, and at the same time, acquire a second occurrence object corresponding to a second information item other than the third information item; If the first occurrence object and the second occurrence object are the same, the corresponding second information item is used as the fourth information item; Establishing a time axis, and arranging the third information item and the fourth information item on the time axis in a corresponding manner according to a time sequence; selecting a fourth information item within a preset first range before and / or after the third information item on the time axis as the fifth information item; performing deep feature extraction on the third information item and the fifth information item to obtain a plurality of fourth features; Determining at least one first confirmation feature corresponding to the third feature based on a preset feature-confirmation feature library; Performing feature matching on the fourth feature and the first verified feature; if a match is found, using the matched first verified feature as the second verified feature; and obtaining a source of the second verified feature, the source including the third information item or the fifth information item; selecting a fourth information item within a preset second range after the source on the time axis as the sixth information item; Determining a compensation analysis model corresponding to the second confirmation feature based on a preset confirmation feature-compensation analysis model library; Inputting the sixth information item into the compensation analysis model to obtain a first analysis result, wherein the first analysis result includes: compensated and uncompensated; When the first analysis result indicates that compensation is not available, determining at least one credit evaluation item corresponding to the second verification feature based on a preset verification feature-credit evaluation item library; Integrate the credit evaluation items, obtain the evaluation results, and complete the credit evaluation.

4. A decision-making system based on corporate credit as claimed in claim 1, characterized in that: The decision module performs the following operations: When it is determined that the second cooperation project cannot be executed, outputting a decision result that the second cooperation project cannot be executed; Otherwise, the output is a decision result of executing the second cooperation project.

5. A decision-making method based on corporate credit, characterized in that: include: Step S1: Acquire multiple cooperation items, where the cooperation items include: a first cooperating enterprise and a corresponding first cooperation project; Step S2: Obtaining the credit information of the first enterprise, performing a credit evaluation based on the credit information, and obtaining an evaluation result; Step S3: Based on the evaluation result, decide whether to execute the corresponding first cooperation project and output the decision result; In step S3, based on the evaluation result, a decision is made as to whether to execute the corresponding first cooperation project, including: When the evaluation result includes a credit evaluation item, the corresponding first enterprise will be regarded as the second enterprise; Select the first cooperation project corresponding to the second enterprise and use it as the second cooperation project; Analyze and split the process of the second cooperation project to obtain multiple first processes; Sort the first process according to the process sequence to obtain a process sequence; Determining, based on a preset credit evaluation item-influence feature library, at least one first influencing feature corresponding to the credit evaluation item included in the evaluation result; Traverse the first process from the starting point to the end point of the process sequence; Performing feature analysis on the traversed first process to obtain at least one fifth feature; Perform feature matching on the fifth feature and the first influencing feature. If the match is correct, use the traversed first process as the second process. At the same time, use the first influencing feature that matches as the second influencing feature. Selecting the first process within a preset third range before and / or after the second process in the process sequence as the third process; Combine the third process with the second process respectively to obtain a group to be analyzed; Obtaining a preset correlation analysis model, inputting the group to be analyzed into the correlation analysis model, and obtaining a second analysis result, the second analysis result including: correlation degree; When the correlation degree is greater than or equal to the preset eighth threshold, the corresponding correlation degree is valid; Query the preset impact feature-impact value library to determine the impact value corresponding to the second impact feature; Summarize the correlation and influence values to obtain the target value; Continue to traverse the first process, and after each traversal, summarize the target values obtained previously to obtain the target value sum; When the target value is greater than or equal to the preset ninth threshold, the second cooperation project cannot be executed and the decision is completed.

6. A decision-making method based on corporate credit as claimed in claim 5, characterized in that: In step S2, obtaining the credit information of the first enterprise includes: Obtaining a preset crawling strategy set, wherein the crawling strategy set includes: a plurality of first crawling strategies; Performing strategy analysis and splitting on the first crawling strategy to obtain multiple first strategies; Arrange the first strategies in order of strategy to obtain a strategy sequence; Traversing the first strategy in sequence from the starting point to the end point of the strategy sequence; Obtain a preset risk assessment model, input the traversed first strategy into the risk assessment model, and obtain a first assessment result, wherein the first assessment result includes: a first risk value and / or a risk direction, wherein the risk direction includes: forward and / or backward; When the assessment result includes a first risk value, if the first risk value is greater than or equal to a preset first threshold, obtaining a first crawling object corresponding to the traversed first strategy, and at the same time, obtaining a first credibility of the first crawling object; if the first credibility is less than or equal to a preset second threshold, stopping the traversal and eliminating the corresponding first crawling strategy; When the evaluation result includes a risk direction, selecting the first strategy in the risk direction of the first strategy traversed in the strategy sequence as the second strategy; Inputting the second strategy into the risk assessment model to obtain a second assessment result, wherein the second assessment result includes: a second risk value; If the second risk value is greater than or equal to a preset third threshold, obtain a second crawling object corresponding to the second strategy, and at the same time, obtain a second credibility of the second crawling object. If the credibility is less than or equal to a preset fourth threshold, stop traversal and eliminate the corresponding first crawling strategy; After all first crawling strategies that need to be eliminated in the first crawling strategies are eliminated, the remaining first crawling strategies are used as second crawling strategies; crawling at least one identity corresponding to the first enterprise based on the second crawling strategy; Acquiring credit record information, and determining at least one first information item corresponding to the identity in the credit record information; Acquire provided information of the first information item, the provided information including: at least one first provider; When the number of the first providers is 1, obtaining a third credibility of the first providers, and if the third credibility is less than or equal to a preset fifth threshold, eliminating the corresponding first information item; When the number of the first providers is greater than 1, obtaining the amount provided by the first providers corresponding to the first information item; The first provider corresponding to the largest provided amount is used as the second provider, and the remaining first providers are used as third providers; Obtaining the guarantee method used by the second provider to guarantee the third provider; Based on the preset guarantee method-guarantee value library, determine the guarantee value corresponding to the guarantee method; obtaining a fourth credibility of the second provider; If the fourth credibility is less than or equal to a preset sixth threshold and / or the guarantee value is less than or equal to a preset seventh threshold, discard the corresponding first information item; After all first information items that need to be eliminated from the first information items are eliminated, the remaining first information items are used as second information items; Integrate each of the second information items to obtain the credit information of the first enterprise, thereby completing the acquisition.

7. A decision-making method based on corporate credit as claimed in claim 6, characterized in that: In step S2, a credit evaluation is performed based on the credit information to obtain an evaluation result, including: performing basic feature extraction on the second information item to obtain at least one first feature; Obtaining a preset suspicious feature library, performing feature matching on the first feature with a second feature in the suspicious feature library, and if a match is found, using the corresponding second information item as a third information item, and simultaneously using the matched second feature as a third feature; Acquire a first occurrence object corresponding to the third information item, and at the same time, acquire a second occurrence object corresponding to a second information item other than the third information item; If the first occurrence object and the second occurrence object are the same, the corresponding second information item is used as the fourth information item; Establishing a time axis, and arranging the third information item and the fourth information item on the time axis in a corresponding manner according to a time sequence; selecting a fourth information item within a preset first range before and / or after the third information item on the time axis as the fifth information item; performing deep feature extraction on the third information item and the fifth information item to obtain a plurality of fourth features; Determining at least one first confirmation feature corresponding to the third feature based on a preset feature-confirmation feature library; Performing feature matching on the fourth feature and the first verified feature; if a match is found, using the matched first verified feature as the second verified feature; and obtaining a source of the second verified feature, the source including the third information item or the fifth information item; selecting a fourth information item within a preset second range after the source on the time axis as the sixth information item; Determining a compensation analysis model corresponding to the second confirmation feature based on a preset confirmation feature-compensation analysis model library; Inputting the sixth information item into the compensation analysis model to obtain a first analysis result, wherein the first analysis result includes: compensated and uncompensated; When the first analysis result indicates that compensation is not available, determining at least one credit evaluation item corresponding to the second verification feature based on a preset verification feature-credit evaluation item library; Integrate the credit evaluation items, obtain the evaluation results, and complete the credit evaluation.

8. A decision-making method based on corporate credit as claimed in claim 5, characterized in that: In step S3, outputting the decision result includes: When it is determined that the second cooperation project cannot be executed, outputting a decision result that the second cooperation project cannot be executed; Otherwise, the output is a decision result of executing the second cooperation project.

Images