Bid invitation data review processing method and system and readable storage medium

Abstract

The invention discloses a bid invitation data review processing method and system and a readable storage medium. The method comprises the steps of obtaining data information associated with each bidder; analyzing and acquiring or capturing feature data in the data information based on user setting, and acquiring derivative data according to the feature data; dividing the feature data into fixed data and floating data based on historical bidding and tendering data and a set judgment rule, and automatically or manually configuring a floating rule and an influence factor for each piece of floating data; setting a bidding data verification model; setting and / or automatically determining an input value of a current influence factor, determining a current floating data value in combination with a floating rule, inputting current to-be-verified feature data and derivative data into the data verification model, and outputting a verification result; according to the scheme, the data information of the bidder can be automatically analyzed, the risk point can be pre-warned, the change data can be tracked, the review risk of a bidding agency is reduced, the workload of review experts is reduced, and the review quality is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of tender data processing, and more specifically, to a tender data review processing method, system and readable storage medium. Background Art

[0002] Bidding is a transaction method used in the purchase and sale of bulk goods, the awarding and contracting of engineering construction projects, and the procurement and supply of service projects under market economic conditions. It is a product of the highly developed commodity economy and an organized way of selecting the best for transactions. The general process of bidding is as follows: in the procurement of goods, engineering and services, the tenderer announces the procurement and requirements in advance to attract many bidders to compete on an equal basis under the same conditions. Finally, after a comprehensive evaluation of many bidders, the winner of the project is selected from them.

[0003] In order to ensure the smooth progress of bidding and procurement, the bidding team needs to verify and sort out the data information provided by the bidder (the above data information is usually recorded in the bid documents) and the bidder's qualifications, and analyze whether the bidder itself and the bidding plan meet the bidding requirements on the premise of ensuring that the data information is true and valid, and avoid risks. At present, in the bidding and procurement process, the query of supplier preliminary evaluation information, whether the award of bids and historical bidding and procurement winning situations are mutually exclusive, and whether there is bid rigging behavior are mostly analyzed by the bidding team offline manual operations. They can only analyze and judge explicit and fixed data information, and it is difficult to handle the inherent correlation and changing data information between data information, which ultimately leads to inefficiency and easy errors in the bidding and procurement data analysis and verification process in each link, which has an adverse impact on the bidding review results. Summary of the invention

[0004] In view of the problem that data search and processing in the existing bidding review process are mostly done manually, which increases the review risk of the bidding agency, reduces the review quality and is inefficient, the first purpose of this application is to propose a bidding data review processing method, which, through the participation of big data and artificial intelligence, can automatically analyze the bidder's data information and warn of risk points, and can track the changed data, reduce the review risk of the bidding agency, reduce the workload of review experts, and improve the review quality. In order to achieve the above method, the second purpose of this application is to propose a bidding data review processing system, and the third purpose is to protect a computer-readable storage medium loaded with a computer for implementing the bidding data review processing method as described above. The specific scheme is as follows:

[0005] A bidding data review processing method, comprising:

[0006] Obtaining data information associated with each bidder and storing it in a first database;

[0007] Analyze and obtain characteristic data in the data information or capture it based on user settings, obtain derivative data corresponding to the characteristic data according to the characteristic data, associate the characteristic data, the derivative data and the bidder information and store them as a second database;

[0008] Based on historical bidding data and according to set judgment rules, the characteristic data is divided into fixed data and floating data, and floating rules and influencing factors are automatically or manually configured for each floating data, and associated and stored as a third database;

[0009] Set up and store bid data validation models;

[0010] Set and / or automatically determine the input value of the current influencing factor, determine the value of the current floating data in combination with the floating rule, input the current feature data to be verified and its derivative data into the data verification model, and output the verification result based on the built-in rules of the model;

[0011] Among them, the data information includes the bidder's equity information data, the bidder's former name data, the bidder's credit data, historical bidding information data, and market policy data related to the bidder's qualifications and production capacity.

[0012] Through the above technical solution, when the system obtains the data information associated with the bidder, it automatically captures the feature data in the data information and inputs it into the data verification model for verification, which can reduce the workload of the bidding team staff in screening data; at the same time, after obtaining the feature data, the corresponding derivative data is generated through the semantic model, such as data that is similar to the description method of the feature data. Even if different bidders have different description methods for certain feature data, it does not affect the final evaluation quality, thereby improving the robustness and generalization ability of the system; for floating data, floating rules and corresponding influencing factors are configured, so that real-time tracking of feature data can be achieved, ensuring that the corresponding feature data is the latest data during the award verification analysis, ensuring the timeliness of the data, improving the evaluation quality of bidding and procurement, and reducing the risks during the later project implementation.

[0013] Further, the derived data includes: nicknames of nouns, former names, synonyms of nouns, conversion of physical units or monetary units, abbreviations of nouns, English translations of nouns, insertion of specific separators in words or paragraphs, or conclusions involving calculation formulas;

[0014] Obtain derivative data corresponding to the feature data, including:

[0015] Based on historical bidding data information and combined with deep learning algorithms, a derivative database or semantic model corresponding to the feature data in various bidding scenarios is established, and each derivative data is sorted and marked for application risks according to the correlation between each derivative data and the feature data;

[0016] The current bidding scenario is determined based on the current bidding project name, and related derivative data is searched from the corresponding derivative database based on the captured feature data, or related derivative data is directly generated through the corresponding semantic model.

[0017] Through the above technical solution, the characteristic data in the data information associated with the bidder can be more comprehensively entered into the data verification model to avoid data errors and improve the quality of bidding data review.

[0018] Furthermore, the floating data includes characteristic data that changes with time, business behavior or market policy factors;

[0019] The method further comprises:

[0020] Analyze the stability of each feature data based on historical bidding data and market policy big data;

[0021] According to the stability of each characteristic data and its relevance to the bidding and procurement project, a corresponding risk value is configured for each characteristic data;

[0022] Count the sum of the risk values ​​of each characteristic data of the bidder, and output warning data if it exceeds the set value; or

[0023] Based on the implementation steps of the bidding and procurement project, each feature data is classified, and the sum of the risk values ​​of each type of feature data is calculated. If it exceeds the set value, the alarm data is output.

[0024] Through the above technical solution, the bidding review personnel can be effectively reminded to review key characteristic data or certain types of data of the bidding projects, thereby improving the review efficiency, avoiding data errors and omissions, and improving the review quality.

[0025] Furthermore, the characteristic data to be verified include: the credit behavior data of the bidder, the equity relationship data of the bidder, the mutually exclusive relationship data of the bid award results of the bidder, and the relationship data between the bidder and the successful bidder of the same project that has been tendered in the early stage;

[0026] The built-in rules of the data validation model include:

[0027] Set data or keywords that affect the bidder's credit rating, and search and match them from the bidder's credit behavior data;

[0028] Based on the shareholder names of each bidder, analyze and trace back to the initial shareholder, and find out whether the names of the superior shareholders of each bidder are the same;

[0029] Analyze the current bidder's award based on historical bidding data information, and find out whether the award and the historical bidding and procurement winning situation are mutually exclusive;

[0030] Based on historical bidding data, obtain the data information of previous successful bidders of the same project, and combine the bidders' equity relationship data to determine the relationship between the current bidder and the previous successful bidder of the same project;

[0031] The risk value and sum of each characteristic data are counted to determine whether they exceed the set threshold.

[0032] Through the above technical solution, it is possible to quickly judge the bidder's data information, eliminate risk items, and reduce the risk of bidding review.

[0033] Furthermore, the method further comprises:

[0034] Establish the association between each type of project and the priority of the validation rules and sort the validation rules according to the priority of each validation rule;

[0035] Establishing a risk database for storing verification results of feature data to be verified;

[0036] Obtain the feature data to be verified and input it into the data verification model, verify the feature data one by one based on the above sorting, and if it does not meet the current verification rules: mark the feature data and / or terminate the verification process and store it in the risk database;

[0037] If the current validation rules are met, the next data validation will be performed until the validation is completed.

[0038] Through the above technical solution, important verification rules can be arranged at the front, thereby reducing the workload of later verification, avoiding invalid verification, and improving review efficiency.

[0039] Furthermore, the method further comprises:

[0040] Associate and store various bidding nodes in the same project directory and their data sharing relationships;

[0041] Determine the data sharing relationship between each bidding node and configure the sharing permissions for data sharing;

[0042] Based on the data sharing relationship between the above bidding nodes, automatically or manually set the follow-up update mechanism of data information between the bidding nodes;

[0043] Automatically query the bidding nodes associated with the current bidding node data information, and retrieve the data information of the associated bidding nodes based on the shared permission verification results;

[0044] When the data information of one of the bidding nodes changes, the data information of the bidding nodes associated with the current bidding node is automatically updated based on the above-mentioned follow-up update mechanism.

[0045] Through the above-mentioned technical solution, it is possible to achieve the interconnection and horizontal sharing of procurement data between nodes in each link of bidding and tendering, promote the coordinated development of bidding and procurement data analysis and verification capabilities at each node, avoid inconsistent data information between the previous and next links in the same project, and reduce the risk of subsequent bidding and evaluation.

[0046] Furthermore, the method further comprises:

[0047] Based on the current bidding project type, the bidding schemes of each bidder are obtained from the historical bidding data, and the differences between the bidding schemes of each bidder are compared pairwise based on the set rules to generate a first relationship database for representing the differences between the bidding schemes of each bidder;

[0048] Obtain the bidding schemes of each bidder in the current bidding project, and compare the differences between the bidding schemes of each bidder based on the set rules, so as to generate a second relationship database for characterizing the differences between the bidding schemes of each bidder in the current project of the same type;

[0049] If the difference between the specific data in the first relational database and the second relational database exceeds a set value, an alarm is output;

[0050] Among them, the above-mentioned specific data include: the ratio of the bid prices of each bidder's bidding proposals, the ratio of the implementation time of each bidder's bidding proposals, and the ratio between the risk values ​​of the bidding data.

[0051] Through the above technical solution, with the help of big data analysis, the differences between bidders in the bidding process of the same type of projects in history can be obtained. Through the analysis and judgment of the above differences, it can be inferred whether there are data anomalies or multiple bidders colluding in the bidding process of the current project, which is convenient for the bidding team reviewers to make accurate judgments on relevant data information, reduce review risks and improve review quality.

[0052] A bidding data review and processing system, comprising:

[0053] A data acquisition module is configured to connect to external data sources via a set API interface or to acquire data information of each bidder through data crawler software;

[0054] A data derivation module is configured to parse and obtain or capture characteristic data in the data information based on user settings, and obtain derivative data corresponding to the characteristic data according to the characteristic data;

[0055] A data floating analysis module is configured to divide the characteristic data into fixed data and floating data based on historical bidding data and according to set judgment rules, and automatically or manually configure floating rules and influencing factors for each floating data;

[0056] The data verification output module has built-in data verification rules, and is configured to determine the value of the current floating data according to the input value of the current influencing factor in combination with the floating rule, input the current feature data to be verified and its derivative data into the data verification model, and output the verification result based on the built-in rules of the model;

[0057] A data storage module, configured as a plurality of databases, for storing data information associated with each bidder, the characteristic data, the derived data and the associated bidder information, and the floating rules and influencing factors corresponding to each floating data;

[0058] Among them, the data information includes the bidder's equity information data, the bidder's former name data, the bidder's credit data, historical bidding information data, and market policy data related to the bidder's qualifications and production capacity.

[0059] Furthermore, the system also includes:

[0060] A data sharing relationship storage module is configured to associate and store various bidding nodes in the same project directory and their data sharing relationships and sharing permissions;

[0061] The data automatic update module has a built-in follow-up update mechanism, which is configured to automatically query the bidding nodes associated with the current bidding node data information, and retrieve the data information of the associated bidding nodes based on the shared permission verification results;

[0062] When the data information of one of the bidding nodes changes, the data information of the bidding nodes associated with the current bidding node is automatically updated based on the above-mentioned follow-up update mechanism.

[0063] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the bidding data review and processing method as described above.

[0064] The above technical solution helps the bidding data review and processing solution described in this application to be conveniently, widely and flexibly applied in bidding data review scenarios.

[0065] Compared with the prior art, the present invention has the following beneficial effects:

[0066] (1) By automatically capturing the characteristic data in the data information and inputting it into the data verification model for verification based on the data information associated with the bidders, the workload of the bidding team staff in screening data can be reduced while ensuring the accuracy of data verification;

[0067] (2) By configuring a dedicated semantic model, corresponding derivative data is generated based on the input feature data. Therefore, even if different bidders have different descriptions of certain feature data in their bids, it will not affect the final evaluation quality, thereby improving the robustness and generalization ability of the system.

[0068] (3) By configuring floating rules and corresponding influencing factors for each floating data, real-time tracking of characteristic data can be achieved, ensuring that the corresponding characteristic data is the latest data during the award verification analysis, ensuring the timeliness of the data, improving the review quality of bidding and procurement, and reducing the risks in the later project implementation, thereby reducing the review risk of the bidding agency, reducing the workload of the review experts, and improving the review quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 This is a schematic diagram of the overall steps of the bidding data review and processing method for this application;

[0070] Figure 2 A schematic diagram of steps for obtaining derivative data corresponding to characteristic data;

[0071] Figure 3 A schematic diagram of the risk warning steps based on data stability;

[0072] Figure 4 It is a schematic diagram of the feature data verification steps;

[0073] Figure 5 A flowchart of the mutually exclusive relationship verification analysis of the tender award results for bidders;

[0074] Figure 6 It is a schematic diagram of the verification result output based on the verification rules;

[0075] Figure 7 This is a schematic diagram of the framework of the bidding data review and processing system for this application.

[0076] Figure numerals: 1. Data acquisition module; 2. Data derivation module; 3. Data floating analysis module; 4. Data verification output module; 5. Data storage module. DETAILED DESCRIPTION

[0077] The present invention will be further described in detail below in conjunction with embodiments and drawings, but the embodiments of the present invention are not limited thereto.

[0078] In the description of this specification, the description with reference to the terms "certain embodiments", "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiments or examples are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0079] A bidding data review processing method, such as Figure 1 As shown, it mainly includes the following steps:

[0080] S100, obtaining data information associated with each bidder and storing it in a first database;

[0081] S200, parsing and acquiring feature data in the data information or capturing feature data based on user settings, acquiring derivative data corresponding to the feature data according to the feature data, associating the feature data, derivative data and bidder information and storing them as a second database;

[0082] S300, based on historical bidding data and according to set judgment rules, the characteristic data is divided into fixed data and floating data, and floating rules and influencing factors are automatically or manually configured for each floating data, and associated and stored as a third database;

[0083] S400, setting and storing a bidding data verification model;

[0084] S500, setting and / or automatically determining the input value of the current influencing factor, determining the value of the current floating data in combination with the floating rule, inputting the current feature data to be verified and its derivative data into the data verification model, and outputting the verification result based on the built-in rules of the model.

[0085] In the above step S100, the data information associated with the bidder includes but is not limited to the bidder's equity information data, the bidder's former name data, the bidder's credit data, historical bidding information data, and market policy data related to the bidder's qualifications and production capacity. The above data information is obtained by setting an API interface or crawler software. In the implementation mode of the present application, the system's data interface is connected to credit reporting websites such as Credit China, the National Enterprise Information Publicity System, and Tianyancha to obtain information data such as the credit behavior record, legal representative, shareholders and holding company of the designated bidder. At the same time, industry blacklists, corporate blacklists and other information are also obtained through industry websites, which is convenient for the later establishment of a bad supplier information database, forming a supplier credit database, and supporting bid opening and evaluation data verification.

[0086] In step S200, the definition of feature data is: data that can be applied to the bidding data review, such as bidder name, organization code, shareholder information, corporate executive information, specific business data, etc. There are two main ways to obtain feature data in step S200. One is to automatically filter out data that may be needed for the current project type through historical bidding information data, using big data and artificial intelligence technology. For example, if corporate executive information data is used in the bidding review of the same type of project, the system automatically captures the corresponding executive information from the collected data information; the second is to capture feature data in the data information based on user settings. For example, if the user specifies to capture corporate supervisor information, the system captures the corresponding data from the collected data information as feature data according to the specified content.

[0087] In actual applications, different data sources usually have different names for the same type of things. Even for historical bidding data, the same structural parts may have different names in different bidding documents. Obviously, if fixed standards are used to filter data information, data omissions will occur, affecting the quality of later reviews. Therefore, in step S200, data derivation needs to be performed based on the current feature data.

[0088] In detail, in actual applications, derived data at least include: aliases of nouns, former names, synonyms of nouns, conversion of physical units or monetary units, abbreviations of nouns, English translations of nouns, insertion of specific separators in vocabulary or paragraphs, or conclusions involving calculation formulas. For example, if Company A used the former name Company B in historical bidding data information, then "Company B" is used as the derived data of "Company A"; the derived data of deacons include managers, supervisors, etc.; if the quotation unit in the relevant data source is US dollars, the system will automatically convert it into RMB quotation according to the current exchange rate and mark it, and the RMB quotation will be used as the derivative data of the original feature data; for example, if the original feature data is "the area of ​​​​A plate is 3m*3m", then the derived data can be set as "the area of ​​​​A plate is 9m 2 (3m*3m)", so that the bidding team can compare the sizes of different plates. By deriving the above characteristic data, the review data can be made more comprehensive, helping the bidding reviewers to quickly make more accurate review results.

[0089] In the implementation mode of the present application, the derivative data corresponding to the characteristic data is obtained, such as Figure 2 As shown, further comprising:

[0090] S210, based on historical bidding data information and combined with deep learning algorithms, establish a derivative database and / or semantic model corresponding to feature data in various bidding scenarios;

[0091] S220, sorting each derivative data according to the correlation between each derivative data and the characteristic data and marking the application risk;

[0092] S230, determining the current bidding scenario based on the current bidding project name, searching for associated derivative data from a corresponding derivative database based on the captured feature data, and / or directly generating associated derivative data via a corresponding semantic model.

[0093] Since the accuracy of the derived data cannot be guaranteed to match the original data information, in the above step S220, each derived data is marked with a corresponding risk value. The derivative data that is closest to the original feature data has a lower risk value, and vice versa. Later, the bidding review staff can review the relevant data based on the above risk value to avoid data errors and omissions, and improve the quality of bidding data review.

[0094] In actual applications, relevant feature data often changes. For example, a bidder is currently subject to administrative penalties, but the above administrative penalties will end after a set time. Obviously, it is not feasible to continue to use the current bidder's administrative penalty status; similarly, the company's operating data or construction costs will also change with the fluctuations in market raw material prices and equipment prices. In order to adapt to the above situation, in the implementation of the present application, as shown in step S300, according to the set judgment rules, such as the stability of the data, the collected feature data is divided into fixed data and floating data. The above division method can be set manually in advance, or it can be automatically completed based on historical big data.

[0095] In practical applications, common floating data include characteristic data that changes with time, business behavior or market policy factors, such as the repayment time of corporate loans, the time limit of administrative penalties, the expected operating income of enterprises, and policy restrictions on a certain industry or business behavior. Obviously, due to the fluctuation of characteristic data, the accuracy of bidding review will be affected, which is very likely to cause data errors and misjudgments. Figure 3 As shown, the bidding data review processing method described in this application also includes a risk warning step based on data stability:

[0096] B100, based on historical bidding data and market policy big data, analyzes the stability of each feature data;

[0097] B200, according to the stability of each characteristic data and its relevance to the bidding and procurement project, configure a corresponding risk value for each characteristic data;

[0098] B310, calculate the sum of the risk values ​​of each characteristic data of the bidder, and output warning data if it exceeds the set value; or

[0099] B320, classifies each characteristic data based on the implementation steps of the bidding and procurement project, and calculates the sum of the risk values ​​of each type of characteristic data. If it exceeds the set value, the alarm data is output.

[0100] The above technical solution can effectively remind bidding review personnel to review key feature data or certain types of data of bidding projects, thereby improving review efficiency, avoiding data errors and omissions, and improving review quality.

[0101] In the implementation mode of the present application, the characteristic data to be verified mainly include: the credit behavior data of the bidder, the equity relationship data of the bidder, the mutually exclusive relationship data of the bidder's award results, and the relationship data between the bidder and the successful bidder of the same project that has been tendered in the early stage.

[0102] In step S400, the built-in rules of the data validation model include at least:

[0103] (1) Set data or keywords that affect the bidder's credit rating, and search and match them from the bidder's credit behavior data. For example, corporate credit rating data, whether there are unexpired administrative penalties, whether the bidder is a dishonest executor, etc.

[0104] (2) Based on the shareholder names of each bidder, the bidders are analyzed and traced back to the initial shareholders to determine whether the names of the superior shareholders of each bidder are the same. For example, tracing back from the bidder to the investors and shareholders at all levels will form an inverted tree diagram. By comparing and checking whether the superior shareholders or investors of two bidders are the same, the relationship between the two bidders can be defined. Based on the defined relationship, it can be determined whether there is a possibility of collusion between the bidders.

[0105] (3) Analyze the current bidder's award based on historical bidding data information, and find out whether the award is mutually exclusive with the award of the historical bidding and procurement winning situation. For example, Company A is one of the investors of Company B, and its bid package is a construction package, while Company B's bid package is a design or acceptance service. In actual application, it is necessary to avoid the above-mentioned Company A and Company B participating in the above-mentioned bidding project at the same time.

[0106] (4) Based on historical bidding data, obtain the data information of previous successful bidders for the same project, and combine the bidders' equity relationship data to determine the relationship between the current bidder and the previous successful bidders for the same project.

[0107] (5) Count the risk values ​​and their sums corresponding to each feature data to determine whether they exceed the set threshold.

[0108] In practical applications, in order to improve the review efficiency, Figure 4 As shown, the bidding data review processing method also includes:

[0109] C100, establish the association relationship between each type of project and the priority of the validation rules and sort the validation rules according to the priority of each validation rule;

[0110] C200, establish a risk database for storing the verification results of the feature data to be verified;

[0111] C300, obtain the feature data to be verified and input it into the data verification model, verify the feature data one by one based on the above sorting, and if it does not meet the current verification rules: mark the feature data and / or terminate the verification process and store it in the risk database;

[0112] If the current validation rules are met, the next data validation will be performed until the validation is completed.

[0113] like Figure 5 The figure shows the process diagram of the verification and analysis of the mutually exclusive relationship of the bid award results of the bidder. Figure 6 It is a schematic diagram of the verification results based on the output of the verification rules.

[0114] In order to achieve the interconnection and horizontal sharing of procurement data between various nodes in the bidding process, and promote the coordinated development of bidding procurement data analysis and verification capabilities at each node, the bidding data review and processing method described in this application also includes:

[0115] D100 associates and stores the bidding nodes and their data sharing relationships in the same project directory. The above data sharing relationships include the transmission of specific data between bidding nodes.

[0116] D200, determine the data sharing relationship between each bidding node and configure the sharing authority of data sharing. In the above sub-steps, configuring authority for bidding reviewers can effectively avoid the overflow of non-current bidding data information and ensure data security.

[0117] D300, based on the data sharing relationship between the above-mentioned bidding nodes, automatically or manually sets the follow-up update mechanism of data information between the bidding nodes. For example, important feature data can be set to be automatically updated, while ordinary feature data can be set to be manually updated.

[0118] D400 automatically queries the bidding nodes associated with the current bidding node data information, and retrieves the data information of the associated bidding nodes based on the shared authority verification result. When the data information of one of the bidding nodes changes, the data information of the bidding nodes associated with the current bidding node is automatically updated based on the above-mentioned follow-up update mechanism.

[0119] By analyzing the historical bidding data, it can be seen that for different bidders, due to their own specific differences in operation and management, the relevant parameters in similar bidding projects will usually show regular differences. For example, due to the problem of operating costs, the quotation schemes of Company A and Company B in similar bidding projects will show a certain proportional relationship. The above proportional relationship is actually a reflection of the company's operating costs and can be regarded as an inherent attribute of the bidder. In order to further enhance the accuracy of data information, the method described in the implementation mode of this application also includes:

[0120] E100, based on the current bidding project type, obtains the bidding proposals of each bidder from the historical bidding data, compares the differences between the bidding proposals of each bidder based on the set rules, and generates a first relationship library for characterizing the differences between the bidding proposals of each bidder, wherein the above-mentioned differences include the quotation of the proposal, the implementation period of the proposal, etc.

[0121] E200, obtaining the bidding proposals of each bidder in the current bidding project, and comparing the differences between the bidding proposals of each bidder based on the set rules, to generate a second relationship database for characterizing the differences between the bidding proposals of each bidder in the current project of the same type.

[0122] The above-mentioned set rules include but are not limited to: obtaining and comparing the price difference or ratio relationship between the quotation proposals of two bidders; obtaining and comparing the difference or ratio relationship between the job levels and quantities of personnel in the specific project teams of the two bidders; obtaining and comparing the difference or ratio relationship between the labor periods of two bids, etc.

[0123] E300, if the difference between the specific data in the first relational database and the second relational database exceeds the set value, for example, the difference between the quotations of Company A and Company B in the historical early bidding data is X, and the difference between the quotations of Company A and Company B in the current bidding project is Y, if the difference between X and Y is greater than the set value, an alarm is output. The above-mentioned specific data includes: the ratio of the quotations of the bidding schemes of each bidder, the ratio of the implementation time of the bidding schemes of each bidder, and the ratio between the risk values ​​of the bidding data.

[0124] Based on the above technical solution, big data analysis can be used to obtain the differences between bidders in the bidding process of similar projects in history. By analyzing and judging the above differences, it can be inferred whether there are data anomalies or collusion among multiple bidders in the bidding process of the current project, which will help the bidding team reviewers make accurate judgments on relevant data information, reduce review risks, and improve review quality.

[0125] A bidding data review and processing system, such as Figure 7As shown, it mainly includes: a data acquisition module 1, a data derivation module 2, a data floating analysis module 3, a data verification output module 4 and a data storage module 5.

[0126] The data acquisition module 1 is configured to connect to the external data source data through a set API interface, or to obtain the data information of each bidder through data crawler software. This application system interfaces with credit websites such as Credit China, National Enterprise Information Publicity System, and Tianyancha to obtain information such as the credit behavior record, legal representative, shareholders, and holding companies of the designated bidder. In practice, the above-mentioned data acquisition module 1 is configured or externally connected with a data format conversion program, which is used to convert source data with different data formats or different naming rules into a set format to facilitate the processing of later data.

[0127] The data derivation module 2 is configured to be data-connected with the data storage module 5 or directly with the data acquisition module 1, and is used to parse and obtain or capture the characteristic data in the data information based on the user setting, and obtain the derivative data corresponding to the characteristic data according to the characteristic data. The derivation rules are as described above, including synonym annotation, unit conversion, calculation result output, etc. The data floating analysis module 3 is configured to divide the characteristic data into fixed data and floating data based on the historical bidding data and according to the set judgment rules, and automatically or manually configure floating rules and influencing factors for each floating data. The above-mentioned judgment rules include extracting and comparing similar items in the historical bidding data to determine whether a certain data has changed in several bidding processes. For example, the raw material quotation value usually changes over time, which is classified as floating data, and the bidder name generally does not change and is defined as fixed data. The above-mentioned floating rules, that is, the change rules of floating data, different floating data correspond to different rules. For example, there is a specific functional relationship between the cost of the product and the price of the raw materials. The above-mentioned functional relationship is the floating rule of the characteristic data of the product cost.

[0128] The data verification output module 4 has built-in data verification rules, is data-connected to the data storage module 5 and the data derivation module 2, and is configured to determine the value of the current floating data based on the input value of the current influencing factor in combination with the floating rule, input the current feature data to be verified and its derivative data into the data verification model, and output the verification result based on the built-in rules of the model. In order to achieve the wide applicability of the data results, in the embodiment of the present application, the data output end of the data verification output module 4 is connected to a format conversion program for outputting the verification results in different data formats, such as EXCEL tables, PDF, etc.

[0129] The data storage module 5 is configured as a plurality of databases, which are respectively used to store data information associated with each bidder, the characteristic data, the derived data and the associated bidder information, and the floating rules and influencing factors corresponding to each floating data. The data information includes bidder equity information data, bidder former name data, bidder credit data, historical bidding information data, and market policy data associated with bidder qualifications and production capacity.

[0130] In the implementation manner of the present application, the data storage module 5 stores at least the following data:

[0131] (1) Collect historical information on winning bids for various types of bidding and procurement projects, and associate project types, procurement bid package information with the project catalog to form a procurement project process information database based on the project catalog and procurement content. The winning bid results form a historical winning bid record database.

[0132] (2) Based on the external enterprise data platform, establish a bidder equity information database to record the bidder equity penetration information.

[0133] (3) Based on the external enterprise data platform and the bidder management information form, establish a bidder's former name information database to record the bidder's former name management list.

[0134] (4) Based on the bid package purchase list, a record of bidders’ bid document purchases is formed to support data analysis and verification, and to provide a data basis for future bidder behavior analysis.

[0135] (5) Based on information from external enterprise data platforms, Credit China, enterprise disclosure systems, industry blacklists, enterprise blacklists, etc., establish a bad supplier information database and form a supplier credit database.

[0136] In order to achieve the interconnection and interoperability of important data, the bidding data review and processing system described in this application also includes: a data sharing relationship storage module and a data automatic update module.

[0137] The data sharing relationship storage module is configured to associate and store the various bidding nodes in the same project directory and their data sharing relationships and sharing permissions, so as to facilitate data sharing between personnel and nodes with the same level of permissions. The data automatic update module is internally configured with a set follow-up update mechanism, which is configured to automatically query the bidding nodes associated with the current bidding node data information, and retrieve the data information of the associated bidding nodes based on the sharing permission verification results. When the data information of one of the bidding nodes changes, the data information of the bidding node associated with the current bidding node is automatically updated based on the above-mentioned follow-up update mechanism, which can avoid data confusion and remind the reviewers of the relevant nodes to pay attention.

[0138] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the bidding data review and processing method as described above.

[0139] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of a computer device, and uses various interfaces and lines to connect the various parts of the entire computer device.

[0140] The storage medium described in the present application may include a high-speed random access memory, and may also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. In practice, it can be used to store computer programs and / or modules. The processor implements various functions of the computer device by running or executing the computer programs and / or modules stored in the storage medium and calling the data stored in the memory.

[0141] It should be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, servers, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code.

[0142] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A bidding data review and processing method, characterized in that: include: Obtaining data information associated with each bidder and storing it in a first database; Analyze and obtain characteristic data in the data information or capture it based on user settings, obtain derivative data corresponding to the characteristic data according to the characteristic data, associate the characteristic data, the derivative data and the bidder information and store them as a second database; Based on historical bidding data and according to set judgment rules, the characteristic data is divided into fixed data and floating data, and floating rules and influencing factors are automatically or manually configured for each floating data, and associated and stored as a third database; Set up and store bid data validation models; Set and / or automatically determine the input value of the current influencing factor, determine the value of the current floating data in combination with the floating rule, input the current feature data to be verified and its derivative data into the data verification model, and output the verification result based on the built-in rules of the model; Among them, the data information includes the bidder's equity information data, the bidder's former name data, the bidder's credit data, historical bidding information data, and market policy data related to the bidder's qualifications and production capacity.

2. The bidding data review and processing method according to claim 1, characterized in that: The derived data include: synonyms of nouns, former names, synonyms of nouns, conversion of physical units or monetary units, abbreviations of nouns, English translations of nouns, insertion of specific separators in words or paragraphs, or conclusions involving calculation formulas; Obtain derivative data corresponding to the feature data, including: Based on historical bidding data information and combined with deep learning algorithms, a derivative database or semantic model corresponding to the feature data in various bidding scenarios is established, and each derivative data is sorted and marked for application risks according to the correlation between each derivative data and the feature data; The current bidding scenario is determined based on the current bidding project name, and related derivative data is searched from the corresponding derivative database based on the captured feature data, or related derivative data is directly generated through the corresponding semantic model.

3. The bidding data review and processing method according to claim 2, characterized in that: The floating data include characteristic data that changes with time, business behavior or market policy factors; The method further comprises: Analyze the stability of each feature data based on historical bidding data and market policy big data; According to the stability of each characteristic data and its relevance to the bidding and procurement project, a corresponding risk value is configured for each characteristic data; Count the sum of the risk values ​​of each characteristic data of the bidder, and output warning data if it exceeds the set value; or Based on the implementation steps of the bidding and procurement project, each feature data is classified, and the sum of the risk values ​​of each type of feature data is calculated. If it exceeds the set value, the alarm data is output.

4. The bidding data review and processing method according to claim 1, characterized in that: The characteristic data to be verified include: the credit behavior data of the bidder, the equity relationship data of the bidder, the mutually exclusive relationship data of the bidder's award results, and the relationship data between the bidder and the successful bidder of the same project that has been tendered in the early stage; The built-in rules of the data validation model include: Set data or keywords that affect the bidder's credit rating, and search and match them from the bidder's credit behavior data; Based on the shareholder names of each bidder, analyze and trace back to the initial shareholder, and find out whether the names of the superior shareholders of each bidder are the same; Analyze the current bidder's award based on historical bidding data information, and find out whether the award and the historical bidding and procurement winning situation are mutually exclusive; Based on historical bidding data, obtain the data information of previous successful bidders of the same project, and combine the bidders' equity relationship data to determine the relationship between the current bidder and the previous successful bidder of the same project; The risk value and sum of each characteristic data are counted to determine whether they exceed the set threshold.

5. The bidding data review and processing method according to claim 4, characterized in that: The method further comprises: Establish the association between each type of project and the priority of the validation rules and sort the validation rules according to the priority of each validation rule; Establishing a risk database for storing verification results of feature data to be verified; Obtain the feature data to be verified and input it into the data verification model, verify the feature data one by one based on the above sorting, and if it does not meet the current verification rules: mark the feature data and / or terminate the verification process and store it in the risk database; If the current validation rules are met, the next data validation will be performed until the validation is completed.

6. The bidding data review and processing method according to claim 2, characterized in that: The method further comprises: Associate and store various bidding nodes in the same project directory and their data sharing relationships; Determine the data sharing relationship between each bidding node and configure the sharing permissions for data sharing; Based on the data sharing relationship between the above bidding nodes, automatically or manually set the follow-up update mechanism of data information between the bidding nodes; Automatically query the bidding nodes associated with the current bidding node data information, and retrieve the data information of the associated bidding nodes based on the shared permission verification results; When the data information of one of the bidding nodes changes, the data information of the bidding nodes associated with the current bidding node is automatically updated based on the above-mentioned follow-up update mechanism.

7. The bidding data review and processing method according to claim 3, characterized in that: The method further comprises: Based on the current bidding project type, the bidding schemes of each bidder are obtained from the historical bidding data, and the differences between the bidding schemes of each bidder are compared pairwise based on the set rules to generate a first relationship database for representing the differences between the bidding schemes of each bidder; Obtain the bidding schemes of each bidder in the current bidding project, and compare the differences between the bidding schemes of each bidder based on the set rules, so as to generate a second relationship database for characterizing the differences between the bidding schemes of each bidder in the current project of the same type; If the difference between the specific data in the first relational database and the second relational database exceeds a set value, an alarm is output; Among them, the above-mentioned specific data include: the ratio of the bid prices of each bidder's bidding proposals, the ratio of the implementation time of each bidder's bidding proposals, and the ratio between the risk values ​​of the bidding data.

8. A bidding data review and processing system, characterized in that: include: A data acquisition module (1) is configured to connect to an external data source through a set API interface or to acquire data information of each bidder through a data crawler software; A data derivation module (2) is configured to parse and obtain or capture characteristic data in the data information based on user settings, and obtain derivative data corresponding to the characteristic data based on the characteristic data; A data floating analysis module (3) is configured to divide the characteristic data into fixed data and floating data based on historical bidding data and according to set judgment rules, and automatically or manually configure floating rules and influencing factors for each floating data; A data verification output module (4) is built with data verification rules, and is configured to determine the value of the current floating data according to the input value of the current influencing factor in combination with the floating rule, input the current feature data to be verified and its derivative data into the data verification model, and output the verification result based on the built-in rules of the model; The data storage module (5) is configured as a plurality of databases, which are respectively used to store data information associated with each bidder, the characteristic data, the derived data and the associated bidder information, and the floating rules and influencing factors corresponding to each floating data; Among them, the data information includes the bidder's equity information data, the bidder's former name data, the bidder's credit data, historical bidding information data, and market policy data related to the bidder's qualifications and production capacity.

9. The bidding data review and processing system according to claim 8, characterized in that: The system further comprises: A data sharing relationship storage module is configured to associate and store various bidding nodes in the same project directory and their data sharing relationships and sharing permissions; The data automatic update module has a built-in follow-up update mechanism, which is configured to automatically query the bidding nodes associated with the current bidding node data information, and retrieve the data information of the associated bidding nodes based on the shared permission verification results; When the data information of one of the bidding nodes changes, the data information of the bidding nodes associated with the current bidding node is automatically updated based on the above-mentioned follow-up update mechanism.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that When the computer program is executed by a processor, the bidding data review and processing method as described in any one of claims 1 to 7 is implemented.

Images