Intelligent selection method and system of bidding information based on data mining technology

Abstract

The present invention discloses a method and system for intelligently screening bidding information based on data mining technology, and relates to the field of big data technology. When the system is running, the project requirements of the enterprise are accurately extracted through natural language processing technology, and the demand information is automatically structured and processed to form a demand data group. Secondly, the system can efficiently parse the bidding documents, extract key information such as technical requirements, budgets, execution cycles and risk parameters, and form a bidding data group. Then, the matching degree calculation module comprehensively compares the project requirements with the bidding information by establishing an intelligent matching algorithm, generates a matching degree score MTC, and automatically screens the bidding projects that meet the needs of the enterprise. At the same time, the system generates a comprehensive secondary evaluation index SC based on historical data, and sorts all bidding projects. Through iterative optimization capabilities, by collecting actual bidding results and project execution feedback, the screening process is continuously optimized to ensure that the screening results are continuously improved.

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and in particular to a method and system for intelligently screening bidding information based on data mining technology. Background Art

[0002] In the invention of Chinese patent application number CN202311403479.9, a method, device, equipment and storage medium for processing the demand category of a tender document are disclosed. The method includes: obtaining at least one bidding text combination corresponding to at least two bidding data, obtaining the response matching degree of the bidding text in each bidding text combination to the bidding text, obtaining the bidding text similarity between the bidding texts corresponding to the two bidding data for any two bidding data, determining the bidding demand correlation between the bidding texts corresponding to the two bidding data according to at least one response matching degree corresponding to the two bidding data and at least one bidding text similarity corresponding to the two bidding data, and clustering each bidding text in each bidding data based on at least one bidding demand correlation to obtain the bidding demand category of the bidding text cluster set. The embodiment of the present invention solves the problem that the text similarity algorithm and the bidding demand text have low matching degree.

[0003] It can be seen from this that analyzing all the bidding information will greatly increase the processing pressure and time, and it is impossible to determine whether the currently processed bidding information files meet the project requirements. If the bidding information that does not meet the requirements is processed and analyzed, it will undoubtedly be ineffective work. In the face of complex bidding documents and their technical requirements, budget and risk information, it is not possible to give priority to judging whether the bidding information meets the requirements. It is also a prerequisite that needs to be solved urgently. Failure to parse a large amount of bidding information to match the bidding information that meets the project will lead to the possibility of invalid work in subsequent work. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a method and system for intelligently screening bidding information based on data mining technology, which solves the problems mentioned in the background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a bidding information intelligent screening system based on data mining technology, including a demand analysis module, a bidding information analysis module, a matching degree calculation module, a scoring and sorting module and an iterative optimization module;

[0006] The demand analysis module extracts demand information from the enterprise's project demand document through natural language processing technology and structures it, including technical requirements TRQ, budget range BRQ and project cycle DRQ, to form a demand data group XZ;

[0007] The bidding information parsing module parses the received bidding information file and extracts information related to technical requirements, budget, project cycle, funding source and additional terms, and obtains extracted technical requirements TBD, extracted budget BBD, extracted execution cycle DBD and extracted risk parameters R through integration to form a bidding data group ZZ;

[0008] The matching degree calculation module establishes a matching degree evaluation algorithm for the demand data group XZ and the bidding data group ZZ, generates a matching degree score MTC between the bidding information file and the project demand, and matches it with a preset project demand matching bidding evaluation threshold PZ to obtain a matching status result of the current bidding information file;

[0009] The scoring and sorting module combines the acquired tender information file with historical data according to the matching status result of the current tender information file, generates a comprehensive secondary evaluation index SC of the tender information, and sorts the comprehensive tender data group ZHZ of the comprehensive secondary evaluation index SC of the tender information;

[0010] The iterative optimization module verifies the sorted comprehensive bidding data group ZHZ bidding information, collects feedback information, including the bidding results and project execution feedback information, and regulates the matching score MTC and the comprehensive secondary evaluation index SC generated by the matching evaluation algorithm.

[0011] Preferably, the demand analysis module includes a natural language processing unit and a data structuring unit;

[0012] The natural language processing unit parses and extracts demand information from the enterprise's project demand document by natural language processing technology. The parsing of the project demand document includes dividing the project demand document into different logical parts, including a technical requirement part, a budget part and a cycle part, and then extracting demand information by using keyword extraction and regular expression technology, including extracting technical requirement information, budget range information and project cycle information in the project demand;

[0013] The data structuring unit performs structured processing on the extracted technical requirement information, budget range information and project cycle information. The result processing includes extracting the frequency of occurrence of different information contents, including technical requirements TRQ, budget range BRQ and project cycle DRQ, to form a demand data group XZ, ​​specifically XZ={TRQ, BRQ, DRQ}.

[0014] Preferably, the bidding information parsing module includes an information extraction unit and a data integration unit;

[0015] The information extraction unit parses the received bidding information file, the parsing includes classifying according to different parts, and extracting technical requirements, budget, project cycle, funding sources and additional terms related information, and using regular expressions to obtain relevant numerical information of the extracted technical requirements, budget, project cycle, funding sources and additional terms related information, the data information includes the frequency of occurrence of different technical requirements, the frequency of occurrence of different budgets, the frequency of occurrence of different project cycles and the frequency of occurrence of different additional terms related information;

[0016] The data integration unit extracts technical requirements TBD, budget BBD, execution period DBD and risk parameters R by integrating relevant numerical information to form a bidding data group ZZ, specifically ZZ={TBD, BBD, DBD, R}.

[0017] Preferably, the matching degree calculation module includes a matching evaluation unit and a state detection unit;

[0018] The matching evaluation unit establishes a matching evaluation algorithm for the demand data group XZ and the bidding data group ZZ to generate a matching score MTC between the bidding information file and the project demand;

[0019] The matching score MTC is obtained by the following calculation formula:

[0020] ;

[0021] Where SIMtech represents the matching degree of technical requirements, SIMbudget represents the matching degree of budget, SIMcycle represents the matching degree of execution cycle, log represents the logarithmic function, R represents the extracted risk parameter, m1, m2 and m3 represent the preset weight values ​​of the matching degree of technical requirements SIMtech, the matching degree of budget SIMbudget and the matching degree of execution cycle SIMcycle respectively, and , , , , the specific value is set by the user;

[0022] SIMtech obtains the matching degree of the technical requirements through the following statistical algorithm formula:

[0023] ;

[0024] In the formula, Indicates the intersection of technical requirements TRQ and extraction technology TBD requirements, specifically indicating the same technical requirements information, It represents the union of the technical requirement TRQ and the extracted technical requirement TBD, specifically it represents the sum of the technical requirement TRQ and the extracted technical requirement TBD;

[0025] The budget matching degree SIMbudget is obtained by the following statistical algorithm formula:

[0026] ;

[0027] In the formula, BRQ represents the budget range, and BBD represents the extracted budget;

[0028] The matching degree of the execution cycle SIMcycle:

[0029] ;

[0030] In the formula, DRQ represents the project cycle and DBD represents the extraction execution cycle.

[0031] Preferably, the state detection unit matches the matching score MTC with a preset project demand matching bidding evaluation threshold PZ to obtain a matching state result of the current bidding information file;

[0032] The matching status result of the current bidding information file is obtained by the following matching method:

[0033] ;

[0034] When the matching return signal status of the matching status result of the current bidding information file is 1, the matching result is obtained, and the matching result of the current bidding information file is marked;

[0035] When the matching return signal status of the matching status result of the current bidding information file is 0, the unmatched result is obtained, and the unmatched result is marked on the current bidding information file.

[0036] Preferably, the scoring and ranking module includes a secondary evaluation unit and an integration unit;

[0037] The secondary evaluation unit combines the acquired tender information file with historical data according to the matching status result of the current tender information file, including the historical tender information execution records and winning records of the company from which the current tender information file comes, and simultaneously performs statistical analysis to obtain the historical score HS and winning rate SR of the tender information file execution of the company from which the current tender information comes, and then performs fitting with the matching score MTC to generate a comprehensive secondary evaluation index SC of the tender information;

[0038] The comprehensive secondary evaluation index SC is obtained by the following calculation formula:

[0039] ;

[0040] The historical score HS is obtained by statistical analysis through the following calculation formula:

[0041] ;

[0042] Where Q represents the project delivery quality status, specifically whether there are quality problems after the project is delivered, and the area with quality problems is compared with the total project construction area to obtain the delivery quality area ratio information, C represents the cost control score, specifically the proportion of the project budget exceeded when the project is delivered, T represents the project delivery cycle score, specifically the proportion of the time delayed when the project is delivered, h1, h2 and h3 represent the preset weight values ​​of the project delivery quality status Q, cost control score and project delivery cycle score T respectively, and , , ,and , the specific value is set by the user;

[0043] The winning bid rate SR is obtained by the following calculation formula:

[0044] ;

[0045] Where ZBL represents the number of projects that won bids within a fixed period, and TBZS represents the total number of projects that bid within a fixed period.

[0046] Preferably, the integration unit marks and associates all the tender information according to the comprehensive secondary evaluation index SC of the tender information to form a comprehensive tender data group ZHZ, and performs bubble sorting;

[0047] Specifically, the bubble sort compares the results of the i-th comprehensive secondary evaluation index SC and the i+1-th comprehensive secondary evaluation index SC in the comprehensive bidding data group ZHZ to exchange positions, and repeats the exchange until there is no more exchange of positions, and stops the bubble sort.

[0048] Preferably, the iterative optimization module includes a feedback collection unit and an algorithm control unit;

[0049] The feedback collection unit verifies the sorted comprehensive bidding data group ZHZ bidding information and collects feedback information, including bidding results and project execution feedback information;

[0050] The collection of the bid result feedback information includes verifying the top five projects in the sorted comprehensive bidding data group ZHZ, and recording the bid results of the projects; if the bid result is a successful bid, the bidding data group ZZ of the winning project is recorded; if the bid result is a failure to win the bid, the reasons for the failure of the top five projects are recorded, and the specific parameter information affecting the failure to win the bid in the bidding data group ZZ is marked, and the algorithm control unit is triggered to execute;

[0051] Project execution feedback information includes collecting project delivery quality status Q, cost control score C and project delivery cycle score T.

[0052] Preferably, the algorithm control unit is triggered to control the matching score MTC and the comprehensive secondary evaluation index SC generated by the matching evaluation algorithm according to the winning results of the top five projects, and the control includes proportional adjustment of m1, m2, m3, h1, h2 and h3;

[0053] Among them, the trigger includes triggering the regulation of the matching score MTC and the comprehensive secondary evaluation index SC generated by the matching evaluation algorithm when the winning bid result is a winning bid.

[0054] The intelligent screening method of bidding information based on data mining technology includes the following steps:

[0055] Step 1: The demand analysis module extracts demand information from the enterprise's project demand document through natural language processing technology and structures it, including technical requirements TRQ, budget range BRQ and project cycle DRQ, to form a demand data group XZ;

[0056] Step 2: The bidding information parsing module parses the received bidding information file and extracts the technical requirements, budget, project cycle, funding sources and additional terms. After integration, the extracted technical requirements TBD, extracted budget BBD, extracted execution cycle DBD and extracted risk parameters R are obtained to form the bidding data group ZZ;

[0057] Step 3: The matching degree calculation module establishes a matching degree evaluation algorithm for the demand data group XZ and the bidding data group ZZ, generates a matching degree score MTC between the bidding information file and the project demand, and matches it with the preset project demand matching bidding evaluation threshold PZ to obtain the matching status result of the current bidding information file;

[0058] Step 4: The scoring and sorting module combines the acquired bidding information file with historical data according to the matching status result of the current bidding information file, generates a comprehensive secondary evaluation index SC of the bidding information, and sorts the comprehensive bidding data group ZHZ of the comprehensive secondary evaluation index SC of the bidding information;

[0059] Step 5: The iterative optimization module verifies the sorted comprehensive bidding data group ZHZ bidding information and collects feedback information, including the winning bid results and project execution feedback information, and regulates the matching score MTC and the comprehensive secondary evaluation index SC generated by the matching evaluation algorithm.

[0060] The present invention provides a bidding information intelligent screening method and system based on data mining technology, which has the following beneficial effects:

[0061] (1) When the system is running, it accurately extracts the project requirements of the enterprise through natural language processing technology, automatically structures and processes the demand information, and forms a demand data group. Secondly, the system can efficiently parse the bidding documents, extract key information such as technical requirements, budget, execution cycle and risk parameters, and form a bidding data group. Then, the matching calculation module establishes an intelligent matching algorithm to comprehensively compare the project requirements with the bidding information, generate a matching score MTC, and automatically screen the bidding projects that meet the needs of the enterprise. At the same time, the system combines historical data to generate a comprehensive secondary evaluation index SC, and sorts all bidding projects to help enterprises give priority to high-potential projects. Finally, through iterative optimization capabilities, by collecting actual bidding results and project execution feedback, the matching score MTC and the comprehensive secondary evaluation index SC are dynamically adjusted for regulation, and the screening process is continuously optimized to ensure that the screening results are continuously improved.

[0062] (2) By introducing the matching evaluation unit and the status detection unit, the matching score MTC between the project requirements and the bidding information is accurately calculated, and through intelligent evaluation of multiple dimensions such as technical requirements, budget, and cycle, it ensures that the selected bidding information is highly in line with the needs of the enterprise. The scoring and sorting module conducts a secondary evaluation by combining historical data, generates a comprehensive secondary evaluation index SC, and accurately sorts the bidding information through bubble sorting. The system improves the quantitative evaluation of the supplier's past performance by combining the historical score HS and the winning rate SR, making the screening process more comprehensive and rigorous. At the same time, bubble sorting ensures that high-potential projects are presented first, allowing enterprises to quickly locate the best bidding opportunities. The system's automation and data-driven capabilities effectively reduce manual screening errors, greatly improve the intelligence of screening, and provide refined matching evaluation and comprehensive sorting functions, enabling enterprises to make efficient decisions in complex bidding environments.

[0063] (3) The feedback collection unit and algorithm control unit in the iterative optimization module ensure dynamic optimization and continuous improvement of the screening process. The system not only verifies the winning results of the top five projects, but also identifies potential problems by collecting the reasons for not winning the bid and specific influencing factors, and triggers algorithm control to regulate the matching score MTC and comprehensive secondary evaluation index SC generated by the matching evaluation algorithm. At the same time, the system collects detailed execution feedback information of the winning projects, including project quality, cost control and delivery cycle, so as to dynamically adjust the scoring weights in the next screening. Through this mechanism, the system can continuously learn from historical results and execution performance, gradually optimize the matching score MTC and comprehensive secondary evaluation index SC, and ensure that subsequent evaluation results are more accurate. Compared with static algorithms, this iterative optimization process greatly improves the flexibility and adaptability of the system, providing enterprises with more accurate and real-time updated bidding information screening and evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 It is a schematic diagram of a block diagram of a bidding information intelligent screening system based on data mining technology of the present invention;

[0065] Figure 2 The figure is a schematic diagram of the steps of the method for intelligently screening bidding information based on data mining technology of the present invention. DETAILED DESCRIPTION

[0066] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0067] Embodiment 1:

[0068] The present invention provides a bidding information intelligent screening system based on data mining technology, please refer to Figure 1 , including demand analysis module, bidding information analysis module, matching degree calculation module, scoring and sorting module and iterative optimization module;

[0069] The demand analysis module extracts demand information from the enterprise's project demand document through natural language processing technology and structures it, including technical requirements TRQ, budget range BRQ and project cycle DRQ, to form a demand data group XZ;

[0070] The bidding information parsing module parses the received bidding information file and extracts information related to technical requirements, budget, project cycle, funding source and additional terms, and obtains extracted technical requirements TBD, extracted budget BBD, extracted execution cycle DBD and extracted risk parameters R through integration to form a bidding data group ZZ;

[0071] The matching degree calculation module establishes a matching degree evaluation algorithm for the demand data group XZ and the bidding data group ZZ, generates a matching degree score MTC between the bidding information file and the project demand, and matches it with a preset project demand matching bidding evaluation threshold PZ to obtain a matching status result of the current bidding information file;

[0072] The scoring and sorting module combines the acquired tender information file with historical data according to the matching status result of the current tender information file, generates a comprehensive secondary evaluation index SC of the tender information, and sorts the comprehensive tender data group ZHZ of the comprehensive secondary evaluation index SC of the tender information;

[0073] The iterative optimization module verifies the sorted comprehensive bidding data group ZHZ bidding information, collects feedback information, including the bidding results and project execution feedback information, and regulates the matching score MTC and the comprehensive secondary evaluation index SC generated by the matching evaluation algorithm.

[0074] In this embodiment, the project requirements of the enterprise are accurately extracted through natural language processing technology, and the demand information is automatically structured and processed to form a demand data group. Secondly, the system can efficiently parse the bidding documents, extract key information such as technical requirements, budget, execution cycle and risk parameters, and form a bidding data group. Then, the matching calculation module establishes an intelligent matching algorithm to comprehensively compare the project requirements with the bidding information, generate a matching score MTC, and automatically screen the bidding projects that meet the needs of the enterprise. At the same time, the system combines historical data to generate a comprehensive secondary evaluation index SC, and sorts all bidding projects to help companies give priority to high-potential projects. Finally, through iterative optimization capabilities, by collecting actual bidding results and project execution feedback, dynamically adjust the matching score MTC and the comprehensive secondary evaluation index SC for regulation, and continuously optimize the screening process to ensure that the screening results are continuously improved.

[0075] Embodiment 2:

[0076] This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the demand parsing module includes a natural language processing unit and a data structuring unit;

[0077] The natural language processing unit parses and extracts demand information from the enterprise's project demand document by natural language processing technology. The parsing of the project demand document includes dividing the project demand document into different logical parts, including a technical requirement part, a budget part and a cycle part, and then extracting demand information by using keyword extraction and regular expression technology, including extracting technical requirement information, budget range information and project cycle information in the project demand;

[0078] The data structuring unit performs structured processing on the extracted technical requirement information, budget range information and project cycle information. The result processing includes extracting the frequency of occurrence of different information contents, including technical requirements TRQ, budget range BRQ and project cycle DRQ, to form a demand data group XZ, ​​specifically XZ={TRQ, BRQ, DRQ}.

[0079] The bidding information parsing module includes an information extraction unit and a data integration unit;

[0080] The information extraction unit parses the received bidding information file, the parsing includes classifying according to different parts, and extracting technical requirements, budget, project cycle, funding sources and additional terms related information, and using regular expressions to obtain relevant numerical information of the extracted technical requirements, budget, project cycle, funding sources and additional terms related information, the data information includes the frequency of occurrence of different technical requirements, the frequency of occurrence of different budgets, the frequency of occurrence of different project cycles and the frequency of occurrence of different additional terms related information;

[0081] The data integration unit extracts technical requirements TBD, budget BBD, execution period DBD and risk parameters R by integrating relevant numerical information to form a bidding data group ZZ, specifically ZZ={TBD, BBD, DBD, R}.

[0082] Embodiment 3:

[0083] This embodiment is explained in Example 2. Please refer to Figure 1 ,Specifically: the matching degree calculation module includes a matching evaluation unit and a state detection unit;

[0084] The matching evaluation unit establishes a matching evaluation algorithm for the demand data group XZ and the bidding data group ZZ to generate a matching score MTC between the bidding information file and the project demand;

[0085] The matching score MTC is obtained by the following calculation formula:

[0086] ;

[0087] Where SIMtech represents the matching degree of technical requirements, SIMbudget represents the matching degree of budget, SIMcycle represents the matching degree of execution cycle, log represents the logarithmic function, R represents the extracted risk parameter, m1, m2 and m3 represent the preset weight values ​​of the matching degree of technical requirements SIMtech, the matching degree of budget SIMbudget and the matching degree of execution cycle SIMcycle respectively, and , , , , the specific value is set by the user;

[0088] SIMtech obtains the matching degree of the technical requirements through the following statistical algorithm formula:

[0089] ;

[0090] In the formula, Indicates the intersection of technical requirements TRQ and extraction technology TBD requirements, specifically indicating the same technical requirements information, It represents the union of the technical requirement TRQ and the extracted technical requirement TBD, specifically it represents the sum of the technical requirement TRQ and the extracted technical requirement TBD;

[0091] The budget matching degree SIMbudget is obtained by the following statistical algorithm formula:

[0092] ;

[0093] In the formula, BRQ represents the budget range, and BBD represents the extracted budget;

[0094] The matching degree of the execution cycle SIMcycle:

[0095] ;

[0096] In the formula, DRQ represents the project cycle and DBD represents the extraction execution cycle.

[0097] The state detection unit matches the matching score MTC with the preset project demand matching bidding evaluation threshold PZ to obtain the matching state result of the current bidding information file;

[0098] The matching status result of the current bidding information file is obtained by the following matching method:

[0099] ;

[0100] When the matching return signal status of the matching status result of the current bidding information file is 1, the matching result is obtained, and the matching result of the current bidding information file is marked;

[0101] When the matching return signal status of the matching status result of the current bidding information file is 0, the unmatched result is obtained, and the unmatched result is marked on the current bidding information file.

[0102] The scoring and ranking module includes a secondary evaluation unit and an integration unit;

[0103] The secondary evaluation unit combines the acquired tender information file with historical data according to the matching status result of the current tender information file, including the historical tender information execution records and winning records of the company from which the current tender information file comes, and simultaneously performs statistical analysis to obtain the historical score HS and winning rate SR of the tender information file execution of the company from which the current tender information comes, and then performs fitting with the matching score MTC to generate a comprehensive secondary evaluation index SC of the tender information;

[0104] The comprehensive secondary evaluation index SC is obtained by the following calculation formula:

[0105] ;

[0106] The historical score HS is obtained by statistical analysis through the following calculation formula:

[0107] ;

[0108] Where Q represents the project delivery quality status, specifically whether there are quality problems after the project is delivered, and the area with quality problems is compared with the total project construction area to obtain the delivery quality area ratio information, C represents the cost control score, specifically the proportion of the project budget exceeded when the project is delivered, T represents the project delivery cycle score, specifically the proportion of the time delayed when the project is delivered, h1, h2 and h3 represent the preset weight values ​​of the project delivery quality status Q, cost control score and project delivery cycle score T respectively, and , , ,and , the specific value is set by the user;

[0109] The winning bid rate SR is obtained by the following calculation formula:

[0110] ;

[0111] Where ZBL represents the number of projects that won bids within a fixed period, and TBZS represents the total number of projects that bid within a fixed period.

[0112] The integration unit marks and associates all the tender information according to the comprehensive secondary evaluation index SC of the tender information to form a comprehensive tender data group ZHZ, and performs bubble sorting;

[0113] Specifically, the bubble sort compares the results of the i-th comprehensive secondary evaluation index SC and the i+1-th comprehensive secondary evaluation index SC in the comprehensive bidding data group ZHZ to exchange positions, and repeats the exchange until there is no more exchange of positions, and stops the bubble sort.

[0114] In this embodiment, by introducing a matching evaluation unit and a status detection unit, the matching score MTC between the project requirements and the bidding information is accurately calculated, and through intelligent evaluation of multiple dimensions such as technical requirements, budget, and cycle, it is ensured that the selected bidding information is highly in line with the needs of the enterprise. The scoring and sorting module performs a secondary evaluation by combining historical data, generates a comprehensive secondary evaluation index SC, and accurately sorts the bidding information through bubble sorting. The system improves the quantitative evaluation of the supplier's past performance by combining the historical score HS and the winning rate SR, making the screening process more comprehensive and rigorous. At the same time, bubble sorting ensures that high-potential projects are presented first, allowing companies to quickly locate the best bidding opportunities. The automation and data-driven capabilities of the system effectively reduce manual screening errors, greatly improve the intelligence of screening, and provide refined matching evaluation and comprehensive sorting functions, enabling companies to make efficient decisions in complex bidding environments.

[0115] Embodiment 4:

[0116] This embodiment is explained in Example 3, please refer to Figure 1 ,Specifically: the iterative optimization module includes a feedback collection unit and an algorithm control unit;

[0117] The feedback collection unit verifies the sorted comprehensive bidding data group ZHZ bidding information and collects feedback information, including bidding results and project execution feedback information;

[0118] The collection of the bid result feedback information includes verifying the top five projects in the sorted comprehensive bidding data group ZHZ, and recording the bid results of the projects; if the bid result is a successful bid, the bidding data group ZZ of the winning project is recorded; if the bid result is a failure to win the bid, the reasons for the failure to win the bid of the top five projects are recorded, and the specific parameter information affecting the failure to win the bid in the bidding data group ZZ is marked, and the algorithm control unit is triggered to execute;

[0119] Project execution feedback information includes collecting project delivery quality status Q, cost control score C and project delivery cycle score T.

[0120] The algorithm control unit is triggered to control the matching score MTC and the comprehensive secondary evaluation index SC generated by the matching evaluation algorithm according to the winning results of the top five projects, and the control includes proportional adjustment of m1, m2, m3, h1, h2 and h3;

[0121] Among them, the trigger includes triggering the regulation of the matching score MTC and the comprehensive secondary evaluation index SC generated by the matching evaluation algorithm when the winning bid result is a winning bid.

[0122] In this embodiment, the dynamic optimization and continuous improvement of the screening process are ensured through the feedback collection unit and algorithm control unit in the iterative optimization module. The system not only verifies the winning results of the top five projects, but also identifies potential problems by collecting the reasons for not winning the bid and specific influencing factors, and triggers algorithm control to regulate the matching score MTC and the comprehensive secondary evaluation index SC generated by the matching evaluation algorithm. At the same time, the system collects detailed execution feedback information of the winning projects, including project quality, cost control and delivery cycle, so as to dynamically adjust the scoring weights in the next screening. Through this mechanism, the system can continuously learn historical results and execution performance, gradually optimize the matching score MTC and the comprehensive secondary evaluation index SC, and ensure that subsequent evaluation results are more accurate. Compared with static algorithms, this iterative optimization process greatly improves the flexibility and adaptability of the system, providing enterprises with more accurate and real-time updated bidding information screening and evaluation.

[0123] Embodiment 5:

[0124] Intelligent screening method of bidding information based on data mining technology, please refer to Figure 2 , specifically: including the following steps:

[0125] Step 1: The demand analysis module extracts demand information from the enterprise's project demand document through natural language processing technology and structures it, including technical requirements TRQ, budget range BRQ and project cycle DRQ, to form a demand data group XZ;

[0126] Step 2: The bidding information parsing module parses the received bidding information file and extracts the technical requirements, budget, project cycle, funding sources and additional terms. After integration, the extracted technical requirements TBD, extracted budget BBD, extracted execution cycle DBD and extracted risk parameters R are obtained to form the bidding data group ZZ;

[0127] Step 3: The matching degree calculation module establishes a matching degree evaluation algorithm for the demand data group XZ and the bidding data group ZZ, generates a matching degree score MTC between the bidding information file and the project demand, and matches it with the preset project demand matching bidding evaluation threshold PZ to obtain the matching status result of the current bidding information file;

[0128] Step 4: The scoring and sorting module combines the acquired bidding information file with historical data according to the matching status result of the current bidding information file, generates a comprehensive secondary evaluation index SC of the bidding information, and sorts the comprehensive bidding data group ZHZ of the comprehensive secondary evaluation index SC of the bidding information;

[0129] Step 5: The iterative optimization module verifies the sorted comprehensive bidding data group ZHZ bidding information and collects feedback information, including the winning bid results and project execution feedback information, and regulates the matching score MTC and the comprehensive secondary evaluation index SC generated by the matching evaluation algorithm.

[0130] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. Intelligent bidding information screening system based on data mining technology, Features: It includes demand analysis module, bidding information analysis module, matching degree calculation module, scoring and sorting module and iterative optimization module; The demand analysis module extracts demand information from the enterprise's project demand document through natural language processing technology and structures it, including technical requirements TRQ, budget range BRQ and project cycle DRQ, to form a demand data group XZ; The bidding information parsing module parses the received bidding information file and extracts information related to technical requirements, budget, project cycle, funding source and additional terms, and obtains extracted technical requirements TBD, extracted budget BBD, extracted execution cycle DBD and extracted risk parameters R through integration to form a bidding data group ZZ; The bidding information parsing module includes an information extraction unit and a data integration unit; The information extraction unit parses the received bidding information file, the parsing includes classifying according to different parts, and extracting technical requirements, budget, project cycle, funding sources and additional terms related information, and using regular expressions to obtain relevant numerical information of the extracted technical requirements, budget, project cycle, funding sources and additional terms related information, the data information includes the frequency of occurrence of different technical requirements, the frequency of occurrence of different budgets, the frequency of occurrence of different project cycles and the frequency of occurrence of different additional terms related information; The data integration unit integrates the relevant numerical information to extract technical requirements TBD, extract budget BBD, extract execution period DBD and extract risk parameter R to form a bidding data group ZZ, specifically ZZ={TBD, BBD, DBD, R}; The matching degree calculation module establishes a matching degree evaluation algorithm for the demand data group XZ and the bidding data group ZZ, generates a matching degree score MTC between the bidding information file and the project demand, and matches it with a preset project demand matching bidding evaluation threshold PZ to obtain a matching status result of the current bidding information file; The matching degree calculation module includes a matching evaluation unit and a state detection unit; The matching evaluation unit establishes a matching evaluation algorithm for the demand data group XZ and the bidding data group ZZ to generate a matching score MTC between the bidding information file and the project demand; The matching score MTC is obtained by the following calculation formula: ； Where SIMtech represents the matching degree of technical requirements, SIMbudget represents the matching degree of budget, SIMcycle represents the matching degree of execution cycle, log represents the logarithmic function, R represents the extracted risk parameter, m1, m2 and m3 represent the preset weight values ​​of the matching degree of technical requirements SIMtech, the matching degree of budget SIMbudget and the matching degree of execution cycle SIMcycle respectively, and , , , , the specific value is set by the user; SIMtech obtains the matching degree of the technical requirements through the following statistical algorithm formula: ； In the formula, Indicates the intersection of technical requirements TRQ and extraction technology TBD requirements, specifically indicating the same technical requirements information, It represents the union of the technical requirement TRQ and the extracted technical requirement TBD, specifically it represents the sum of the technical requirement TRQ and the extracted technical requirement TBD; The budget matching degree SIMbudget is obtained by the following statistical algorithm formula: ； In the formula, DRQ represents the project cycle, and DBD represents the extraction execution cycle; The state detection unit matches the matching score MTC with the preset project demand matching bidding evaluation threshold PZ to obtain the matching state result of the current bidding information file; The matching status result of the current bidding information file is obtained by the following matching method: ； When the matching return signal status of the matching status result of the current bidding information file is 1, the matching result is obtained, and the matching result is marked for the current bidding information file; When the matching return signal status of the matching status result of the current bidding information file is 0, the non-matching result is obtained, and the non-matching result is marked for the current bidding information file; The scoring and sorting module combines the acquired tender information file with historical data according to the matching status result of the current tender information file, generates a comprehensive secondary evaluation index SC of the tender information, and sorts the comprehensive tender data group ZHZ of the comprehensive secondary evaluation index SC of the tender information; The scoring and ranking module includes a secondary evaluation unit and an integration unit; The secondary evaluation unit combines the acquired tender information file with historical data according to the matching status result of the current tender information file, including the historical tender information execution records and winning records of the company from which the current tender information file comes, and simultaneously performs statistical analysis to obtain the historical score HS and winning rate SR of the tender information file execution of the company from which the current tender information comes, and then performs fitting with the matching score MTC to generate a comprehensive secondary evaluation index SC of the tender information; The comprehensive secondary evaluation index SC is obtained by the following calculation formula: ； The historical score HS is obtained by statistical analysis through the following calculation formula: ； Where Q represents the project delivery quality status, specifically whether there are quality problems after the project is delivered, and the area with quality problems is compared with the total project construction area to obtain the delivery quality area ratio information, C represents the cost control score, specifically the proportion of the project budget exceeded when the project is delivered, T represents the project delivery cycle score, specifically the proportion of the time delayed when the project is delivered, h1, h2 and h3 represent the preset weight values ​​of the project delivery quality status Q, cost control score and project delivery cycle score T respectively, and , , ,and , the specific value is set by the user; The winning bid rate SR is obtained by the following calculation formula: ； In the formula, ZBL represents the number of projects that won bids in a fixed period, and TBZS represents the total number of projects that bid in a fixed period; The integration unit marks and associates all the tender information according to the comprehensive secondary evaluation index SC of the tender information to form a comprehensive tender data group ZHZ, and performs bubble sorting; Specifically, the bubble sort compares the results of the i-th comprehensive secondary evaluation index SC and the i+1-th comprehensive secondary evaluation index SC in the comprehensive bidding data group ZHZ to exchange positions, and repeats the exchange until there is no exchange of positions, and stops the bubble sort; The iterative optimization module verifies the sorted comprehensive bidding data group ZHZ bidding information, collects feedback information, including the bidding results and project execution feedback information, and regulates the matching score MTC and the comprehensive secondary evaluation index SC generated by the matching evaluation algorithm; The iterative optimization module includes a feedback collection unit and an algorithm control unit; The feedback collection unit verifies the sorted comprehensive bidding data group ZHZ bidding information and collects feedback information, including bidding results and project execution feedback information; The collection of the bid result feedback information includes verifying the top five projects in the sorted comprehensive bidding data group ZHZ, and recording the bid results of the projects; if the bid result is a successful bid, the bidding data group ZZ of the winning project is recorded; if the bid result is a failure to win the bid, the reasons for the failure to win the bid of the top five projects are recorded, and the specific parameter information affecting the failure to win the bid in the bidding data group ZZ is marked, and the algorithm control unit is triggered to execute; Project execution feedback information includes collecting project delivery quality status Q, cost control score C and project delivery cycle score T.

2. According to claim 1, the intelligent bidding information screening system based on data mining technology, Features: The demand analysis module includes a natural language processing unit and a data structuring unit; The natural language processing unit parses and extracts demand information from the enterprise's project demand document by natural language processing technology. The parsing of the project demand document includes dividing the project demand document into different logical parts, including a technical requirement part, a budget part and a cycle part, and then extracting demand information by using keyword extraction and regular expression technology, including extracting technical requirement information, budget range information and project cycle information in the project demand; The data structuring unit performs structured processing on the extracted technical requirement information, budget range information and project cycle information. The result processing includes extracting the frequency of occurrence of different information contents, including technical requirements TRQ, budget range BRQ and project cycle DRQ, to form a demand data group XZ, ​​specifically XZ={TRQ, BRQ, DRQ}.

3. According to claim 1, the intelligent bidding information screening system based on data mining technology, Features: The algorithm control unit triggers the control of the matching score MTC and the comprehensive secondary evaluation index SC generated by the matching evaluation algorithm according to the winning results of the top five projects, and the control includes proportional adjustment of m1, m2, m3, h1, h2 and h3; Among them, the trigger includes triggering the regulation of the matching score MTC and the comprehensive secondary evaluation index SC generated by the matching evaluation algorithm when the winning bid result is a winning bid.

4. A method for intelligently screening bidding information based on data mining technology, comprising the intelligent screening system for bidding information based on data mining technology as described in any one of claims 1 to 3 above, Features: The following steps are involved: Step 1: The demand analysis module extracts demand information from the enterprise's project demand document through natural language processing technology and structures it, including technical requirements TRQ, budget range BRQ and project cycle DRQ, to form a demand data group XZ; Step 2: The bidding information parsing module parses the received bidding information file and extracts the technical requirements, budget, project cycle, funding sources and additional terms. After integration, the extracted technical requirements TBD, extracted budget BBD, extracted execution cycle DBD and extracted risk parameters R are obtained to form the bidding data group ZZ; Step 3: The matching degree calculation module establishes a matching degree evaluation algorithm for the demand data group XZ and the bidding data group ZZ, generates a matching degree score MTC between the bidding information file and the project demand, and matches it with the preset project demand matching bidding evaluation threshold PZ to obtain the matching status result of the current bidding information file; Step 4: The scoring and sorting module combines the acquired bidding information file with historical data according to the matching status result of the current bidding information file, generates a comprehensive secondary evaluation index SC of the bidding information, and sorts the comprehensive bidding data group ZHZ of the comprehensive secondary evaluation index SC of the bidding information; Step 5: The iterative optimization module verifies the sorted comprehensive bidding data group ZHZ bidding information and collects feedback information, including the winning bid results and project execution feedback information, and regulates the matching score MTC and the comprehensive secondary evaluation index SC generated by the matching evaluation algorithm.

Images