A method and system for identifying and analyzing bid-rigging behavior based on big data
By conducting qualification review and information sequence analysis of bidders, an identification matrix is built to judge the bidding behavior between bidders, which solves the problem of difficult to accurately identify bidding subjects in the existing technology, and achieves a more accurate analysis results.
Patent Information
- Application Number
- CN202410504274.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-04-25
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-04-25
Smart Images

Figure CN118410298B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of bidding and tendering, and in particular to a method and system for identifying and analyzing bid-rigging and collusion based on big data. Background Art
[0002] Bid rigging usually refers to an agreement among several bidders to raise or lower their bid prices in a coordinated manner in order to restrict competition, exclude other bidders, and allow a certain stakeholder to win the bid.
[0003] However, the existing methods and systems for identifying and analyzing bid-rigging behaviors based on big data only determine whether there is bid-rigging or bid-rigging among bidders by analyzing whether there is a relationship between the bidders participating in the bidding. However, the relationships between bidders are often complex and diverse, and may involve multiple levels and dimensions. It may be difficult to fully and accurately reveal bid-rigging and bid-rigging behaviors through relationship analysis alone. For example, some entities may collude through complex associations and hidden means, which may not be easily discovered by simple relationship analysis. It is impossible to obtain more accurate bid-rigging and bid-rigging identification and analysis results for each bidder based on the uploaded information of all bidders. For example, the patent with the publication number of "CN110992059A" and the name of the patent is "A method for identifying and analyzing bid-rigging and bid-collusion based on big data", and the method includes the following steps: With the help of big data technology, an analysis model for identifying bid-rigging and bid-collusion behaviors is established - a community detection model based on complex networks, by constructing a community relationship network between the bidders participating in the tender, the network can dynamically modify the weights according to the corporate behavior and integrity relationship feedback, and at the same time use the community detection algorithm to discover the potential community relationship, and judge whether there is bid-rigging and bid-collusion between the bidders. Specifically, the model is applied to the complex network of corporate bidding behavior, and the hidden laws behind the transaction data are revealed through data experimental results, and the transaction trajectory and behavioral characteristics of bid-rigging and bid-collusion are portrayed, which reflects the internal root causes of bid-rigging and bid-collusion behaviors to a certain extent. However, this patent only judges whether there is bid-rigging and bid-collusion between the bidders by analyzing whether there is a relationship between the bidders participating in the tender, which has great limitations. It is impossible to mine the hidden relationship based on the technical information and business information of multiple bidders, and it is impossible to obtain the bid-rigging and bid-collusion behavior identification and analysis results of each bidder more accurately based on the uploaded information of all bidders.
[0004] Therefore, the present invention proposes a method and system for identifying and analyzing bid-rigging behavior based on big data. Summary of the invention
[0005] The present invention provides a method and system for identifying and analyzing bid-rigging behaviors based on big data, which is used to conduct a qualification review of the corresponding bidders according to all uploaded information of each bidder, obtain the first qualification review result of the corresponding bidder, screen out the bidders that do not have the ability to undertake the bidding project, conduct a second qualification review of each bidder whose first qualification review result is passed, obtain the second qualification review result of the corresponding bidder, determine all bidders that have an associated relationship and may have bid-rigging behaviors, obtain the business information sequence and technical information sequence of the corresponding bidder according to all uploaded information of each inspected bidder, so as to facilitate the subsequent accurate acquisition of the first predicted bid-rigging and second predicted bid-rigging for each inspected bidder, and obtain the first predicted bid-rigging and second predicted bid-rigging for each inspected bidder according to the information uploaded by all inspected bidders. All first predicted bidders for each investigated bidder are obtained according to the business information sequence, all second predicted bidders for each investigated bidder are obtained according to the technical information sequence of all investigated bidders, and the identification matrix of the corresponding investigated bidder is obtained according to the business information sequence of all first predicted bidders for each investigated bidder and the technical information sequence of all second predicted bidders, so as to facilitate the judgment and analysis on whether there is bid-rigging between the investigated bidder and the corresponding all first predicted bid-rigging and all second predicted bid-rigging. The identification analysis results of bid-rigging of all investigated bidders are accurately obtained according to the identification matrix of all investigated bidders, that is, it is judged whether there is bid-rigging between multiple bidders from both business information and technical information.
[0006] The present invention provides a method for identifying and analyzing bid-rigging behaviors based on big data, comprising:
[0007] S1: Obtain all uploaded information of all bidders, and conduct a qualification review of the corresponding bidder based on all uploaded information of each bidder, obtain the first qualification review result of the corresponding bidder, and conduct a second qualification review on each bidder whose first qualification review result is passed, and obtain the second qualification review result of the corresponding bidder;
[0008] S2: All bidders whose second qualification review results are passed are taken as the inspection bidders, and based on all the uploaded information of each inspection bidder, the business information sequence and technical information sequence of the corresponding inspection bidder are obtained;
[0009] S3: Based on the business information sequences of all the inspected bidders, all the first predicted bidders that may collude in the bid for each inspected bidder are obtained; based on the technical information sequences of all the inspected bidders, all the second predicted bidders that may collude in the bid for each inspected bidder are obtained; and based on the business information sequences of all the first predicted bidders that may collude in the bid for each inspected bidder and the technical information sequences of all the second predicted bidders that may collude in the bid for each inspected bidder, an identification matrix of the corresponding inspected bidder is obtained; based on the identification matrix of all the inspected bidders, identification and analysis results of the collusion behavior of all the inspected bidders are obtained.
[0010] Preferably, the bid-rigging behavior identification and analysis method based on big data obtains all uploaded information of all bidders, and conducts a qualification review of the corresponding bidder based on all uploaded information of each bidder to obtain the first qualification review result of the corresponding bidder, including:
[0011] Obtain all uploaded information of all bidders, including the bid documents and all qualification certificates of the bidders;
[0012] If the qualification certificates of each category of bidders of each bidder meet the preset certificate requirements of the corresponding category, the first qualification review result of the corresponding bidder will be set to passed; otherwise, the first qualification review result of the corresponding bidder will be set to failed, and the first qualification review results of all bidders will be obtained.
[0013] Preferably, the method for identifying and analyzing bid-rigging behavior based on big data conducts a secondary qualification review on each bidder whose first qualification review result is passed, and obtains the second qualification review result of the corresponding bidder, including:
[0014] Obtain the company names of all bidders whose first qualification review results are passed, and obtain the actual beneficiaries of the corresponding bidders based on the company name of each bidder and the preset equity penetration analysis platform;
[0015] Compare the actual beneficiaries of all bidders. If the actual beneficiaries of different bidders are the same, the second qualification review results of all bidders with the same actual beneficiaries will be set as failed;
[0016] Among all bidders whose first qualification review results are passed, the second qualification review results of all bidders except those whose second qualification review results are failed will be set as passed.
[0017] Preferably, the method for identifying and analyzing bid-rigging based on big data, S2: taking all bidders whose second qualification review results are passed as the bidders for investigation, and obtaining the business information sequence and technical information sequence of the corresponding bidders for investigation based on all uploaded information of each bidder for investigation, including:
[0018] All bidders whose second qualification review results are passed are regarded as inspected bidders, and the business information of the corresponding inspected bidder is obtained based on the bid of each inspected bidder, wherein the business information includes the total bid price and the price list of main items;
[0019] Obtaining technical information of the corresponding inspected bidder based on the bid of each inspected bidder, wherein the technical information includes the project plan;
[0020] Based on the business information of each inspected bidder, a business information sequence of the corresponding inspected bidder is obtained; based on the technical information of each inspected bidder, a technical information sequence of the corresponding inspected bidder is obtained.
[0021] Preferably, the method for identifying and analyzing bid-rigging based on big data obtains the business information sequence of the corresponding bidder based on the business information of each bidder under investigation, and obtains the technical information sequence of the corresponding bidder based on the technical information of each bidder under investigation, including:
[0022] The total bid price of each investigated bidder is taken as the first sequence element of the business information sequence, and all main items in the main item price list are sorted according to the position of the main items in the main item price list to obtain the main item sorting result of each investigated bidder;
[0023] The price value of each main item in the main item price list is used as a sequence element of the business information sequence, and the relative positions of the sequence elements corresponding to all the main items in the main item price list are consistent with the main item sorting result, so as to obtain the business information sequence of the corresponding inspection bidder;
[0024] Eliminate all non-statistical words in the project plan of each inspected bidder, obtain the key text of the corresponding project plan, take the words that appear more than 2 times in the key text of each project plan as the plan words, obtain the word frequency of each plan word in the project plan of each inspected bidder, and sort all the plan words in the project plan of each inspected bidder from large to small according to the word frequency, and obtain the sorting result of the plan words of the corresponding inspected bidder;
[0025] The first preset number of solution words in the solution word ranking result of each investigated bidder are used as selected solution words, and based on the word frequency of each selected solution word and the total word frequency of all solution words in the corresponding solution word ranking result, the sequence element representation of the corresponding selected solution word is determined;
[0026] The sequence element representations of all selected proposal words of each inspected bidder are used as the sequence elements of the technical information sequence of the corresponding inspected bidder to obtain the technical information sequence of the corresponding inspected bidder, wherein the sequence positions of all sequence elements in the technical information sequence are consistent with the sequence order of the first appearance of all sequence elements in the corresponding project proposal.
[0027] Preferably, the method for identifying and analyzing bid-rigging behavior based on big data obtains all first predicted bid-rigging parties of each investigated bidder based on the business information sequence of all investigated bidders, including:
[0028] Based on the business information sequences of the two inspected bidders, the similarity of the business information of the two inspected bidders is obtained, which is:
[0029] ;
[0030] in, is the business information similarity of the two investigated bidders currently being calculated, The first of the two currently calculated bidders in the business information sequence of the first bidder sequence elements, The first one in the business information sequence of the second of the two currently calculated bidders sequence elements, The number of sequence elements of the business information sequence for a single investigated bidder;
[0031] Among all the inspected bidders, two inspected bidders whose business information similarity is greater than a preset first similarity are determined to be first predicted bid-collusion parties;
[0032] Obtain all first predicted collusion parties for each investigated bidder.
[0033] Preferably, the method for identifying and analyzing bid-rigging behavior based on big data obtains all second predicted bid-rigging parties of each investigated bidder based on the technical information sequence of all investigated bidders, including:
[0034] Based on the technical information sequences of the two inspected bidders, the similarity of the technical information of the two inspected bidders is obtained, which is:
[0035] ;
[0036] in, is the similarity of technical information of the two investigated bidders currently calculated, The first of the two currently calculated inspection bidders in the technical information sequence of the first inspection bidder. sequence elements, The second of the two currently calculated bidders in the technical information sequence of the second bidder sequence elements, The number of sequence elements for a single sequence of technical information of a candidate for investigation;
[0037] Among all the inspected bidders, two inspected bidders whose technical information similarity is greater than a preset second similarity are determined to be the second predicted bid-collusive parties;
[0038] Obtain all second predicted collusion parties for each investigated bidder.
[0039] Preferably, the big data-based bid-rigging behavior identification and analysis method obtains the identification matrix of the corresponding investigated bidder based on the business information sequence of all first predicted bid-rigging parties and the technical information sequence of all second predicted bid-rigging parties of each investigated bidder, including:
[0040] According to the principle of descending similarity of business information with the corresponding investigated bidder, all first predicted bid-collusive parties of each investigated bidder are defined ordinal numbers starting from 1, and the ranking result of the first predicted bid-collusive parties of each investigated bidder is obtained;
[0041] According to the principle of descending similarity of technical information with the corresponding inspected bidder, all the second predicted bidders for bid collusion of each inspected bidder are defined ordinal numbers starting from 1, and the ranking result of the second predicted bidders for bid collusion of each inspected bidder is obtained;
[0042] Based on the first predicted bid-collusion ranking result and the second predicted bid-collusion ranking result of each investigated bidder, the identification matrix of the corresponding investigated bidder is obtained, which is:
[0043] ;
[0044] in, is the identification matrix of a single inspected bidder, is the first sequence element in the business information sequence of the first predicted bidder with the ordinal number 1 corresponding to the bidder under investigation, is the second sequence element in the business information sequence of the first predicted bidder with the ordinal number 1 corresponding to the bidder under investigation, is the first predicted bidder in the business information sequence of the bidder with the ordinal number 1. sequence elements, is the number of sequence elements of the business information sequence of a single investigated bidder, is the first sequence element in the business information sequence of the first predicted bidder with ordinal number 2 corresponding to the bidder under investigation, is the second sequence element in the business information sequence of the first predicted bidder with the ordinal number 2 corresponding to the bidder under investigation, is the first predicted bidder in the business information sequence of the bidder with the ordinal number 2. sequence elements, The ordinal number of the corresponding bidder is The first sequence element in the business information sequence of the first predicted bidder, The ordinal number of the corresponding bidder is The second sequence element in the business information sequence of the first predicted bidder, The ordinal number of the corresponding bidder is The first prediction of the business information sequence of the bidder sequence elements, is the number of bidders that are colluding in the first prediction of the bidders to be investigated, is the first sequence element in the technical information sequence of the second predicted bidder with the ordinal number 1 corresponding to the investigated bidder, is the second sequence element in the technical information sequence of the second predicted bidder with ordinal number 1 corresponding to the bidder under investigation, is the first in the technical information sequence of the second predicted bidder with the ordinal number 1 corresponding to the bidder under investigation. sequence elements, is the number of sequence elements of the technical information sequence of a single inspection bidder, is the first sequence element in the technical information sequence of the second predicted bidder with the ordinal number 2 corresponding to the bidder under investigation, is the second sequence element in the technical information sequence of the second predicted bidder with the ordinal number 2 corresponding to the bidder under investigation, is the technical information sequence of the second predicted bidder with the ordinal number 2 corresponding to the bidder under investigation. sequence elements, The ordinal number of the corresponding bidder is The first sequence element in the technical information sequence of the second predicted bidder, The ordinal number of the corresponding bidder is The second sequence element in the technical information sequence of the second predicted bidder, The ordinal number of the corresponding bidder is The second prediction of the technical information sequence of the bid-rigging party sequence elements, The second predicted number of collusion parties corresponding to the bidder under investigation.
[0045] Preferably, the bid-rigging behavior identification and analysis method based on big data obtains the bid-rigging behavior identification and analysis results of all investigated bidders based on the identification matrix of all investigated bidders, including:
[0046] If the rank of the identification matrix of a single inspected bidder is greater than a preset rank threshold, the identification analysis result of the bid-rigging behavior of the corresponding inspected bidder is determined to be a bid-rigging behavior.
[0047] The present invention provides a system for identifying and analyzing bid-rigging behaviors based on big data, which is used to execute any one of the methods for identifying and analyzing bid-rigging behaviors based on big data in Embodiments 1 to 9, including:
[0048] The review module is used to obtain all uploaded information of all bidders, and conduct a qualification review of the corresponding bidder based on all uploaded information of each bidder to obtain the first qualification review result of the corresponding bidder, and conduct a second qualification review of each bidder whose first qualification review result is passed to obtain the second qualification review result of the corresponding bidder;
[0049] A processing module, used to take all bidders whose second qualification review results are passed as the inspection bidders, and obtain the business information sequence and technical information sequence of the corresponding inspection bidder based on all uploaded information of each inspection bidder;
[0050] An analysis module is used to obtain all first predicted bidders that may collude in bidding for each investigated bidder based on the business information sequences of all investigated bidders, obtain all second predicted bidders that may collude in bidding for each investigated bidder based on the technical information sequences of all investigated bidders, and obtain an identification matrix of the corresponding investigated bidder based on the business information sequences of all first predicted bidders that may collude in bidding and the technical information sequences of all second predicted bidders that may collude in bidding for each investigated bidder, and obtain identification analysis results of collusion behavior of all investigated bidders based on the identification matrices of all investigated bidders.
[0051] Compared with the prior art, the beneficial effects of the present invention are as follows: a qualification review is conducted on the corresponding bidders according to all the uploaded information of each bidder, a first qualification review result of the corresponding bidder is obtained, bidders that do not have the ability to undertake the bidding project are screened out, a second qualification review is conducted on each bidder whose first qualification review result is passed, a second qualification review result of the corresponding bidder is obtained, all bidders that have an associated relationship and may have bid-rigging and collusion behaviors are determined, and the business information sequence and technical information sequence of the corresponding bidder are obtained according to all the uploaded information of each inspected bidder, so as to facilitate the subsequent accurate acquisition of the first predicted bid-rigging and second predicted bid-rigging and collusion parties of each inspected bidder, and the business information sequence of all inspected bidders is obtained. All first predicted bid-rigging parties of each investigated bidder are obtained in a column, all second predicted bid-rigging parties of each investigated bidder are obtained according to the technical information sequences of all investigated bidders, and an identification matrix of the corresponding investigated bidder is obtained according to the business information sequences of all first predicted bid-rigging parties and the technical information sequences of all second predicted bid-rigging parties of each investigated bidder, so as to facilitate the judgment and analysis on whether there is bid-rigging behavior between the investigated bidder and the corresponding all first predicted bid-rigging parties and all second predicted bid-rigging parties. According to the identification matrix of all investigated bidders, the identification and analysis results of bid-rigging behavior of all investigated bidders are accurately obtained, that is, whether there is bid-rigging behavior between multiple bidders is judged from both business information and technical information.
[0052] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structure specifically pointed out in the written application document.
[0053] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0054] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:
[0055] Figure 1 This is a flow chart of a method for identifying and analyzing bid-rigging behavior based on big data in an embodiment of the present invention;
[0056] Figure 2 Schematic diagram of a system for identifying and analyzing bid-rigging behavior based on big data in an embodiment of the present invention. DETAILED DESCRIPTION
[0057] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention. Embodiment 1:
[0058] The present invention provides a method for identifying and analyzing bid-rigging behavior based on big data. Figure 1 ,include:
[0059] S1: Obtain all uploaded information of all bidders, and conduct a qualification review of the corresponding bidder based on all uploaded information of each bidder, obtain the first qualification review result of the corresponding bidder, and conduct a second qualification review on each bidder whose first qualification review result is passed, and obtain the second qualification review result of the corresponding bidder;
[0060] S2: All bidders whose second qualification review results are passed are taken as the inspection bidders, and based on all the uploaded information of each inspection bidder, the business information sequence and technical information sequence of the corresponding inspection bidder are obtained;
[0061] S3: Based on the business information sequences of all the inspected bidders, all the first predicted bidders that may collude in the bid for each inspected bidder are obtained; based on the technical information sequences of all the inspected bidders, all the second predicted bidders that may collude in the bid for each inspected bidder are obtained; and based on the business information sequences of all the first predicted bidders that may collude in the bid for each inspected bidder and the technical information sequences of all the second predicted bidders that may collude in the bid for each inspected bidder, an identification matrix of the corresponding inspected bidder is obtained; based on the identification matrix of all the inspected bidders, identification and analysis results of the collusion behavior of all the inspected bidders are obtained.
[0062] In this embodiment, the bidder is an enterprise that responds to the tender and participates in the bidding competition.
[0063] In this embodiment, a qualification review is a process of conducting a comprehensive, objective and fair assessment and review of the qualifications of enterprises (bidders) participating in the bidding competition to ensure that they have the ability and conditions to undertake the bidding project.
[0064] In this embodiment, the first qualification review result is the result of a comprehensive, objective and fair evaluation and review of the various qualifications of the enterprises (bidders) participating in the bidding competition, and the first qualification review result includes pass and fail.
[0065] In this embodiment, the secondary qualification review is a preliminary review process to determine whether there is any relationship between different enterprises (bidders) participating in the bidding competition and whether there may be any collusion in bidding.
[0066] In this embodiment, the second qualification review result is the result of a preliminary review of whether there is any association relationship between different enterprises (bidders) participating in the bidding competition and whether there may be collusion in bidding, and the second qualification review result includes pass and fail.
[0067] In this embodiment, the business information sequence is a sequence constructed based on the business information in the bid document of the investigated bidder and reflects the main features of the business information in the bid document of the investigated bidder.
[0068] In this embodiment, the technical information sequence is a sequence constructed based on the technical information in the bid document of the investigated bidder and reflects the main features of the technical information in the bid document of the investigated bidder.
[0069] In this embodiment, the first predicted bid-rigging party is the remaining investigated bidders predicted based on the business information sequences of all investigated bidders that may engage in bid-rigging with the corresponding investigated bidder.
[0070] In this embodiment, the second predicted bidders engaging in bid rigging are the remaining bidders that are predicted to be likely to engage in bid rigging with the corresponding bidder based on the technical information sequences of all the bidders under investigation.
[0071] In this embodiment, the identification matrix is a matrix constructed by using the business information sequences of all first predicted bid-rigging parties and the technical information sequences of all second predicted bid-rigging parties of each investigated bidder as matrix elements, and contains some performance characteristics of whether the corresponding investigated bidder has engaged in bid-rigging behavior.
[0072] In this embodiment, the bid-rigging behavior identification analysis result is a behavior identification result of whether the corresponding bidder has engaged in bid-rigging behavior, which is obtained based on the identification matrix of each investigated bidder and a preset rank threshold.
[0073] The beneficial effects of the above technology are as follows: according to all the uploaded information of each bidder, a qualification review is conducted on the corresponding bidder to obtain the first qualification review result of the corresponding bidder, and the bidders that do not have the ability to undertake the bidding project are screened out; for each bidder whose first qualification review result is passed, a second qualification review is conducted on the bidder to obtain the second qualification review result of the corresponding bidder, and all bidders with associated relationships and possible bid-rigging behaviors are determined; according to all the uploaded information of each inspected bidder, the business information sequence and technical information sequence of the corresponding inspected bidder are obtained, so as to facilitate the subsequent accurate acquisition of the first predicted bid-rigging and second predicted bid-rigging of each inspected bidder; according to the business information sequences of all inspected bidders, the all first predicted bidders for each investigated bidder are obtained, all second predicted bidders for each investigated bidder are obtained according to the technical information sequences of all investigated bidders, and an identification matrix of the corresponding investigated bidder is obtained according to the business information sequences of all first predicted bidders for each investigated bidder and the technical information sequences of all second predicted bidders, so as to facilitate the judgment and analysis on whether there is bid-rigging between the investigated bidder and the corresponding all first predicted bid-rigging and all second predicted bid-rigging, and accurately obtain the identification and analysis results of the bid-rigging of all investigated bidders according to the identification matrix of all investigated bidders, that is, to judge whether there is bid-rigging between multiple bidders from both business information and technical information. Embodiment 2:
[0074] On the basis of Example 1, a bid-rigging behavior identification and analysis method based on big data obtains all uploaded information of all bidders, and conducts a qualification review of the corresponding bidder based on all uploaded information of each bidder, and obtains the first qualification review result of the corresponding bidder, including:
[0075] Obtain all uploaded information of all bidders, including the bid documents and all qualification certificates of the bidders;
[0076] If the qualification certificates of each category of bidders of each bidder meet the preset certificate requirements of the corresponding category, the first qualification review result of the corresponding bidder will be set to passed; otherwise, the first qualification review result of the corresponding bidder will be set to failed, and the first qualification review results of all bidders will be obtained.
[0077] In this embodiment, the bid is a document submitted by the bidder to the tendering unit according to the conditions and requirements of the tender document, which includes the total bid price, the price list of main items, the project plan, etc.
[0078] In this embodiment, the qualification certificates are all qualification certificates related to bidding uploaded by the bidder and required by the tenderer, such as business license, tax registration certificate, organization code certificate, construction qualification certificate, etc.
[0079] In this embodiment, the preset certificate requirements are the requirements for the qualification certificates of each type of bidders pre-set by the tendering party, for example, the certificate requirement for the bidder's construction qualification certificate is the general construction contracting qualification certificate.
[0080] The beneficial effects of the above technology are: a qualification review of the corresponding bidder is conducted based on all uploaded information of each bidder, the first qualification review result of the corresponding bidder is obtained, and the bidders that do not have the ability to undertake the bidding project are screened out, which is convenient for the subsequent secondary qualification review. Embodiment 3:
[0081] On the basis of Example 1, a method for identifying and analyzing bid-rigging behavior based on big data is used to conduct a secondary qualification review on each bidder whose first qualification review result is passed, and obtain the second qualification review result of the corresponding bidder, including:
[0082] Obtain the company names of all bidders whose first qualification review results are passed, and obtain the actual beneficiaries of the corresponding bidders based on the company name of each bidder and the preset equity penetration analysis platform;
[0083] Compare the actual beneficiaries of all bidders. If the actual beneficiaries of different bidders are the same, the second qualification review results of all bidders with the same actual beneficiaries will be set as failed;
[0084] Among all bidders whose first qualification review results are passed, the second qualification review results of all bidders except those whose second qualification review results are failed will be set as passed.
[0085] In this embodiment, the enterprise name is the full name of the bidder in the industrial and commercial registration (business license).
[0086] In this embodiment, the preset equity penetration analysis platform is a pre-set platform for analyzing the actual beneficiaries of the corresponding bidders according to the equity results of each bidder, such as the existing Qichacha platform.
[0087] In this embodiment, the actual beneficiary is the person who enjoys the rights of corporate asset income, participation in major decisions and selection of managers. In this embodiment, the actual beneficiary is all shareholders who own more than 25% of the shares of the bidding party.
[0088] The beneficial effects of the above technology are: a secondary qualification review is conducted on each bidder whose first qualification review result is passed, and the second qualification review result of the corresponding bidder is obtained, and from the perspective of the actual beneficiary of the enterprise, all bidders with associated relationships and possible collusion in bidding are determined. Embodiment 4:
[0089] Based on Example 1, a method for identifying and analyzing bid-rigging behavior based on big data, S2: all bidders whose second qualification review results are passed are taken as inspected bidders, and based on all uploaded information of each inspected bidder, the business information sequence and technical information sequence of the corresponding inspected bidder are obtained, including:
[0090] All bidders whose second qualification review results are passed are regarded as inspected bidders, and the business information of the corresponding inspected bidder is obtained based on the bid of each inspected bidder, wherein the business information includes the total bid price and the price list of main items;
[0091] Obtaining technical information of the corresponding inspected bidder based on the bid of each inspected bidder, wherein the technical information includes the project plan;
[0092] Based on the business information of each inspected bidder, a business information sequence of the corresponding inspected bidder is obtained; based on the technical information of each inspected bidder, a technical information sequence of the corresponding inspected bidder is obtained.
[0093] In this embodiment, the total bid price is the total price quoted by the bidders during this bidding process.
[0094] In this embodiment, the main item price list is a table filled with the main material prices during the bidding process, where the main materials are the materials that are indispensable for the project, such as steel prices and cement prices.
[0095] In this embodiment, the project plan is a project implementation plan formulated by each inspected bidder in response to the tender document issued by the tendering party.
[0096] The beneficial effects of the above technology are: based on all the uploaded information of each inspected bidder, the business information and technical information of the corresponding inspected bidder is obtained, and a business information sequence and a technical information sequence are constructed based on the business information and technical information of each inspected bidder, so as to realize the separate serialization representation of the business information and technical information in all the uploaded information of the inspected bidder, so as to facilitate the subsequent calculation of the serialization results to accurately obtain the first predicted collusion party and the second predicted collusion party of each inspected bidder. Embodiment 5:
[0097] On the basis of Example 4, a method for identifying and analyzing bid-rigging based on big data, based on the business information of each investigated bidder, obtains a business information sequence of the corresponding investigated bidder, and based on the technical information of each investigated bidder, obtains a technical information sequence of the corresponding investigated bidder, including:
[0098] The total bid price of each investigated bidder is taken as the first sequence element of the business information sequence, and all main items in the main item price list are sorted according to the position of the main items in the main item price list to obtain the main item sorting result of each investigated bidder;
[0099] The price value of each main item in the main item price list is used as a sequence element of the business information sequence, and the relative positions of the sequence elements corresponding to all the main items in the main item price list are consistent with the main item sorting result, so as to obtain the business information sequence of the corresponding inspection bidder;
[0100] Eliminate all non-statistical words in the project plan of each inspected bidder, obtain the key text of the corresponding project plan, take the words that appear more than 2 times in the key text of each project plan as the plan words, obtain the word frequency of each plan word in the project plan of each inspected bidder, and sort all the plan words in the project plan of each inspected bidder from large to small according to the word frequency, and obtain the sorting result of the plan words of the corresponding inspected bidder;
[0101] The first preset number of solution words in the solution word ranking result of each investigated bidder are used as selected solution words, and based on the word frequency of each selected solution word and the total word frequency of all solution words in the corresponding solution word ranking result, the sequence element representation of the corresponding selected solution word is determined;
[0102] The sequence element representations of all selected proposal words of each inspected bidder are used as the sequence elements of the technical information sequence of the corresponding inspected bidder to obtain the technical information sequence of the corresponding inspected bidder, wherein the sequence positions of all sequence elements in the technical information sequence are consistent with the sequence order of the first appearance of all sequence elements in the corresponding project proposal.
[0103] In this embodiment, the position of the main item in the main item price list is based on the upper and lower positions of the main item in the main item price list, and there is only one main item in a row of the main item price list.
[0104] In this embodiment, the main item sorting result is the result of sorting all the main items in the main item price list according to the positions of the main items in the main item price list.
[0105] In this embodiment, non-statistical words are words in the project proposals of the bidders that cannot reflect the characteristics of the proposals, such as stop words.
[0106] In this embodiment, the key text is the text in the project proposal of the bidder that can reflect the characteristics of the proposal.
[0107] In this embodiment, the word frequency is the number of times each solution word appears in the key text.
[0108] In this embodiment, the scheme word ranking result is the result of sorting all the scheme words in the project plan of the inspected bidder from large to small according to the word frequency of each scheme word in the project plan of the inspected bidder, and when the word frequencies of multiple scheme words are the same, they are sorted in the order of the first appearance of the corresponding scheme words in the project plan.
[0109] In this embodiment, the preset number is a preset number of selected solution words, for example, 5.
[0110] In this embodiment, the total word frequency is the sum of the word frequencies of all solution words in the solution word ranking result.
[0111] In this embodiment, based on the word frequency of each selected solution word and the total word frequency of all solution words in the corresponding solution word ranking result, the sequence element representation of the corresponding selected solution word is determined, that is: the sum of the quotient of the word frequency of each selected solution word and the total word frequency of all solution words in the corresponding solution word ranking result and 1 is determined, and the natural logarithm function value of the obtained sum is used as the sequence element representation of the corresponding selected solution word.
[0112] The beneficial effects of the above technology are: constructing a business information sequence and a technical information sequence based on the business information and technical information of each inspected bidder, so as to facilitate the subsequent accurate acquisition of the first predicted collusion party and the second predicted collusion party of each inspected bidder. This embodiment gives in detail a specific implementation method for constructing a business information sequence and a technical information sequence based on the business information and technical information of each inspected bidder. Embodiment 6:
[0113] On the basis of Example 1, a method for identifying and analyzing bid-rigging behavior based on big data obtains all first predicted bid-rigging parties of each investigated bidder based on the business information sequences of all investigated bidders, including:
[0114] Based on the business information sequences of the two inspected bidders, the similarity of the business information of the two inspected bidders is obtained, which is:
[0115] ;
[0116] in, is the business information similarity of the two investigated bidders currently being calculated, The first of the two currently calculated bidders in the business information sequence of the first bidder sequence elements, The first one in the business information sequence of the second of the two currently calculated bidders sequence elements, The number of sequence elements of the business information sequence for a single investigated bidder;
[0117] Among all the inspected bidders, two inspected bidders whose business information similarity is greater than a preset first similarity are determined to be first predicted bid-collusion parties;
[0118] Obtain all first predicted collusion parties for each investigated bidder.
[0119] In this embodiment, the business information similarity is a numerical value obtained based on the business information sequences of the two investigated bidders and reflects the similarity of the business information in the bids of the two investigated bidders.
[0120] In this embodiment, the preset first similarity is a pre-set similarity threshold used to determine whether two investigated bidders are first predicted bid-rigging parties to each other.
[0121] The beneficial effects of the above technology are: based on the business information sequences of the two inspected bidders, the similarity between the business information of the two inspected bidders is accurately quantified, and the quantified business information similarity is compared with the preset similarity threshold to obtain all the first predicted bidders who may engage in bid-rigging for each inspected bidder, thereby achieving preliminary screening of bidders that may engage in bid-rigging with the corresponding inspected bidders. Embodiment 7:
[0122] On the basis of Example 1, a method for identifying and analyzing bid-rigging behavior based on big data, based on the technical information sequences of all investigated bidders, obtains all second predicted bid-rigging parties of each investigated bidder, including:
[0123] Based on the technical information sequences of the two inspected bidders, the similarity of the technical information of the two inspected bidders is obtained, which is:
[0124] ;
[0125] in, is the similarity of technical information of the two investigated bidders currently calculated, The first of the two currently calculated inspection bidders in the technical information sequence of the first inspection bidder. sequence elements, The second of the two currently calculated bidders in the technical information sequence of the second bidder sequence elements, The number of sequence elements for a single sequence of technical information of a candidate for investigation;
[0126] Among all the inspected bidders, two inspected bidders whose technical information similarity is greater than a preset second similarity are determined to be the second predicted bid-collusive parties;
[0127] Obtain all second predicted collusion parties for each investigated bidder.
[0128] In this embodiment, the technical information similarity is a numerical value obtained based on the technical information sequences of the two investigated bidders and reflecting the similarity of the technical information in the bids of the two investigated bidders.
[0129] In this embodiment, the preset second similarity is a pre-set similarity threshold used to determine whether two investigated bidders are the second predicted bid-rigging parties to each other.
[0130] The beneficial effects of the above technology are: based on the technical information sequences of the two inspected bidders, the similarity between the technical information of the two inspected bidders is accurately quantified, and the quantified technical information similarity is compared with the preset second similarity threshold to obtain all the second predicted bidders who may engage in bid-rigging for each inspected bidder, thereby further screening the bidders who may engage in bid-rigging with the corresponding inspected bidders. Embodiment 8:
[0131] On the basis of Example 1, the big data-based bid-rigging behavior identification and analysis method obtains the identification matrix of the corresponding bidder under investigation based on the business information sequences of all first predicted bid-rigging parties and the technical information sequences of all second predicted bid-rigging parties of each bidder under investigation, including:
[0132] According to the principle of descending similarity of business information with the corresponding investigated bidder, all first predicted bid-collusive parties of each investigated bidder are defined ordinal numbers starting from 1, and the ranking result of the first predicted bid-collusive parties of each investigated bidder is obtained;
[0133] According to the principle of descending similarity of technical information with the corresponding inspected bidder, all the second predicted bidders for bid collusion of each inspected bidder are defined ordinal numbers starting from 1, and the ranking result of the second predicted bidders for bid collusion of each inspected bidder is obtained;
[0134] Based on the first predicted bid-collusion ranking result and the second predicted bid-collusion ranking result of each investigated bidder, the identification matrix of the corresponding investigated bidder is obtained, which is:
[0135] ;
[0136] in, is the identification matrix of a single inspected bidder, is the first sequence element in the business information sequence of the first predicted bidder with the ordinal number 1 corresponding to the bidder under investigation, is the second sequence element in the business information sequence of the first predicted bidder with the ordinal number 1 corresponding to the bidder under investigation, is the first predicted bidder in the business information sequence of the bidder with the ordinal number 1. sequence elements, is the number of sequence elements of the business information sequence of a single investigated bidder, is the first sequence element in the business information sequence of the first predicted bidder with ordinal number 2 corresponding to the bidder under investigation, is the second sequence element in the business information sequence of the first predicted bidder with the ordinal number 2 corresponding to the bidder under investigation, is the first predicted bidder in the business information sequence of the bidder with the ordinal number 2. sequence elements, The ordinal number of the corresponding bidder is The first sequence element in the business information sequence of the first predicted bidder, The ordinal number of the corresponding bidder is The second sequence element in the business information sequence of the first predicted bidder, The ordinal number of the corresponding bidder is The first prediction of the business information sequence of the bidder sequence elements, is the number of bidders that are colluding in the first prediction of the bidders to be investigated, is the first sequence element in the technical information sequence of the second predicted bidder with the ordinal number 1 corresponding to the investigated bidder, is the second sequence element in the technical information sequence of the second predicted bidder with ordinal number 1 corresponding to the bidder under investigation, is the first in the technical information sequence of the second predicted bidder with the ordinal number 1 corresponding to the bidder under investigation. sequence elements, is the number of sequence elements of the technical information sequence of a single inspection bidder, is the first sequence element in the technical information sequence of the second predicted bidder with the ordinal number 2 corresponding to the bidder under investigation, is the second sequence element in the technical information sequence of the second predicted bidder with the ordinal number 2 corresponding to the bidder under investigation, is the technical information sequence of the second predicted bidder with the ordinal number 2 corresponding to the bidder under investigation. sequence elements, The ordinal number of the corresponding bidder is The first sequence element in the technical information sequence of the second predicted bidder, The ordinal number of the corresponding bidder is The second sequence element in the technical information sequence of the second predicted bidder, The ordinal number of the corresponding bidder is The second prediction of the technical information sequence of the bid-rigging party sequence elements, The second predicted number of collusion parties corresponding to the bidder under investigation.
[0137] In this embodiment, the ranking result of the first predicted bidders for collusion is the result of defining the ordinal numbers of all the first predicted bidders for each investigated bidder starting from 1 according to the principle of descending similarity with the business information of the corresponding investigated bidder. When multiple predicted bidders for collusion have the same business information similarity with the corresponding investigated bidder, the sorting order of the corresponding multiple predicted bidders for collusion in the ranking result of the first predicted bidders for collusion is determined according to the principle of descending value of the first sequence element in the corresponding business information sequence. If the values of the first sequence elements in multiple business information sequences are the same, the sorting is continued according to the principle of descending value of the second sequence element in the business information sequence until all predicted bidders for collusion with the same business information similarity with the corresponding investigated bidder are sorted.
[0138] In this embodiment, the ranking result of the second predicted bidders for collusion is the result of defining the ordinal numbers of all the second predicted bidders for each investigated bidder starting from 1 according to the principle of descending similarity in technical information with the corresponding investigated bidder. When multiple predicted bidders for collusion have the same technical information similarity with the corresponding investigated bidder, the sorting order of the corresponding multiple predicted bidders in the ranking result of the second predicted bidders for collusion is determined according to the principle of descending numerical values of the first sequence element in the corresponding technical information sequence. If the numerical values of the first sequence elements in multiple technical information sequences are the same, the sorting is continued according to the principle of descending numerical values of the second sequence element in the technical information sequence until all predicted bidders for collusion with the same technical information similarity with the corresponding investigated bidder are sorted.
[0139] The beneficial effect of the above technology is: based on the business information sequences of all first predicted bidders who have colluded in bids and the technical information sequences of all second predicted bidders who have colluded in bids for each investigated bidder, an identification matrix of the corresponding investigated bidder is obtained, that is, a quantitative analysis of the performance characteristics of whether there is collusion between other investigated bids that may have collusion with the corresponding investigated bidder is achieved. This embodiment provides a specific implementation method for constructing an identification matrix of the corresponding investigated bidder based on the business information sequences of all first predicted bidders who have colluded in bids and the technical information sequences of all second predicted bidders who have colluded in bids for each investigated bidder. Embodiment 9:
[0140] Based on the embodiment, a method for identifying and analyzing bid-rigging behaviors based on big data obtains the identification and analysis results of bid-rigging behaviors of all investigated bidders based on the identification matrix of all investigated bidders, including:
[0141] If the rank of the identification matrix of a single inspected bidder is greater than a preset rank threshold, the identification analysis result of the bid-rigging behavior of the corresponding inspected bidder is determined to be a bid-rigging behavior.
[0142] In this embodiment, the preset rank threshold is a preset threshold of the rank of the identification matrix used as a reference for determining whether there is bid rigging between the corresponding investigated bidder and all corresponding first predicted bid rigging parties and all second predicted bid rigging parties.
[0143] The beneficial effects of the above technology are: according to the identification matrix of all investigated bidders, the identification and analysis results of the bid-rigging behavior of all investigated bidders are accurately obtained, and whether there is bid-rigging behavior between the investigated bidders and the corresponding all first predicted bid-rigging parties and all second predicted bid-rigging parties is analyzed.
[0144] Embodiment 10:
[0145] The present invention provides a system for identifying and analyzing bid-rigging behaviors based on big data, which is used to execute any one of the methods for identifying and analyzing bid-rigging behaviors based on big data in Embodiments 1 to 9, including:
[0146] The review module is used to obtain all uploaded information of all bidders, and conduct a qualification review of the corresponding bidder based on all uploaded information of each bidder to obtain the first qualification review result of the corresponding bidder, and conduct a second qualification review of each bidder whose first qualification review result is passed to obtain the second qualification review result of the corresponding bidder;
[0147] A processing module, used to take all bidders whose second qualification review results are passed as the inspection bidders, and obtain the business information sequence and technical information sequence of the corresponding inspection bidder based on all uploaded information of each inspection bidder;
[0148] An analysis module is used to obtain all first predicted bidders that may collude in bidding for each investigated bidder based on the business information sequences of all investigated bidders, obtain all second predicted bidders that may collude in bidding for each investigated bidder based on the technical information sequences of all investigated bidders, and obtain an identification matrix of the corresponding investigated bidder based on the business information sequences of all first predicted bidders that may collude in bidding and the technical information sequences of all second predicted bidders that may collude in bidding for each investigated bidder, and obtain identification analysis results of collusion behavior of all investigated bidders based on the identification matrices of all investigated bidders.
[0149] The beneficial effects of the above technology are as follows: according to all the uploaded information of each bidder, a qualification review is conducted on the corresponding bidder to obtain the first qualification review result of the corresponding bidder, and the bidders that do not have the ability to undertake the bidding project are screened out; for each bidder whose first qualification review result is passed, a second qualification review is conducted to obtain the second qualification review result of the corresponding bidder, and all bidders with associated relationships and possible bid-rigging behaviors are determined; according to all the uploaded information of each inspected bidder, the business information sequence and technical information sequence of the corresponding inspected bidder are obtained, so as to facilitate the subsequent accurate acquisition of the first predicted bid-rigging and second predicted bid-rigging of each inspected bidder; according to the business information sequences of all inspected bidders, the all first predicted bidders for each investigated bidder are obtained, all second predicted bidders for each investigated bidder are obtained according to the technical information sequences of all investigated bidders, and an identification matrix of the corresponding investigated bidder is obtained according to the business information sequences of all first predicted bidders for each investigated bidder and the technical information sequences of all second predicted bidders, so as to facilitate the judgment and analysis on whether there is bid-rigging between the investigated bidder and the corresponding all first predicted bid-rigging and all second predicted bid-rigging, and accurately obtain the identification and analysis results of the bid-rigging of all investigated bidders according to the identification matrix of all investigated bidders, that is, to judge whether there is bid-rigging between multiple bidders from both business information and technical information.
[0150] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.
Claims
1. A method for identifying and analyzing bid-rigging behavior based on big data, characterized in that: include: S1: Obtain all uploaded information of all bidders, and conduct a qualification review of the corresponding bidder based on all uploaded information of each bidder, obtain the first qualification review result of the corresponding bidder, and conduct a second qualification review on each bidder whose first qualification review result is passed, and obtain the second qualification review result of the corresponding bidder; S2: All bidders whose second qualification review results are passed are taken as the inspection bidders, and based on all the uploaded information of each inspection bidder, the business information sequence and technical information sequence of the corresponding inspection bidder are obtained; S3: based on the business information sequences of all the inspected bidders, all the first predicted bid-rigging parties of each inspected bidder are obtained; based on the technical information sequences of all the inspected bidders, all the second predicted bid-rigging parties of each inspected bidder are obtained; based on the business information sequences of all the first predicted bid-rigging parties of each inspected bidder and the technical information sequences of all the second predicted bid-rigging parties of each inspected bidder, an identification matrix of the corresponding inspected bidder is obtained; based on the identification matrices of all the inspected bidders, identification and analysis results of the bid-rigging behaviors of all the inspected bidders are obtained; Wherein, all first predicted bid-collusion parties of each investigated bidder are obtained based on the business information sequences of all investigated bidders, including: Based on the business information sequences of the two inspected bidders, the similarity of the business information of the two inspected bidders is obtained, which is: Among them, β is the similarity of the business information of the two investigated bidders currently calculated, Q i is the i-th sequence element in the business information sequence of the first of the two investigated bidders currently being calculated, W i is the ith sequence element in the business information sequence of the second of the two investigated bidders currently being calculated, and n is the number of sequence elements in the business information sequence of a single investigated bidder; Among all the inspected bidders, two inspected bidders whose business information similarity is greater than a preset first similarity are determined to be first predicted bid-collusion parties; Obtain all first predicted bid-collusion parties of each inspected bidder; Among them, based on the technical information sequence of all inspected bidders, all second predicted bid-rigging parties of each inspected bidder are obtained, including: Based on the technical information sequences of the two inspected bidders, the similarity of the technical information of the two inspected bidders is obtained, which is: Among them, γ is the similarity of the technical information of the two investigated bidders currently calculated, R i is the ith sequence element in the technical information sequence of the first of the two inspected bidders currently being calculated, T i is the ith sequence element in the technical information sequence of the second of the two inspected bidders currently being calculated, and m is the number of sequence elements in the technical information sequence of a single inspected bidder; Among all the inspected bidders, two inspected bidders whose technical information similarity is greater than a preset second similarity are determined to be the second predicted bid-collusive parties; Obtain all second predicted bidders for each inspected bidder; Wherein, based on the business information sequences of all first predicted bid-collusive bidders and the technical information sequences of all second predicted bid-collusive bidders of each investigated bidder, an identification matrix of the corresponding investigated bidder is obtained, including: According to the principle of descending similarity of business information with the corresponding investigated bidder, all first predicted bid-collusive parties of each investigated bidder are defined ordinal numbers starting from 1, and the ranking result of the first predicted bid-collusive parties of each investigated bidder is obtained; According to the principle of descending similarity of technical information with the corresponding inspected bidder, all the second predicted bidders for bid collusion of each inspected bidder are defined ordinal numbers starting from 1, and the ranking result of the second predicted bidders for bid collusion of each inspected bidder is obtained; Based on the first predicted bid-collusion ranking result and the second predicted bid-collusion ranking result of each investigated bidder, the identification matrix of the corresponding investigated bidder is obtained, which is: Where E is the identification matrix of a single inspected bidder, k 11 is the first sequence element in the business information sequence of the first predicted bidder with the ordinal number 1 corresponding to the bidder under investigation, k 12 is the second sequence element in the business information sequence of the first predicted bidder with the ordinal number 1 corresponding to the bidder under investigation, k 1n is the nth sequence element in the business information sequence of the first predicted bidder with the ordinal number 1 corresponding to the investigated bidder, n is the number of sequence elements in the business information sequence of a single investigated bidder, k 21 is the first sequence element in the business information sequence of the first predicted bidder with the ordinal number 2 corresponding to the bidder under investigation, k 22 is the second sequence element in the business information sequence of the first predicted bidder with the ordinal number 2 corresponding to the bidder under investigation, k 2n is the nth sequence element in the business information sequence of the first predicted bidder with the ordinal number 2, corresponding to the bidder under investigation, k j1 is the first sequence element in the business information sequence of the first predicted bidder with ordinal number j, corresponding to the bidder under investigation, k j2 is the second sequence element in the business information sequence of the first predicted bidder with ordinal number j, corresponding to the bidder under investigation, k jn is the nth sequence element in the business information sequence of the first predicted bidder with ordinal number j corresponding to the investigated bidder, j is the number of the first predicted bidder corresponding to the investigated bidder, u 11 is the first sequence element in the technical information sequence of the second predicted bidder with the ordinal number 1 corresponding to the bidder under investigation, u 12 is the second sequence element in the technical information sequence of the second predicted bidder with the ordinal number 1 corresponding to the bidder under investigation, u 1m is the mth sequence element in the technical information sequence of the second predicted bidder with ordinal number 1 corresponding to the investigated bidder, m is the number of sequence elements in the technical information sequence of a single investigated bidder, u 21 is the first sequence element in the technical information sequence of the second predicted bidder with the ordinal number 2, corresponding to the bidder under investigation, u 22 is the second sequence element in the technical information sequence of the second predicted bidder with the ordinal number 2 corresponding to the bidder under investigation, u 2m is the mth sequence element in the technical information sequence of the second predicted bidder with the ordinal number 2, corresponding to the bidder under investigation, u g1 is the first sequence element in the technical information sequence of the second predicted bidder with ordinal number g, corresponding to the bidder under investigation, u g2 is the second sequence element in the technical information sequence of the second predicted bidder with ordinal number g, corresponding to the bidder under investigation, u gm is the mth sequence element in the technical information sequence of the second predicted bid-collusive party with ordinal number g corresponding to the investigated bidder, and g is the number of the second predicted bid-collusive party corresponding to the investigated bidder; Among them, based on the identification matrix of all inspected bidders, the identification and analysis results of the bid-rigging behaviors of all inspected bidders are obtained, including: If the rank of the identification matrix of a single inspected bidder is greater than a preset rank threshold, the identification analysis result of the bid-rigging behavior of the corresponding inspected bidder is determined to be a bid-rigging behavior.
2. The method for identifying and analyzing bid-rigging behavior based on big data according to claim 1 is characterized in that: Obtain all uploaded information of all bidders, and conduct a qualification review of the corresponding bidder based on all uploaded information of each bidder, and obtain the first qualification review result of the corresponding bidder, including: Obtain all uploaded information of all bidders, including the bid documents and all qualification certificates of the bidders; If the qualification certificates of each category of bidders of each bidder meet the preset certificate requirements of the corresponding category, the first qualification review result of the corresponding bidder will be set to passed; otherwise, the first qualification review result of the corresponding bidder will be set to failed, and the first qualification review results of all bidders will be obtained.
3. The method for identifying and analyzing bid-rigging behavior based on big data according to claim 1 is characterized in that: Conduct a second qualification review for each bidder whose first qualification review result is passed, and obtain the second qualification review result of the corresponding bidder, including: Obtain the company names of all bidders whose first qualification review results are passed, and obtain the actual beneficiaries of the corresponding bidders based on the company name of each bidder and the preset equity penetration analysis platform; Compare the actual beneficiaries of all bidders. If the actual beneficiaries of different bidders are the same, the second qualification review results of all bidders with the same actual beneficiaries will be set as failed; Among all bidders whose first qualification review results are passed, the second qualification review results of all bidders except those whose second qualification review results are failed will be set as passed.
4. The method for identifying and analyzing bid-rigging behavior based on big data according to claim 1 is characterized in that: S2: All bidders whose second qualification review results are passed are taken as the inspection bidders. Based on all the uploaded information of each inspection bidder, the business information sequence and technical information sequence of the corresponding inspection bidder are obtained, including: All bidders whose second qualification review results are passed are regarded as inspected bidders, and the business information of the corresponding inspected bidder is obtained based on the bid of each inspected bidder, wherein the business information includes the total bid price and the price list of main items; Obtaining technical information of the corresponding inspected bidder based on the bid of each inspected bidder, wherein the technical information includes the project plan; Based on the business information of each inspected bidder, a business information sequence of the corresponding inspected bidder is obtained; based on the technical information of each inspected bidder, a technical information sequence of the corresponding inspected bidder is obtained.
5. The method for identifying and analyzing bid-rigging behavior based on big data according to claim 4 is characterized in that: Based on the business information of each inspected bidder, a business information sequence of the corresponding inspected bidder is obtained; based on the technical information of each inspected bidder, a technical information sequence of the corresponding inspected bidder is obtained, including: The total bid price of each investigated bidder is taken as the first sequence element of the business information sequence, and all main items in the main item price list are sorted according to the position of the main items in the main item price list to obtain the main item sorting result of each investigated bidder; The price value of each main item in the main item price list is used as a sequence element of the business information sequence, and the relative positions of the sequence elements corresponding to all the main items in the main item price list are consistent with the main item sorting result, so as to obtain the business information sequence of the corresponding inspection bidder; Eliminate all non-statistical words in the project plan of each inspected bidder, obtain the key text of the corresponding project plan, take the words that appear more than 2 times in the key text of each project plan as the plan words, obtain the word frequency of each plan word in the project plan of each inspected bidder, and sort all the plan words in the project plan of each inspected bidder from large to small according to the word frequency, and obtain the sorting result of the plan words of the corresponding inspected bidder; The first preset number of solution words in the solution word ranking result of each investigated bidder are used as selected solution words, and based on the word frequency of each selected solution word and the total word frequency of all solution words in the corresponding solution word ranking result, the sequence element representation of the corresponding selected solution word is determined; The sequence element representations of all selected proposal words of each inspected bidder are used as the sequence elements of the technical information sequence of the corresponding inspected bidder to obtain the technical information sequence of the corresponding inspected bidder, wherein the sequence positions of all sequence elements in the technical information sequence are consistent with the sequence order of the first appearance of all sequence elements in the corresponding project proposal.
6. A big data-based bid-rigging behavior identification and analysis system, characterized in that: The method for identifying and analyzing bid-rigging behavior based on big data according to any one of claims 1 to 5 comprises: The review module is used to obtain all uploaded information of all bidders, and conduct a qualification review of the corresponding bidder based on all uploaded information of each bidder to obtain the first qualification review result of the corresponding bidder, and conduct a second qualification review of each bidder whose first qualification review result is passed to obtain the second qualification review result of the corresponding bidder; A processing module, used to take all bidders whose second qualification review results are passed as the inspection bidders, and obtain the business information sequence and technical information sequence of the corresponding inspection bidder based on all uploaded information of each inspection bidder; An analysis module is used to obtain all first predicted bidders that may collude in bidding for each investigated bidder based on the business information sequences of all investigated bidders, obtain all second predicted bidders that may collude in bidding for each investigated bidder based on the technical information sequences of all investigated bidders, and obtain an identification matrix of the corresponding investigated bidder based on the business information sequences of all first predicted bidders that may collude in bidding and the technical information sequences of all second predicted bidders that may collude in bidding for each investigated bidder, and obtain identification analysis results of collusion behavior of all investigated bidders based on the identification matrices of all investigated bidders.
Citation Information
Cited By
Bidding enterprise abnormal behavior detection method based on bidding and tendering data
CN121834536A
