Big data processing-based bidding behavior identification method and system

Abstract

The invention, which belongs to the field of the constellation bid identification technology, discloses a big data processing-based constellation bid behavior identification method and system, the system comprises an address acquisition module, a comparison module, a processing module, a similarity analysis module, a calculation module, a weighted calculation evaluation module and an early warning module, the address acquisition module, the comparison module, the processing module, the similarity analysis module, the calculation module, the weighted calculation evaluation module and the early warning module; and the module is used for acquiring IP address information and MAC address information of all bidding parties, copying and storing all the address information to the address database, and the module can be automatically or manually started to acquire equipment address information when bidding files submitted by all the parties participating in bidding are submitted through the network interface. According to the method and the device, the accuracy of identifying bidding behaviors is improved by comprehensively analyzing the address information of the bidders, the similarity of the bidding file contents, the bidding behaviors and other data in multiple dimensions, a large amount of bidding data can be quickly processed by utilizing a big data processing technology, the detection period is shortened, and the working efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of bid rigging identification, and specifically to a bid rigging identification method and system based on big data processing. Background Art

[0002] In bidding activities, bid rigging and bid collusion is a common illegal means, which undermines the principle of fair competition and harms the interests of the tenderer. Traditional bid rigging and bid collusion identification mainly relies on manual review, which is inefficient and easily affected by subjective factors. With the development of Internet technology, electronic bidding has become a trend, and the generation of a large amount of data has made it possible to automatically identify bid rigging and bid collusion. However, the existing bid rigging and bid collusion identification methods and technologies mostly focus on single-dimensional analysis, which makes it difficult to comprehensively and effectively identify complex and changeable bid rigging and bid collusion. Therefore, in order to be able to identify bid rigging and bid collusion in multiple dimensions, big data technology is used to detect bid data. A bid rigging and bid collusion identification method and system based on big data processing is proposed. Summary of the invention

[0003] The purpose of the present invention is to provide a method and system for identifying bid rigging and collusion based on big data processing, so as to solve the problem mentioned in the above background technology that the existing methods and technologies for identifying bid rigging and collusion mostly focus on analysis of a single dimension, and it is difficult to comprehensively and effectively identify complex and changeable bid rigging and collusion.

[0004] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a method and system for identifying bid-rigging behavior based on big data processing, comprising an address acquisition module, a comparison module, a processing module, a similarity analysis module, a calculation module, a weighted calculation evaluation module and an early warning module;

[0005] As a further preferred embodiment of the technical solution: the address acquisition module is used to acquire the IP address information and MAC address information of all bidders, and copy and store all the address information into the address database. The module can be started automatically or manually, and acquires the device address information of the bidding documents submitted by the parties participating in the bidding through the network interface. After the information is collected, it will be copied and stored in a special address database for subsequent analysis;

[0006] The comparison module is used to compare the acquired address information with the address information in the address database. If there is more than one address information in the address database that is the same as the address information, an abnormality message is output. If there is no duplication, the next step is entered. The comparison module checks the address database to confirm whether two or more different bidders use the same address information. If duplication is found, an abnormality alarm process is triggered. Otherwise, the bidding document is allowed to enter the subsequent analysis stage.

[0007] As a further preferred embodiment of the present technical solution: the processing module is used to obtain the bidding documents of the bidders, and to disassemble them into titles, chapters and contents, etc., and to store them by categories. The processing module is responsible for preliminary processing of key parts of the bidding documents such as titles, chapter titles and text contents, and to classify and store such information according to a predefined format;

[0008] As a further preferred embodiment of the technical solution: the similarity analysis module is used to perform similarity analysis on the chapters of the bidding document and the chapters of other bidding documents, and to perform similarity analysis on the content of the bidding document and the content of other bidding documents, so as to obtain a first similarity and a second similarity, and the similarity analysis module adopts a similarity algorithm of text cosine similarity and Jaccard similarity to evaluate the similarity between the chapters and contents of different bidding documents;

[0009] As a further preferred embodiment of the technical solution: the calculation module is used to determine the bid similarity according to the preset first weight, the second weight, the first similarity and the second similarity, and the calculation module calculates the overall similarity score of the bid documents according to the preset weight and similarity analysis results, the weight reflects the relative importance of the chapters and contents in the overall evaluation, and the similarity score directly reflects the similarity between the bid documents;

[0010] As a further preferred embodiment of the present technical solution: the weighted calculation evaluation module is used to weight the bidding behavior, bid document information, and the comprehensive score of the bidder by weighted calculation. The weighted calculation evaluation module comprehensively considers the bidder's behavior, the information quality of the bid document, and the background factors of the bidder, and gives a comprehensive score to each bidder through a specific scoring mechanism;

[0011] As a further preferred embodiment of the present technical solution: an early warning module is used to output alarm information according to the result of weighted calculation. The early warning module determines whether there is bid rigging or collusion based on the result of comprehensive scoring. Once an abnormality is found, an abnormality report will be immediately generated and relevant personnel will be notified;

[0012] As a further preferred embodiment of the present technical solution: the weighted calculation evaluation module is also used to perform weighted calculation on the bidding behavior, bid document information, and comprehensive scores of the bidders, perform comprehensive scores according to preset weight coefficients, and output alarm information through the early warning module. The weighted calculation evaluation module not only calculates single scores, but also combines scores of bidding behavior, bid document information, and bidders' background information, and obtains a final comprehensive score through preset weight coefficients. This comprehensive score result will be directly fed back to the early warning module to determine whether there is bid rigging or collusion, and trigger an alarm when necessary;

[0013] As a further preferred embodiment of the technical solution: the similarity analysis module is also used to analyze the format consistency of the bidding documents, including font, font size, and line spacing, and incorporate the format similarity into the first similarity calculation; the address acquisition module is also used to regularly update the address database, and can automatically detect and delete invalid or expired address records. In addition to collecting address information during the bidding process, the address acquisition module will also regularly check the address database to ensure that the address information therein is up to date, and any invalid or expired address records will be automatically deleted to ensure the accuracy and effectiveness of the database;

[0014] As a further preferred embodiment of the present technical solution: the calculation module can set different first weights and second weights according to different project types to meet the special needs of different types of projects, and the calculation module can adjust the first weights and second weights according to different types of projects.

[0015] Compared with the prior art, the present invention has the following beneficial effects:

[0016] 1. The present invention improves the accuracy of identifying bid-rigging and bid-collusion behaviors by comprehensively analyzing data of multiple dimensions such as bidders' address information, similarity of bid document contents, and bid behaviors. By using big data processing technology, a large amount of bid data can be quickly processed, thus shortening the detection cycle and improving work efficiency.

[0017] 2. The calculation module in the present invention can adjust the weight of similarity according to the characteristics of different projects, so that the system can adapt to the special needs of different types of projects. It not only pays attention to the content similarity of the bidding documents, but also takes into account factors such as format consistency, making the detection more comprehensive. The address acquisition module can automatically update the address database and automatically delete invalid or expired address records, reducing the workload of manual maintenance and ensuring the accuracy and effectiveness of the database. The early warning module can output alarm information in real time according to the results of weighted calculation. Once abnormal behavior is found, an abnormal report is immediately generated and relevant personnel are notified to ensure that the problem can be handled in a timely manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a structural schematic diagram of a method and system for identifying bid-rigging behavior based on big data processing according to the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Example

[0021] See also Figure 1 As shown, the present invention provides a technical solution: a method and system for identifying bid-rigging behavior based on big data processing, including an address acquisition module, a comparison module, a processing module, a similarity analysis module, a calculation module, a weighted calculation evaluation module and an early warning module.

[0022] In this embodiment, specifically: the address acquisition module is used to obtain the IP address information and MAC address information of all bidders, and copy and store all the address information to the address database. The module can be started automatically or manually, and obtains the device address information of the bidding documents submitted by the parties participating in the bidding through the network interface. After the information is collected, it will be copied and stored in a special address database for subsequent analysis;

[0023] The comparison module is used to compare the acquired address information with the address information in the address database. If there is more than one address information in the address database that is the same as the address information, an exception message is output. If there is no duplication, the next step is entered. The comparison module will check the address database to confirm whether there are two or more different bidders using the same address information. If duplication is found, the exception alarm process is triggered. Otherwise, the bidding document is allowed to enter the subsequent analysis stage.

[0024] In this embodiment, specifically: the processing module is used to obtain the bidder's bid document, and disassemble it into titles, chapters and contents, etc., and store them in categories. The processing module is responsible for preliminary processing of key parts of the bid document such as titles, chapter titles, and text contents, and classifies and stores this information according to a predefined format.

[0025] In this embodiment, specifically: the similarity analysis module is used to perform similarity analysis on the chapters of the bidding document and the chapters of other bidding documents, and to perform similarity analysis on the content of the bidding document and the content of other bidding documents to obtain a first similarity and a second similarity. The similarity analysis module uses similarity algorithms such as text cosine similarity and Jaccard similarity to evaluate the similarity between chapters and contents of different bidding documents.

[0026] In this embodiment, specifically: the calculation module is used to determine the bid similarity according to the preset first weight, second weight, first similarity and second similarity. The calculation module calculates the overall similarity score of the bid document according to the preset weight and similarity analysis results. The weight reflects the relative importance of chapters and contents in the overall evaluation, and the similarity score directly reflects the similarity between the bid documents.

[0027] In this embodiment, specifically: the weighted calculation evaluation module is used to weight the bidder's bidding behavior, bid document information, and the bidder's comprehensive score. The weighted calculation evaluation module comprehensively considers the bidder's behavior, the information quality of the bid document, and the bidder's background factors, and gives a comprehensive score to each bidder through a specific scoring mechanism.

[0028] In this embodiment, specifically: the early warning module is used to output alarm information according to the result of weighted calculation. The early warning module determines whether there is bid rigging or collusion based on the result of comprehensive scoring. Once an abnormality is found, an abnormality report will be generated immediately and relevant personnel will be notified.

[0029] In this embodiment, specifically: the weighted calculation evaluation module is also used to perform weighted calculation on the bidder's bidding behavior, bid document information, and the bidder's comprehensive score, perform a comprehensive score based on a preset weight coefficient, and output alarm information through the early warning module. The weighted calculation evaluation module not only calculates a single score, but also combines the scores of the bidding behavior, bid document information, and the bidder's background information, and derives the final comprehensive score through the preset weight coefficient. This comprehensive score result will be directly fed back to the early warning module to determine whether there is any collusion in bidding, and trigger an alarm when necessary.

[0030] In this embodiment, specifically: the similarity analysis module is also used to analyze the format consistency of the bidding documents, including font, font size, and line spacing, and incorporates the format similarity into the first similarity calculation; the address acquisition module is also used to regularly update the address database, and can automatically detect and delete invalid or expired address records. In addition to collecting address information during the bidding process, the address acquisition module will also regularly check the address database to ensure that the address information therein is up to date, and any invalid or expired address records will be automatically deleted to ensure the accuracy and validity of the database.

[0031] In this embodiment, specifically: the calculation module can set different first weights and second weights according to different project types to meet the special needs of different types of projects, and the calculation module can adjust the first weights and second weights according to different types of projects.

[0032] Working principle or structural principle:

[0033] Step 1: Obtain the address information of all bidders, including IP address information and MAC address information, and copy and store all address information into the address database;

[0034] Step 2: Compare the address information with the address information in the address database; if there is more than one address information in the address database that is the same as the address information, output abnormal information; if there is no duplication, proceed to the next step;

[0035] Step 3: Obtain the bidding documents of the bidders, disassemble the bidding documents into titles, chapters, contents, etc., and store them by categories;

[0036] Step 4: Performing similarity analysis on the sections of the bidding document and the sections of other bidding documents to obtain a first similarity; performing similarity analysis on the content of the bidding document and the content of other bidding documents to obtain a second similarity;

[0037] Step 5: Determine the bid similarity according to the preset first weight, second weight, first similarity and second similarity;

[0038] Step 6: Compare the bid similarity with the preset similarity threshold; if the bid similarity is greater than the preset similarity threshold, it is determined that there is bid rigging and collusion, and an abnormality report is generated; if the bid similarity is less than the preset similarity threshold, it is determined to be qualified and subsequent analysis continues;

[0039] Step 7: Use the project bidding module to issue the project bid, and use the behavior tracking module to track the bidding behavior information, including the IP address of the bidding device, bidding time, bidding document format, bidding document file upload method, etc., and use the scoring mechanism to score the bidding behavior;

[0040] Step 8: Obtain the bidding data of the current bidding, and analyze the bidding document information through the information analysis module, including the bidding subject, project keywords, bidding price, text similarity, uncommon word number similarity and key element area matching similarity, and classify, store and mark the bidding document information, and score the bidding document information through the scoring mechanism;

[0041] Step 9: Use the background investigation module to query the bidder’s business information, including shareholder relationships, joint research and development, registered address, previous transaction relationships, etc., and classify and compare the bidder’s background, and score the bidder through a scoring mechanism;

[0042] Step 10: By weighted calculation of the bidder's bidding behavior, bid document information, and the bidder's comprehensive score, the alarm information is then output through the early warning module; wherein the weighted calculation evaluation module is also used to perform weighted calculation of the bidder's bidding behavior, bid document information, and the bidder's comprehensive score, perform a comprehensive score according to a preset weight coefficient, and output the alarm information through the early warning module. The weighted calculation evaluation module not only calculates a single score, but also combines the scores of the bidding behavior, bid document information, and the bidder's background, and obtains the final comprehensive score through the preset weight coefficient. This comprehensive score result will be directly fed back to the early warning module to determine whether there is collusion in bidding and trigger an alarm when necessary. The similarity analysis module is also used to analyze the format consistency of the bidding documents, including font, font size, and line spacing, and incorporate the format similarity into the first similarity calculation. The address acquisition module is also used to periodically update the address database, and can automatically detect and delete invalid or expired address records to ensure the accuracy and validity of the address database. Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations may be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method and system for identifying bid-rigging behavior based on big data processing, characterized in that: It includes an address acquisition module, a comparison module, a processing module, a similarity analysis module, a calculation module, a weighted calculation evaluation module and an early warning module.

2. According to claim 1, a method and system for identifying bid-rigging behavior based on big data processing, characterized in that: The address acquisition module is used to obtain the IP address information and MAC address information of all bidders, and copy and store all the address information in the address database. The module can be started automatically or manually, and obtains the device address information of the bidding documents submitted by the parties participating in the bidding through the network interface. After the information is collected, it will be copied and stored in a special address database for subsequent analysis; The comparison module is used to compare the acquired address information with the address information in the address database. If there is more than one address information in the address database that is the same as the address information, an exception message is output. If there is no duplication, the next step is entered. The comparison module will check the address database to confirm whether there are two or more different bidders using the same address information. If duplication is found, the exception alarm process is triggered. Otherwise, the bidding document is allowed to enter the subsequent analysis stage.

3. According to claim 2, a method and system for identifying bid-rigging behavior based on big data processing, characterized in that: The processing module is used to obtain the bid documents of the bidder, disassemble them, divide them into titles, chapters and contents, and store them in categories. The processing module is responsible for preliminary processing of key parts of the bid documents such as titles, chapter titles, and text contents, and classifies and stores this information in a predefined format.

4. According to claim 3, a method and system for identifying bid-rigging based on big data processing, characterized in that: The similarity analysis module is used to perform similarity analysis on the chapters of the bidding document and the chapters of other bidding documents, and to perform similarity analysis on the content of the bidding document and the content of other bidding documents to obtain a first similarity and a second similarity. The similarity analysis module uses similarity algorithms such as text cosine similarity and Jaccard similarity to evaluate the similarity between chapters and contents of different bidding documents.

5. According to claim 4, a method and system for identifying bid-rigging behavior based on big data processing, characterized in that: The calculation module is used to determine the bid similarity based on a preset first weight, a second weight, a first similarity and a second similarity. The calculation module calculates the overall similarity score of the bid document based on the preset weight and similarity analysis results. The weight reflects the relative importance of chapters and contents in the overall evaluation, while the similarity score directly reflects the degree of similarity between the bid documents.

6. The method and system for identifying bid-rigging behavior based on big data processing according to claim 5, characterized in that: The weighted calculation evaluation module is used to calculate the bidder's bidding behavior, bid document information, and the bidder's comprehensive score through weighted calculation. The weighted calculation evaluation module comprehensively considers the bidder's behavior, the information quality of the bid document, and the bidder's background factors, and gives a comprehensive score to each bidder through a specific scoring mechanism.

7. The method and system for identifying bid-rigging behavior based on big data processing according to claim 6 is characterized by: The early warning module is used to output alarm information according to the result of weighted calculation. The early warning module determines whether there is bid rigging or collusion based on the result of comprehensive scoring. Once an abnormality is found, an abnormality report will be generated immediately and relevant personnel will be notified.

8. The method and system for identifying bid-rigging behavior based on big data processing according to claim 7 is characterized by: The weighted calculation evaluation module is also used to perform weighted calculation on the bidding behavior, bid information, and comprehensive scores of bidders, perform comprehensive scores based on preset weight coefficients, and output alarm information through the early warning module. The weighted calculation evaluation module not only calculates individual scores, but also combines scores for bidding behavior, bid information, and background information of bidders, and obtains the final comprehensive score through preset weight coefficients. This comprehensive score result will be directly fed back to the early warning module to determine whether there is any collusion in bidding, and trigger an alarm when necessary.

9. The method and system for identifying bid-rigging behavior based on big data processing according to claim 8, characterized in that: The similarity analysis module is also used to analyze the format consistency of the bidding documents, including font, font size, and line spacing, and incorporate the format similarity into the first similarity calculation. The address acquisition module is also used to regularly update the address database and can automatically detect and delete invalid or expired address records. In addition to collecting address information during the bidding process, the address acquisition module will also regularly check the address database to ensure that the address information therein is up to date. Any invalid or expired address records will be automatically deleted to ensure the accuracy and effectiveness of the database.

10. The method and system for identifying bid-rigging behavior based on big data processing according to claim 9, characterized in that: The calculation module can set different first weights and second weights according to different project types to meet the special needs of different types of projects. The calculation module can adjust the first weights and second weights according to different types of projects.

Images