Construction method and construction system of power industry bid auditing model and storage medium

Abstract

The embodiment of the invention provides a construction method and a construction system of a power industry bid auditing model, and a storage medium, and belongs to the technical field of power auditing assistance. The construction method comprises the following steps: acquiring multiple categories of bidding data in the power industry; acquiring audit sample data according to the bidding data; constructing a supplier associated data set for the audit sample data, and obtaining a frequent item set according to the supplier associated data set and the audit sample data; obtaining a supplier strong association rule according to the frequent item set; and obtaining a bid auditing conclusion according to the supplier strong association rule. According to the method, the strong association between suppliers can be determined by adopting a mode of obtaining the supplier strong association rule according to the frequent item set, so that whether a surrounding mark behavior exists or not can be quickly and effectively judged, and the auditing accuracy and efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the field of power audit auxiliary technology, and in particular to a method, system and storage medium for constructing a bid-rigging audit model in the power industry. Background Art

[0002] "Bid rigging" is an unfair competition behavior that often occurs during the bidding process. This behavior usually manifests itself as collusion between bidders to influence the bidding results through prior negotiation and cooperation in order to exclude other competitors and increase their chances of winning the bid. In order to effectively curb this behavior, it is crucial to establish an effective audit model to effectively audit and supervise bid rigging.

[0003] At present, the situation of bid rigging in the bidding and tendering of the power industry is mainly judged based on the subjective ability of auditors, and there is no unified standard to measure the situation of bid rigging, which leads to low efficiency and poor effect of bid rigging audit in the power industry.

[0004] The inventors of the present application discovered during the process of implementing the present invention that the above-mentioned solutions in the prior art have the defects of low efficiency and poor effect. Summary of the invention

[0005] The purpose of the embodiments of the present invention is to provide a method, system and storage medium for constructing a bid-rigging audit model in the power industry. The method, system and storage medium for constructing a bid-rigging audit model in the power industry have the function of improving the efficiency and effectiveness of bid-rigging audits in the power industry.

[0006] In order to achieve the above purpose, an embodiment of the present invention provides a method for constructing a bid-rigging audit model in the power industry, comprising:

[0007] Obtain bidding data for multiple categories in the power industry;

[0008] Acquiring audit sample data based on the bidding data;

[0009] Constructing a supplier-related data set for the audit sample data, and acquiring frequent item sets based on the supplier-related data set and the audit sample data;

[0010] Acquire supplier strong association rules according to the frequent itemsets;

[0011] The conclusion of the bid rigging audit is obtained based on the supplier strong association rules.

[0012] Optionally, the bidding data includes the bidding year, bidding date, business category, bidding batch number / package number, purchasing unit code, bidding unit code, winning unit code and name.

[0013] Optionally, the categories of the power industry include power grid maintenance and power grid infrastructure.

[0014] Optionally, obtaining audit sample data based on the bid data includes adopting an annual random sampling method.

[0015] Optionally, constructing a supplier association data set for the audit sample data includes supplier name, business category, and tender batch number / package number.

[0016] Optionally, acquiring frequent itemsets according to the supplier-related data set and the audit sample data includes:

[0017] Presetting a support threshold for the audit sample data;

[0018] Initialize the size k of the current item set to 1;

[0019] Scan the audit sample data to obtain the current candidate data set and the candidate support of each item set in the candidate data set;

[0020] Determine whether the candidate support of the item set in the current candidate data set is greater than or equal to the support threshold;

[0021] In the case where it is determined that the candidate support of the item set in the current candidate data set is greater than or equal to the support threshold, determining that the item set is a frequent item set;

[0022] Obtain the number of elements in the current frequent item set;

[0023] Determine whether the number of elements in the frequent item set is less than k-1;

[0024] When it is determined that the number of elements in the frequent item set is less than k-1, removing the element;

[0025] Self-connecting the frequent item sets to obtain a new candidate data set;

[0026] According to the new candidate data set, delete the elements in the audit sample data that are not in the new candidate data set to obtain the new audit sample data;

[0027] Update the size of the current itemset according to formula (1):

[0028] k=k+1, (1)

[0029] Among them, k is the size of the current itemset;

[0030] Return to the step of scanning the audit sample data to obtain the current candidate data set and the corresponding candidate support.

[0031] Optionally, obtaining the current candidate data set and the corresponding candidate support includes:

[0032] Obtain the candidate support of the item set in the candidate data set according to formula (2);

[0033]

[0034] Among them, Z i is the candidate support of the i-th item set in the candidate data set, n i is the number of occurrences of the i-th item set, N is the total number of item sets in the candidate data set, and i is an integer number.

[0035] Optionally, acquiring a supplier strong association rule according to the frequent itemsets includes:

[0036] Obtaining the confidence of each subset in the final frequent item set;

[0037] The supplier strong association rules are screened according to the confidence of each subset in the final frequent item set.

[0038] On the other hand, the present invention also provides a system for constructing a bid-rigging audit model according to any of the above construction methods, comprising:

[0039] A sample data collection unit for obtaining bidding data of multiple categories in the power industry;

[0040] A frequent itemset extraction unit, connected to the sample data collection unit, for extracting frequent itemsets;

[0041] A strong association rule extraction unit, connected to the frequent itemset extraction unit, and configured to obtain a supplier strong association rule according to the frequent itemsets;

[0042] The audit analysis unit is connected to the strong association rule extraction unit and is used to perform audit analysis according to the supplier strong association rule and obtain an audit conclusion.

[0043] On the other hand, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, wherein the instructions are used to be read by a machine so that the machine executes any of the construction methods described above.

[0044] Through the above technical scheme, the construction method, construction system and storage medium of the power industry bid rigging audit model provided by the present invention obtain bid data of multiple categories in the power industry, and then screen the bid data to obtain audit sample data. A supplier association data set can be constructed for the audit sample data, and then a frequent item set can be obtained. The supplier strong association rules can be obtained based on the frequent item set, and the existence of bid rigging can be determined based on the supplier strong association rules; by adopting the method of obtaining the supplier strong association rules based on the frequent item set, the strong correlation between suppliers can be clarified, and then it can be quickly and effectively determined whether there is bid rigging, thereby improving the accuracy and efficiency of the audit.

[0045] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0047] Figure 1 is a flow chart of a method for constructing a bid-rigging audit model in the power industry according to an embodiment of the present invention;

[0048] Figure 2 It is a flowchart of obtaining frequent item sets in a method for constructing a bid-rigging audit model in the power industry according to an embodiment of the present invention;

[0049] Figure 3 It is a flowchart of obtaining supplier strong association rules in a method for constructing a bid-rigging audit model in the power industry according to an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The specific implementation of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present invention, and is not used to limit the embodiment of the present invention.

[0051] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application are in compliance with the relevant provisions of national laws and regulations. In the embodiments of this application, some existing solutions in the industry such as certain software, components, and models may be mentioned, which should be considered as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of this application, but it does not mean that the applicant has or will necessarily use the solution.

[0052] Figure 1The following is a flow chart of a method for constructing a bid-rigging audit model in the power industry according to an embodiment of the present invention. Figure 1 In, the construction method may include:

[0053] In step S10, multiple categories of bidding data in the power industry are obtained. The bidding categories for the power industry may include power grid maintenance, power grid infrastructure, etc., and the comparison bidding data may include the bidding year, bidding date, business category, bidding batch number / package number, purchasing unit code, bidding unit code, winning unit code and (unit) name, etc.

[0054] In step S11, audit sample data is obtained based on the bidding data. Among them, the bidding data contains a lot of redundant data and irrelevant data, so such data can be eliminated. At the same time, for bid rigging audits, it is generally carried out in an annual random sampling manner, that is, the bidding data of a certain year is selected to obtain the corresponding audit sample data.

[0055] In step S12, a supplier-associated data set is constructed for the audit sample data, and frequent item sets are obtained based on the supplier-associated data set and the audit sample data. Among them, since the selection of suppliers can provide selection opinions for the audit bidding colluded information, the selection of business categories can provide opinions for the selection of audit bidding colluded information of the corresponding suppliers and bidding batch numbers / package numbers. Therefore, the supplier-associated data set can include Q={S,B,P}, where S is the supplier, B is the business category, and P is the bidding batch number / package number. For the above three types of information, at most a frequent "3-item set" may be obtained. Then, the frequent item sets of the audit sample data can be extracted based on the supplier-associated data set, and the frequent item sets are also the final frequent item sets. Specifically, assuming that there are n bidding suppliers, m bidding batch numbers / package numbers, and p business categories, the initial candidate item set contains a total of n+m+p items.

[0056] In step S13, the supplier strong association rules are obtained according to the frequent itemsets, wherein the supplier strong association rules in the annual audit sample data can be determined according to the final frequent itemsets.

[0057] In step S14, the bid rigging audit conclusion is obtained according to the supplier strong association rule. After the supplier strong association rule is obtained, the strong correlation between suppliers can be obtained according to the supplier strong association rule, and it can be determined whether there is bid rigging, that is, the bid rigging audit conclusion can be obtained.

[0058] In step S10 to step S14, first obtain bidding data of multiple categories in the power industry, and then select audit sample data in an annual random sampling manner. Construct a supplier association data set for the audit sample data, and extract frequent item sets from the audit sample data based on the supplier association data set to obtain the final frequent item set. Determine the supplier strong association rule based on the frequent item set, and the supplier strong association rule reflects the bidding relationship between suppliers. Then, it can be determined whether there is a bid rigging situation based on the supplier strong association rule, that is, the bid rigging audit conclusion can be obtained.

[0059] In the traditional power industry, bid rigging in bidding is mainly judged based on the subjective ability of auditors, and there is no unified standard to measure bid rigging, which leads to low efficiency and poor results in power bid rigging audit. In this embodiment of the present invention, the strong correlation between suppliers can be clarified by obtaining the strong association rules of suppliers based on frequent item sets, and then it can be quickly and effectively determined whether there is bid rigging, which improves the accuracy and efficiency of the audit.

[0060] In this embodiment of the present invention, after obtaining the supplier association data set and the audit sample data, the frequent itemsets can be extracted from the audit sample data. The specific extraction steps can be as follows: Figure 2 Specifically, Figure 2 In the construction method, the construction method may further include:

[0061] In step S120, a support threshold of the audit sample data is preset.

[0062] In step S121, the size k of the current item set is initialized to 1.

[0063] In step S122, the audit sample data is scanned to obtain the candidate support of the current candidate data set and each item set in the candidate data set. k Specifically, the calculation of the candidate support of the item set in the candidate data set may include the following formula (2):

[0064] According to formula (2), the candidate support of the item set in the candidate data set is obtained;

[0065]

[0066] Among them, Z i is the candidate support of the i-th item set in the candidate data set, n i is the number of occurrences of the i-th item set, N is the total number of item sets in the candidate data set, and i is an integer number.

[0067] In step S123, it is determined whether the candidate support of the item set in the current candidate data set is greater than or equal to the support threshold. k A certain item set / item set in is compared with the preset support threshold.

[0068] In step S124, when it is determined that the candidate support of the item set in the current candidate data set is greater than or equal to the support threshold, the item set is determined to be a frequent item set. If the candidate support of the item set / item set is greater than or equal to the support threshold, the item set is a frequent item set, otherwise it is not a frequent item set.

[0069] In step S125, the number of elements in the current frequent itemset is obtained.

[0070] In step S126, it is determined whether the number of elements in the frequent itemset is less than k-1.

[0071] In step S127, when it is determined that the number of elements in the frequent itemset is less than k-1, the element is removed.

[0072] In step S128, the frequent itemsets are self-connected to obtain a new candidate data set. The new candidate data set is C k+1 .

[0073] In step S129, elements in the audit sample data that are not in the new candidate data set are deleted according to the new candidate data set to obtain new audit sample data. k+1 , get new sample data, and replace the previous sample data with the new sample data.

[0074] In step S130, the size of the current itemset is updated according to formula (1):

[0075] k=k+1, (1)

[0076] Where k is the size of the current itemset. Specifically, the size of the itemset is the number of elements in the itemset.

[0077] In step S131, the step of scanning the audit sample data to obtain the current candidate data set and the corresponding candidate support is returned. The acquisition condition for frequent item sets may include until no more frequent item sets can be generated, that is, when the candidate support of all item sets in the current candidate item set is less than the preset support.

[0078] In step S120 to step S131, when extracting the final frequent item set, firstly, simply count the frequencies of all items containing one element, and find those item sets that are greater than or equal to the preset support threshold, and then obtain the one-dimensional frequent item set L1. Then start the loop process until no more frequent item sets with better dimensions are generated. Specifically, the loop process may include: in the kth step, generate a k-dimensional candidate item set according to the k-1-dimensional frequent item set generated in the k-1 step. Since the number of elements appearing in the set can be counted when generating k-1 frequent item sets, for a certain element, if its count number is less than k-1, the element can be deleted, thereby eliminating the large-size combination caused by the element. Because if a certain element is to become an element of a k-dimensional item set, the number of counts of the element in the k-1-dimensional frequent item set must reach k-2, otherwise it is impossible to generate a k-dimensional item set. Then, according to the Apriori algorithm, check whether all the k-1-dimensional item sets of the new k-dimensional frequent item set are already included in the k-1-dimensional frequent item set that has been obtained. If one of them is not included, the combination is also deleted, so that a truly useful candidate item set of k-dimensional frequent itemsets can be obtained. After obtaining this candidate item set, each sample / transaction in the audit sample data can be scanned. If the sample / transaction contains at least one candidate item set C k , then the sample / transaction is retained. Otherwise, the sample / transaction record is exchanged with the sample / transaction record at the end of the audit sample data that is not marked for deletion, and the sample / transaction record moved to the end of the audit sample data is marked for deletion. After the entire audit sample data is scanned, it is updated to become the new audit sample data. Therefore, as k increases, the amount of new audit sample data records is greatly reduced, which shortens the scanning time of the next sample and saves the I / O consumption of the computer system, that is, effectively saves computing power. Specifically, the above-mentioned items are the k-sets formed by each selection of elements, such as {supplier}{tender batch number / package number, business category}, a transaction is a complete record, including {business type, supplier, tender batch number / package number}, and an element is a single element in a transaction, such as a supplier.

[0079] In this embodiment of the present invention, after obtaining the final frequent item set, the supplier strong association rule can be obtained according to the frequent item set. The specific steps can be as follows: Figure 3 Specifically, Figure 3 In the construction method, the construction method may further include:

[0080] In step S140, the confidence of each subset in the final frequent itemset is obtained. The confidence can be calculated by using association rules known to those skilled in the art, including dividing the support of the union before and after the association rule by the support before the association rule.

[0081] In step S141, the supplier strong association rules are screened according to the confidence of each subset in the final frequent itemset. For the identification of strong association rules, generally a confidence threshold is preset, and the confidence of each subset is compared with the confidence threshold. If the confidence of the subset is greater than or equal to the confidence threshold, it means that the association rule is a strong association rule.

[0082] In this embodiment of the present invention, the strong association rules obtained when the preset support threshold is 0.05 and the preset confidence threshold is 0.7 are taken as an example, as shown in Table 1.

[0083] Table 1 Strong association rules table

[0084]

[0085]

[0086] According to the sequence number 1 in Table 1, it can be seen that supplier 435 has a close relationship with supplier 735, supplier 653, and supplier 479, which means that the above suppliers often collude in bidding when supplier 435 bids. Therefore, the audit suspicious information generated according to this strong association rule can be that supplier 435 often colludes in bidding with supplier 735, supplier 653, and supplier 479.

[0087] According to No. 2 in Table 1, it can be seen that supplier 246 has a close relationship with supplier 675 and supplier 613, which means that the above suppliers often collude in bidding when supplier 246 bids. Therefore, the audit suspicious information generated according to this strong association rule can be that supplier 246 often colludes with supplier 675 and supplier 613.

[0088] According to No. 3 in Table 1, it can be seen that supplier 640 has a close relationship with supplier 730, supplier 118, and supplier 159, which means that the above suppliers often collude in bidding when supplier 640 bids. Therefore, the audit suspicious information generated according to this strong association rule can be that supplier 640 often colludes in bidding with supplier 730, supplier 118, and supplier 159.

[0089] According to No. 4 in Table 1, it can be seen that supplier 339 has a close relationship with supplier 101, supplier 263, supplier 320, and supplier 219, which means that the above suppliers often collude in bidding when supplier 339 bids. Therefore, the audit suspicious information generated according to this strong association rule can be that supplier 339 often engages in collusion in bidding with supplier 101, supplier 263, supplier 320, and supplier 219.

[0090] On the other hand, the present invention also provides a system for constructing a bid-rigging audit model in the power industry. Specifically, the system may include a sample data collection unit, a frequent itemset extraction unit, a strong association rule extraction unit, and an audit analysis unit.

[0091] The sample data collection unit is used to obtain bidding data of multiple categories in the power industry. The frequent item set extraction unit is connected to the sample data collection unit to extract frequent item sets. The strong association rule extraction unit is connected to the frequent item set extraction unit to obtain the supplier strong association rule according to the frequent item set. The audit analysis unit is connected to the strong association rule extraction unit to perform audit analysis according to the supplier strong association rule and obtain audit conclusions or audit clue analysis information.

[0092] On the other hand, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, and the instructions are used to be read by a machine so that the machine executes any of the above construction methods.

[0093] Through the above technical scheme, the construction method, construction system and storage medium of the power industry bid rigging audit model provided by the present invention obtain bid data of multiple categories in the power industry, and then screen the bid data to obtain audit sample data. The supplier association data set can be constructed for the audit sample data, and then the frequent item set can be obtained. The supplier strong association rules can be obtained according to the frequent item set, and the existence of bid rigging can be determined according to the supplier strong association rules; the strong association between suppliers can be clarified by obtaining the supplier strong association rules according to the frequent item set, and then it can be quickly and effectively determined whether there is bid rigging, thereby improving the accuracy and efficiency of the audit. Specifically, the present invention generates supplier bid rigging audit clue analysis information by performing deep data mining on bid audit data of categories such as maintenance and power grid infrastructure, providing technical support for in-depth analysis of bid audit data of categories such as maintenance and power grid infrastructure and auditing of supplier bid rigging, bringing new technical means to improving the accuracy and efficiency of audits and maintaining fair and transparent management of the power bidding market.

[0094] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0095] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0096] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0097] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0098] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0099] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0100] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0101] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0102] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for constructing a bid-rigging audit model in the power industry, characterized in that: include: Obtain bidding data for multiple categories in the power industry; Acquiring audit sample data based on the bidding data; Constructing a supplier-related data set for the audit sample data, and acquiring frequent item sets based on the supplier-related data set and the audit sample data; Acquire supplier strong association rules according to the frequent itemsets; The conclusion of the bid rigging audit is obtained based on the supplier strong association rules.

2. The construction method according to claim 1, characterized in that: The bidding data includes the bidding year, bidding date, business category, bidding batch number / package number, purchasing unit code, bidding unit code, winning unit code and name.

3. The construction method according to claim 1 or 2, characterized in that: The categories of the power industry include power grid maintenance and power grid infrastructure.

4. The construction method according to claim 1, characterized in that: Obtaining audit sample data based on the bidding data includes adopting an annual random sampling method.

5. The construction method according to claim 1, characterized in that: Constructing a supplier association data set for the audit sample data includes supplier name, business category, and tender batch number / package number.

6. The construction method according to claim 5, characterized in that: Acquiring frequent itemsets according to the supplier association data set and the audit sample data includes: Presetting a support threshold for the audit sample data; Initialize the size k of the current item set to 1; Scan the audit sample data to obtain the current candidate data set and the candidate support of each item set in the candidate data set; Determine whether the candidate support of the item set in the current candidate data set is greater than or equal to the support threshold; In the case where it is determined that the candidate support of the item set in the current candidate data set is greater than or equal to the support threshold, determining that the item set is a frequent item set; Obtain the number of elements in the current frequent item set; Determine whether the number of elements in the frequent item set is less than k-1; When it is determined that the number of elements in the frequent item set is less than k-1, removing the element; Self-connecting the frequent item sets to obtain a new candidate data set; According to the new candidate data set, delete the elements in the audit sample data that are not in the new candidate data set to obtain the new audit sample data; Update the size of the current itemset according to formula (1): k=k+1, (1) Among them, k is the size of the current itemset; Return to the step of scanning the audit sample data to obtain the current candidate data set and the corresponding candidate support.

7. The construction method according to claim 6, characterized in that: Obtaining the current candidate data set and the corresponding candidate support includes: Obtain the candidate support of the item set in the candidate data set according to formula (2); Among them, Z i is the candidate support of the i-th item set in the candidate data set, n i is the number of occurrences of the i-th item set, N is the total number of item sets in the candidate data set, and i is an integer number.

8. The construction method according to claim 1, characterized in that: Acquiring supplier strong association rules based on the frequent itemsets includes: Obtaining the confidence of each subset in the final frequent item set; The supplier strong association rules are screened according to the confidence of each subset in the final frequent item set.

9. A system for constructing a bid-rigging audit model according to the construction method according to any one of claims 1 to 8, characterized in that: include: A sample data collection unit for obtaining bidding data of multiple categories in the power industry; A frequent itemset extraction unit, connected to the sample data collection unit, for extracting frequent itemsets; A strong association rule extraction unit, connected to the frequent itemset extraction unit, and configured to obtain a supplier strong association rule according to the frequent itemsets; The audit analysis unit is connected to the strong association rule extraction unit and is used to perform audit analysis according to the supplier strong association rule and obtain an audit conclusion.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and the instructions are used to be read by a machine so that the machine executes the construction method according to any one of claims 1 to 8.

Images