CN114462960A - A method and system for automatic qualification examination in electronic bidding - Google Patents

Abstract

The invention discloses a qualification auditing method in electronic bidding, which comprises the steps of constructing a bidding validity auditing model, analyzing extracted bidding documents and data, and obtaining the validity of the current bid of a bidding enterprise; constructing a bidding security auditing model, and identifying and uniquely authenticating related data of a bidding system to obtain the security of the current bidding of the bidding document; establishing a bidding matching auditing model, and analyzing the catalog of the bidding document to obtain the matching of the bidding request content and the contents of the bidding document; establishing a bidding authenticity auditing model, recognizing and extracting key information of qualification and certificate pictures in a bidding document by using an OCR (optical character recognition), and verifying authenticity to confirm the bidding qualification and the authenticity of the document; and outputting the bidding qualification auditing result of the current bidding enterprise and sending an automatic alarm for the unqualified condition. The method greatly improves the validity, accuracy, precision and efficiency of the audit, and reduces the time and cost of manual audit.

Description

A method and system for automatic qualification examination in electronic bidding

technical field

The invention relates to the fields of data analysis and electronic bidding, in particular to an automatic qualification examination scheme in electronic bidding.

Background technique

Social development and engineering construction are inseparable from bidding and tendering. In recent years, with the deep integration of the Internet and the field of bidding, electronic bidding has become more and more widespread. Electronic bidding can not only improve the efficiency of bidding and procurement, reduce the time, labor, material cost and workload of the bidding process, but also reduce the risk of human intervention in the bidding process. However, in the bidding activities, problems such as bid rigging, bid rigging, fraud, and information mismatch have always been unavoidable, which are the development of the bidding field to varying degrees.

Qualification review before bid evaluation is a key link in the bidding process. Qualification review is often carried out manually. The problems of low review efficiency and high labor costs are more prominent, and there is currently a lack of bidding qualification review in electronic bidding. Therefore, there is a need for an intelligent, efficient, and automated technical solution for qualification review that can deeply analyze text information.

SUMMARY OF THE INVENTION

In order to solve the above-mentioned deficiencies in the prior art, the present invention proposes a qualification review scheme for electronic bidding, which can carry out qualification review work more comprehensively, objectively, efficiently and accurately, and avoid bid-rigging and bid-rigging to a certain extent. , fraud, information mismatch and other problems, reducing invalid bidding, improving bidding efficiency and accuracy, and saving workload have a very good auxiliary role.

In order to achieve the above purpose, the technical solution of the present invention provides a method for automatic qualification examination in electronic bidding, comprising the following steps:

Step 1. Build a bid validity review model, analyze the extracted bid documents and data, and obtain the validity of this bid by the bidding enterprise;

Step 2. Build a bidding security review model, identify and uniquely certify the relevant data of the bidding system, and obtain the security of the bidding documents for this bid;

Step 3, constructing a bid matching review model, analyzing the catalogue of the bidding documents, and obtaining the matching between the content of the bidding requirements and the content of the bidding documents;

Step 4. Build a bidding authenticity review model, use OCR to identify and extract key information of qualifications and certificate pictures in bidding documents, and conduct authenticity verification to confirm the authenticity of bidding qualifications and documents;

Step 5: According to the analysis results of the above steps, output the bidding qualification review result of the current bidding enterprise, and issue an automatic alarm for unqualified conditions.

Moreover, the validity review model described in step 1 is to automatically prepare pre-qualification, bidding document review, valid time review and enterprise credit review for the bidding enterprise, so as to realize validity review. The implementation process of step 1 includes the following sub-steps:

A) Prepare pre-qualification, check the application letter for pre-qualification, legal representative's identity certificate, power of attorney, applicant's basic information table, financial status table in recent years, status table of similar projects completed in recent years, under construction and List of newly undertaken projects, lawsuits and arbitrations that have occurred in recent years, confirm whether the pre-bidding company is qualified for preparation, and mark the results;

B) Review of bidding documents, check whether the bidding system has received the bidding letter and security documents of the bidding company, confirm whether the bidding company has uploaded the necessary bidding documents, and mark the results;

C) Valid time review, check the time when the company uploads the bidding documents and the time when the paper version of the bidding materials is mailed in the bidding system, compare with the predetermined bidding deadline, confirm whether the bidding documents are valid, and mark the results;

D) Enterprise credit review, check the credit status of the bidding enterprise through public information, check whether the credit status of the bidding enterprise complies with the provisions of the bidding documents and national laws and regulations, and mark the results;

E) According to the marking results of steps A), B), C), and D), the bidding validity review result is obtained.

Moreover, the security audit model described in step 2 is to automatically identify the legality of the tenderee's behavior during the bidding process, and uniquely authenticate the bidding documents before and after, so as to ensure the safety of the bidding documents in the bidding system, and avoid the bidding documents from being unauthorised. In the process of collusion or malicious tampering, the implementation process of step 2 includes the following sub-steps:

A) Obtain the identity authentication information and authority information of the registrant from the bidding system, compare and confirm the above information with the preset information of the system, ensure the safety and legality of the login behavior, and mark the result;

B) When the bidder uploads the bidding documents from the bidding system, the unique identification code of the bidding document is obtained through the MD5 algorithm and stored and encrypted with the file. When the bid is opened, the unique identification code of the bidding document is calculated by the MD5 algorithm again, and the unique identification code of the documents is compared. If the unique identification code of the document is exactly the same, it means that the bidder’s bidding document is completely consistent with the bid opening site document, and there is no collusion modification or malicious tampering in the middle, then the bidding document will be decrypted, and the comparison result of the document identification code will be marked;

C) According to the marking results of steps A) and B), the bidding security audit result is obtained.

Moreover, the matching verification model described in step 3 is to calculate the matching of the bidding file name and the matching of the bidding file directory through character matching, and the implementation process of step 3 includes the following sub-steps:

A) According to the requirements of the bidding documents, extract the keywords α ₁ ,α ₂ ,...,α _m in the name of the bidding documents respectively, and form a keyword list of the bidding document name [α ₁ ,α ₂ ,...,α _m ], where m is the number of keywords in the bidding document name, α _i ≠α _j , when the keyword label i≠j and i,j∈{1,2,3,...,m}, input each document Rank the importance of , and calculate the corresponding file preference weight list [k ₁ , _{k 2} ,...,km ] using AHP, where k ₁ + _{k 2} +...+km =1;

B) According to the requirements of the bidding documents, extract the keywords β ₁ , β ₂ ,...,β _n of the titles at all levels in the content of the bidding documents, respectively, to form a list of keywords in the bidding catalog [β ₁ ,β ₂ ,... ,β _n ], where n is the number of keywords in the bidding catalog, β _i ≠β _j , when the keyword labels i≠j and i,j∈{1,2,3,...,n}, input Rank the importance of each title, and use the AHP to calculate the corresponding title preference weight list [l ₁ ,l ₂ ,...,l _n ], where l ₁ +l ₂ +...+l _n =1;

C) Extract the file names of the bidding documents, set M file names, form a list of bidding file names [a ₁ , a ₂ , ..., a _M ], extract the file directory of the bidding documents, and have N file directories, Form a list of bid document directories [b ₁ ,b ₂ ,...,b _N ];

则将文件名关键词α_i所对应的文件偏好权重k_i进行输出；D) Put each keyword α _i in the bidding document name keyword list [α ₁ ,α ₂ ,...,α _m ] into the bidding document name list [a ₁ ,a ₂ ,...,a _M ] Character matching is performed on each filename a _I of, if

Then output the file preference weight _ki corresponding to the file name keyword α _i ;

E) Calculate the file name matching of the bidding documents according to the following file name matching calculation formula:

File name matching = Σk _i *100%

则将目录关键词β_j所对应的标题偏好权重l_j进行输出；F) Put each keyword β _j in the bidding document directory keyword list [β ₁ ,β ₂ ,...,β _n ] into the bidding document name list [b ₁ ,b ₂ ,...,b _N ] Every file in the directory b _J is matched, if

Then output the title preference weight l _j corresponding to the catalog keyword β _j ;

G) Calculate the file directory matching of the bidding documents according to the following file directory matching calculation formula:

File directory matching = Σl _j *100%

H) Compare the file name matching and file directory matching with the corresponding preset thresholds, and output the matching audit result.

Moreover, the authenticity review model described in step 4 uses OCR to identify and extract key information of qualifications and certificate pictures in bidding documents, and automatically performs security verification to ensure the authenticity of bidding qualifications and documents. The implementation process of step 4 includes the following: substep,

A) Use OCR optical character recognition to scan the bidding text, then analyze and process the image file, obtain the qualification name, qualification number, and qualification holder name information in the qualification picture in the bidding document, and store it as a qualification list;

B) Upload the various information of the qualification list to the corresponding inspection website, carry out the authenticity inspection, and output the inspection results;

C) Enter the bidder's name and qualification list into the bidder's qualification database for filing;

D) When the bidding enterprise participates in the bidding, the above steps A)-C) can be carried out to supplement the qualification or create a new enterprise file;

E) The minimum required qualifications required by the current bidding project are necessary elements. When the enterprise qualifications are higher than the bidding qualifications or have additional high-quality qualifications as the preferred elements, compare the enterprise qualifications with the qualifications required for the bidding, and output the qualification grade of the bidding enterprise;

F) Obtain the qualification inspection result and qualification grade of the bidding enterprise according to steps A)-E).

Moreover, the implementation process of step 5 includes the following sub-steps,

A) Compare the effectiveness audit results, safety audit results, matching audit results and authenticity audit results of the bidding enterprise with the corresponding thresholds or conditions, and output the qualification audit results of the bidding enterprise;

B) Calculate the results of all enterprise qualification audits. If the number of enterprises that have passed the qualification audit is greater than or equal to the predetermined threshold, the bid for this project is valid and can be evaluated. Otherwise, the bid is invalid and needs to be re-tendered.

The present invention provides an automatic qualification examination system in electronic bidding, which is used for realizing the above-mentioned automatic qualification examination method in electronic bidding.

Also, the following modules are included,

The first module is used to construct a bid validity review model, analyze the extracted bid documents and data, and obtain the validity of the bid by the bidding enterprise;

The second module is used to build a bidding security review model, identify and uniquely certify the relevant data of the bidding system, and obtain the security of the bidding documents for this bid;

The third module is used to build a bid matching audit model, analyze the catalog of bidding documents, and obtain the matching between the content of the bidding requirements and the content of the bidding documents;

The fourth module is used to build a bidding authenticity review model, use OCR to identify and extract key information of qualifications and certificate pictures in bidding documents, and conduct authenticity verification to confirm the authenticity of bidding qualifications and documents;

The fifth module is used to output the bidding qualification review result of the current bidding enterprise according to the analysis result of the above steps, and issue an automatic alarm for the unqualified situation.

Alternatively, it includes a processor and a memory, the memory is used to store program instructions, and the processor is used to invoke the stored instructions in the memory to execute the above-mentioned method for automatic qualification examination in electronic bidding.

Alternatively, a readable storage medium is included, and a computer program is stored on the readable storage medium, and when the computer program is executed, the above-mentioned method for automatic qualification examination in electronic bidding is implemented.

Through the solution of the present invention, the automatic review of electronic bidding qualifications is realized, the validity, accuracy, precision and efficiency of review of bidding enterprises and bidding documents are greatly improved, and the time and cost of manual review are reduced. It is more comprehensive, the audit result is more objective, the present invention has strong applicability, and is easy to use and popularize.

The solution of the invention is simple and convenient to implement and has strong practicability, solves the problems of low practicability and inconvenient practical application of the related technologies, can improve user experience, and has important market value.

Description of drawings

FIG. 1 is a schematic diagram of the composition and structure of a method for automatic qualification examination in electronic bidding according to an embodiment of the present invention.

Detailed ways

The technical solutions of the present invention will be specifically described below with reference to the accompanying drawings and embodiments.

In the embodiment of the present invention, as shown in FIG. 1 , the automatic qualification verification method in electronic bidding includes the following steps:

Step 1. Build a validity review model, analyze the extracted bidding documents and data, and obtain the validity of the bidding company's bid:

The validity review model is to conduct pre-qualification review, bidding document review, valid time review and enterprise credit review for bidding enterprises, so as to realize validity review.

The preferred implementation process adopted in the embodiment is as follows:

A) Prepare pre-qualification, check the application letter for pre-qualification, legal representative's identity certificate, power of attorney, applicant's basic information table, financial status table in recent years, status table of similar projects completed in recent years, under construction and List of newly undertaken projects, lawsuits and arbitrations that have occurred in recent years, etc., to confirm whether the pre-bidding company is qualified for preparation, and mark the results;

B) Review of bidding documents, check whether the bidding system has received the bidding letter and security documents of the bidding company, confirm whether the bidding company has uploaded the necessary bidding documents, and mark the results;

C) Valid time review, check the time when the company uploads the bidding documents and the time when the paper version of the bidding materials is mailed in the bidding system, compare with the predetermined bidding deadline, confirm whether the bidding documents are valid, and mark the results;

D) Enterprise credit review, check the credit status of the bidding enterprise through public information, check whether the credit status of the bidding enterprise complies with the provisions of the bidding documents and national laws and regulations, and mark the results;

E) According to the marking results of steps A), B), C), and D), the bidding validity review result is obtained.

Step 2. Build a security audit model, identify and uniquely authenticate the relevant data of the bidding system, and obtain the security of the bidding documents for this bid:

The security audit model described is to identify the legitimacy of the tenderee's behavior during the bidding process, and to uniquely authenticate the bidding documents before and after, to ensure the security of the bidding documents in the bidding system, and to prevent the bidding documents from being colluded or modified halfway or maliciously. tamper.

A) Obtain the identity authentication information and authority information of the registrant from the bidding system, compare and confirm the above information with the preset information of the system, ensure the safety and legality of the login behavior, and mark the result;

B) When the bidder uploads the bidding documents in the bidding system, the system obtains the unique identification code of the bidding document through the MD5 algorithm and stores and encrypts it with the file. When the bid is opened, the bidding system calculates the unique identification code of the bidding document again through the MD5 algorithm. Consistent, indicating that the bidder's bidding documents are completely consistent with the bid opening site documents, and there is no collusion between the bidder and the bidder or malicious tampering with hacking techniques, then the system decrypts the bidding documents and marks the comparison results of the document identification codes;

The MD5 algorithm is the message digest algorithm, a widely used cryptographic hash function that can generate a 128-bit (16-bit) hash value to ensure complete and consistent information transmission. The embodiment preferably uses the MD5 algorithm to obtain the unique identification code of the bidding document.

C) According to the marking results of steps A) and B), the bidding security audit result is obtained.

Step 3. Build a matching review model, analyze the catalogue of the bidding documents, and obtain the matching between the content of the bidding requirements and the content of the bidding documents:

The described matching check model is to calculate the matching of the bidding file name and the matching of the bidding file directory through character matching, and step 3 includes:

According to the requirements of the bidding documents, the keywords α ₁ ,α ₂ ,...,α _m in the name of the bidding documents are extracted respectively to form a keyword list [α ₁ ,α ₂ ,...,α _m ] of the bidding document name, where α _i ≠α _j , when i≠j and the keyword labels i,j∈{1,2,3,...,m}, the important information of each document can be input according to the pre-obtained experience of the bid evaluation experts. Ranking, using AHP to calculate the corresponding document preference weight list [k ₁ ,k ₂ ,...,km ], where _{k 1} +k ₂ +...+km =1; where _m is the bid The number of keywords in the file name; AHP is specifically implemented as the prior art, and will not be repeated in the present invention;

According to the requirements of the bidding documents, the keywords β ₁ , β ₂ ,...,β _n of the titles of all levels in the content of the bidding documents are extracted respectively to form a keyword list of the bidding catalog [β ₁ ,β ₂ ,...,β _n ], where β _i ≠β _j , when i≠j and i,j∈{1,2,3,...,n}, input the importance ranking of each title according to the pre-obtained bid evaluation expert experience, and use AHP calculates the corresponding title preference weight list [l ₁ ,l ₂ ,...,l _n ], where l ₁ +l ₂ +...+l _n =1; where n is the keyword in the bidding catalog Number of;

Extract the file names of the bidding documents, set M file names, form a list of bidding file names [a ₁ , a ₂ , ..., a _M ], extract the file directory of the bidding documents, set N file directories, and form a bidding document file directory list [b ₁ ,b ₂ ,...,b _N ];

则将文件名关键词α_i所对应的文件偏好权重k_i进行输出；Compare each keyword α _i in the bidding document name keyword list [α ₁ ,α ₂ ,...,α _m ] with each keyword in the bidding document name list [a ₁ ,a ₂ ,...,a _M ]. a filename _aI for character matching, if

Then output the file preference weight _ki corresponding to the file name keyword α _i ;

The file name matching of bid documents is calculated according to the following file name matching calculation formula:

File name matching = Σk _i *100%,

The formula is to accumulate the file preference weights _ki corresponding to all successfully matched keywords α _i .

则将目录关键词β_j所对应的标题偏好权重l_j进行输出；Compare each keyword β _j in the keyword list [β ₁ ,β ₂ ,...,β _n ] of the bidding document directory with each keyword in the bidding document name list [b ₁ ,b ₂ ,...,b _N ]. A file directory b _J is matched, if

Then output the title preference weight l _j corresponding to the catalog keyword β _j ;

The file directory matching of the bidding documents is calculated according to the following file directory matching calculation formula:

File directory matching = Σl _j *100%,

The formula is to accumulate the title preference weights l _j corresponding to all successfully matched keywords β _j .

Compare the file name matching and file directory matching with the corresponding preset thresholds, (the principle of threshold value is usually related to the actual situation of the project. 90%-95%), and output the matching audit result.

Document matching refers to whether the bidders have all the documents required by the bidding, and the documents are usually not less, so it is reasonable to set the prediction to 100%. Directory matching is whether the directory in the bid is consistent with the required directory. Usually, the number of lines (that is, keywords) in the directory will reach 20-30, and one or two less will have little effect, so the preferred value is 90%. -95%. In actual operation, the threshold can be specifically set in advance by the tenderer according to specific circumstances such as tendering experience and project type.

For the convenience of implementation and reference, the implementation process of step 3 in the embodiment is provided as follows:

A) According to the requirements of the bidding documents, put forward the keywords in the name of the bidding documents, such as "bidding quotation", "commercial part", "technical part", and form the keyword list of the bidding document name 1 ["quotation", "commercial part" ", "Technical part"], obtain the importance ranking of each file according to the experience of the evaluation experts, and use the AHP to calculate the corresponding file preference weight list [0.3, 0.2, 0.5];

B) According to the requirements of the bidding documents, extract the keywords of the titles at all levels in the content of the bidding documents, such as "bid letter", "technical parameters", "performance", and form a keyword list of the bidding catalog 2 ["bid letter", "Technical Parameters", "Performance"], obtain the importance ranking of each document according to the experience of the evaluation experts, and use the AHP to calculate the corresponding title preference weight list [0.3, 0.6, 0.1];

C) Extract the file name of the bidding document, and form the bidding document name list 3 ["bid offer document", "commercial bid part document", "technical bid part document"];

D) Extract the file directory of the bidding documents, and form the list of bidding documents directory 4 ["1. Bidding Letter", "2.3 Product Technical Parameters", "6.5 Company's Supply Performance of the Same Type of Products"];

则“报价”关键词匹配成功，其对应文件偏好权重为0.3；E) Compare the first keyword "quotation" in the keyword list 1 of the bidding document with each of the bidding document name list 3 ["bidding quotation document", "commercial bid part document", "technical bid part document"] File name for character matching to get a string

Then the "quote" keyword matches successfully, and its corresponding file preference weight is 0.3;

则将文件名关键词α_i所对应的文件偏好权重k_i进行输出；F) Sequentially combine each keyword α _i of the bidding document name keyword list [α ₁ ,α ₂ ,...,α _m ] with the bidding document name list [a ₁ ,a ₂ ,..., _am ] matches each filename in _ai , if

Then output the file preference weight _ki corresponding to the file name keyword α _i ;

G) Calculate the file name matching of the bidding documents according to the file name matching calculation formula:

File name matchingness=Σk _i *100%, the formula is to accumulate the file preference weights k _i corresponding to all successfully matched keywords α _i .

则将目录关键词β_j所对应的标题偏好权重l_j进行输出；H) Put each keyword β _i in the bidding document directory keyword list [β ₁ ,β ₂ ,...,β _n ] into the bidding document name list [b ₁ ,b ₂ ,...,b _n ] matches every filename b _j of , if

Then output the title preference weight l _j corresponding to the catalog keyword β _j ;

1) according to the document catalog matching calculation formula, calculate the document catalog matching of bidding documents:

File directory matchingness=Σl _j *100%, the formula is to accumulate the title preference weights l _j corresponding to all successfully matched keywords β _j .

J) Compare the file name matching and file directory matching with a predetermined threshold, and output the matching audit result.

Step 4. Build an authenticity review model, use OCR to identify and extract the key information of qualifications and certificate pictures in bidding documents, and conduct authenticity verification to confirm the authenticity of bidding qualifications and documents:

The authenticity review model described is to use OCR to identify and extract the key information of qualifications and certificate pictures in bidding documents, and to carry out security validation to ensure the authenticity of bidding qualifications and documents.

A) Scan the bidding text with OCR optical character recognition technology, and then analyze and process the image file to obtain the qualification (certificate) name, qualification (certificate) number, qualification (certificate) holder in the qualification (certificate) picture in the bidding document Information such as someone's name, and save it as a list of qualifications;

B) Upload the various information of the qualification list to the corresponding inspection website, carry out the authenticity inspection, and output the inspection results;

C) Enter the bidder's name and qualification list into the bidder's qualification database for filing;

D) When the bidding enterprise participates in the bidding, the above steps A)-C) can be carried out to supplement the qualification or create a new enterprise file;

E) The minimum required qualifications required for the current bidding project are necessary elements. When the enterprise qualifications are higher than the bidding qualifications or have additional high-quality qualifications as the preferred elements, compare the enterprise qualifications with the qualifications required for the bidding, and output the qualification grade of the bidding enterprise. When the bidding enterprise meets the minimum required qualification conditions, it will be marked as qualified bidding enterprise qualification; when the bidding enterprise qualification is higher than the qualification required for bidding, the bidding enterprise will be marked with qualification classification according to the predetermined high-quality qualification conditions;

F) Obtain the qualification inspection result and qualification grade of the bidding enterprise according to steps A)-E).

Step 5. Build and output qualification audit results

A) Compare the effectiveness audit results, safety audit results, matching audit results and authenticity audit results of the bidding enterprise with the corresponding preset thresholds or conditions;

B) The predetermined condition of the validity audit is that in step 1, the results of A), B), C), and D) all meet the standard. If all the results meet the standard, the output validity audit is passed. If there is any non-compliance, the output is valid. If the sex audit is not passed, the items that do not meet the standards will be output;

C) The predetermined condition of the security audit is that the results of A) and B) in step 2 all meet the standards. If all the results meet the standards, the output security audit is passed. If there is any non-compliance, the output security audit is not passed, and output items that do not meet the standard;

D) The predetermined threshold of the matching audit is 100% file name matching, 95% file directory matching, if all meet the standards, the output matching audit is passed, if there is any non-compliance, the output matching audit is not passed;

E) The predetermined condition of the authenticity check has been repeated in E) of step 5;

F) According to the predetermined thresholds of B), C), D), and E) above, determine the qualification review result of the bidding enterprise, and the specific method is that if the four output results are all passed, the bidding qualification review of the output enterprise is passed. , if the comparison result fails, output the doubtful bidding qualification of the enterprise, and output the doubtful items, which can be set to issue an automatic alarm and submit to manual for secondary review;

G) Calculate the results of all enterprise qualification audits. If the number of enterprises that have passed the qualification audit is greater than or equal to 3, the bid for this project is valid and can be evaluated.

H) Output the results of F) and G) above.

During specific implementation, the method proposed by the technical solution of the present invention can be realized by those skilled in the art by using computer software technology to realize the automatic running process. The system device for implementing the method is, for example, a computer-readable storage medium storing a computer program corresponding to the technical solution of the present invention, and a computer that runs the corresponding computer program. The computer equipment of the program should also be within the protection scope of the present invention.

In some possible embodiments, an automatic qualification review system in electronic bidding is provided, including the following modules. The first module is used to construct a bidding validity review model, analyze the extracted bidding documents and data, and obtain a bid The validity of this bid by the enterprise;

The second module is used to build a bidding security review model, identify and uniquely certify the relevant data of the bidding system, and obtain the security of the bidding documents for this bid;

The third module is used to build a bid matching audit model, analyze the catalog of bidding documents, and obtain the matching between the content of the bidding requirements and the content of the bidding documents;

The fourth module is used to build a bidding authenticity review model, use OCR to identify and extract key information of qualifications and certificate pictures in bidding documents, and conduct authenticity verification to confirm the authenticity of bidding qualifications and documents;

The fifth module is used to output the bidding qualification review result of the current bidding enterprise according to the analysis result of the above steps, and issue an automatic alarm for the unqualified situation.

In some possible embodiments, an automatic qualification review system in electronic bidding is provided, including a processor and a memory, the memory is used to store program instructions, and the processor is used to call the stored instructions in the memory to execute the above-mentioned one Qualification automatic review method in electronic bidding.

In some possible embodiments, an automatic qualification review system in electronic bidding is provided, including a readable storage medium, where a computer program is stored on the readable storage medium, and when the computer program is executed, the above-mentioned implementation is realized. An automatic qualification verification method in electronic bidding.

The specific embodiments described herein are merely illustrative of the spirit of the invention. Those skilled in the art to which the present invention pertains can make various modifications or additions to the described specific embodiments or substitute in similar manners, but will not deviate from the spirit of the present invention or go beyond the definitions of the appended claims range.

Claims (10)

1. An automatic qualification auditing method in electronic bidding is characterized by comprising the following steps:

step 1, establishing a bidding validity auditing model, and analyzing the extracted bidding document files and data to obtain the validity of the current bid of the bidding enterprise;

step 2, constructing a bidding security auditing model, and identifying and uniquely authenticating relevant data of the bidding system to obtain the security of the current bidding of the bidding document;

step 3, constructing a bidding matching auditing model, and analyzing the catalog of the bidding document to obtain the matching of the bidding request content and the contents of the bidding document;

step 4, establishing a bidding authenticity auditing model, recognizing and extracting key information of qualification and certificate pictures in the bidding document by using an OCR (optical character recognition), and verifying authenticity to confirm the bidding qualification and the authenticity of the document;

and 5, outputting the bidding qualification auditing result of the current bidding enterprise according to the analysis result of the step, and sending an automatic alarm to the unqualified condition.

2. The method for automatically auditing qualification in electronic bidding according to claim 1, wherein: the validity checking model in the step 1 is to automatically conduct qualification pre-checking, bidding document checking, valid time checking and enterprise credit checking on bidding enterprises so as to realize validity checking, and the implementation process in the step 1 comprises the following substeps,

A) compiling qualification prequalification, checking a qualification prequalification application letter submitted in a bidding system, a legal representative person identification, an authorization submission book, an applicant basic condition table, a recent financial condition table, a recent similar project condition table, a project condition table under construction and new acceptance, a recent litigation and arbitration condition material, confirming whether a prebidding enterprise has qualification, and marking a result;

B) the method comprises the following steps of bidding document examination, checking whether a bidding system receives a bidding function document and a guarantee fund document of a bidding enterprise, confirming whether the bidding enterprise uploads a necessary bidding document, and marking a result;

C) checking the effective time, checking the time of uploading the bidding documents and the time of mailing paper-edition bidding materials by enterprises in the bidding system, comparing the time with the preset bidding deadline time, confirming whether the bidding documents have effectiveness, and marking the result;

D) enterprise credit examination, inquiring the credit condition of the bidding enterprise through public information, checking whether the credit condition of the bidding enterprise accords with the requirements of bidding documents and national laws and regulations, and marking results;

E) and obtaining a bidding validity auditing result according to the marking results of the steps A), B), C) and D).

3. The method for automatically auditing qualification in electronic bidding according to claim 1, wherein: the safety auditing model in step 2 is used for automatically identifying the legality of the behavior of the tenderer in the bidding process, uniquely authenticating the tenderer before and after the tenderer's bidding document, ensuring the safety of the state of the bidding document in the bidding system, and avoiding the bidding document from being modified or maliciously tampered by halfway collusion, the implementation process in step 2 comprises the following substeps,

A) acquiring identity authentication information and authority information of a login person from a bidding system, comparing and confirming the information with system preset information, ensuring the safety and the legality of login behaviors, and marking results;

B) when a bidder uploads a bid file from a bidding system, the unique identification code of the bid file is obtained through an MD5 algorithm and is stored and encrypted along with the file, the unique identification code of the bid file is calculated through an MD5 algorithm again during bid opening, the unique identification codes of the file are compared, the unique identification codes of the file are completely consistent, the file which is bid by the bidder is completely consistent with a bid opening field file, serial modification and malicious tampering do not occur in the midway, the bid file is decrypted, and the comparison result of the identification codes of the file is marked;

C) and obtaining a bidding safety auditing result according to the marking results in the steps A) and B).

4. The method for automatically auditing qualification in an electronic bid according to claim 1, characterized by: the matching auditing model of step 3 is to calculate the matching of the bidding document name and the matching of the bidding document directory by character matching, the step 3 implementation process comprises the following substeps,

A) according to the requirements of the bidding documents, respectively extracting the keywords alpha in the bidding document names₁,α₂,...,α_mForm a keyword list [ alpha ] of the file names to be bid on₁,α₂,...,α_m]Where m is the number of keywords in the bid document name, α_i≠α_jWhen the key word label i is not equal to j and i, j belongs to {1,2, 3.. eta., m }, inputting the importance sequence of each file, and calculating a corresponding file preference weight list [ k ] by utilizing an analytic hierarchy process₁,k₂,...,k_m]Wherein k is₁+k₂+...+k_m＝1；

B) According to the requirement of the bidding document, extracting keywords beta of each grade title in the contents of the bidding document₁,β₂,...,β_nForm a keyword list [ beta ] of the bidding catalog₁,β₂,...,β_n]Where n is the number of keywords in the bid catalog, β_i≠β_jWhen the key word label i is not equal to j and i, j belongs to {1,2, 3.. once, n }, inputting the importance sequence of each title, and calculating a corresponding title preference weight list [ l ] by utilizing an analytic hierarchy process₁,l₂,...,l_n]Wherein l is₁+l₂+...+l_n＝1；

C) Extracting file names of the bidding files, setting M file names to form a bidding file name list [ a ]₁,a₂,...,a_M]Extracting the file directory of the bid document, setting N file directories to form bid document directory list [ b₁,b₂,...,b_N]；

D) Will bid on the file name keyword list [ alpha ]₁,α₂,...,α_m]Each keyword a of_iAnd a list of bid filenames [ a ]₁,a₂,...,a_M]Each file name a in_IPerforming character matching if

The file name keyword alpha is_iCorresponding file preference weight k_iOutputting;

E) and calculating the file name matching property of the bid file according to the following file name matching calculation formula:

matching property of file name ∑ k_i*100％

F) Will bid on the file directory keyword list [ beta ]₁,β₂,...,β_n]Each keyword of (b)_jAnd a list of bid filenames [ b ]₁,b₂,...,b_N]Each file directory b in_JMatch is made if

Then the directory key beta is used_jCorresponding title preference weight l_jOutputting;

G) and calculating the file directory matching of the bid file according to the following file directory matching calculation formula:

file directory matching ═ Σ l_j*100％

H) And respectively comparing the file name matching property and the file directory matching property with corresponding preset thresholds, and outputting a matching property checking result.

5. The method for automatically auditing qualification in electronic bidding according to claim 1, wherein: the authenticity verification model in step 4 is to utilize OCR to recognize and extract key information of qualification and certificate pictures in the bidding document, automatically perform security verification and ensure the authenticity of the bidding qualification and document, the implementation process in step 4 comprises the following substeps,

A) scanning the bidding text by utilizing OCR optical character recognition, analyzing and processing the image file, acquiring the qualification name, the qualification number and the name information of a qualification holder in a qualification picture in the bidding file, and storing the qualification name, the qualification number and the name information as a qualification list;

B) uploading each item of information of the qualification list to a corresponding checking website, checking the authenticity, and outputting a checking result;

C) inputting the name and qualification list of the bidding enterprise into a qualification database of the bidding enterprise for filing;

D) when the bidding enterprises participate in bidding, the steps A) to C) can be carried out, and qualification supplement or new enterprise filing is carried out;

E) the minimum requirement qualification required by the current bid inviting project is a necessary element, when the enterprise qualification is higher than the bid inviting requirement qualification or has additional high-quality qualification as a preferred element, the enterprise qualification is compared with the qualification required by the bid inviting, and the bid enterprise qualification grade is output;

F) and obtaining qualification checking results and qualification grades of the bidding enterprises according to the steps A) to E).

6. The method for automatically auditing qualification in electronic bidding according to claim 1 or 2 or 3 or 4 or 5, characterized in that: the step 5 implementation process includes the following sub-steps,

A) respectively comparing the validity audit result, the security audit result, the matching audit result and the authenticity audit result of the bidding enterprise with corresponding thresholds or conditions, and outputting the qualification audit result of the bidding enterprise;

B) and calculating qualification auditing results of all enterprises, wherein if the number of the enterprises passing the qualification auditing is more than or equal to a preset threshold value, the bidding of the item is effective and can be subjected to bid evaluation, otherwise, the bidding is ineffective and the bid is required to be re-tendered.

7. An automatic qualification auditing system in electronic bidding is characterized in that: method for automated qualification audit in electronic bidding implementing one of the claims 1-6.

8. An automated qualification audit system in electronic bidding according to claim 7, wherein: the bidding effectiveness auditing method comprises a first module and a second module, wherein the first module is used for constructing a bidding effectiveness auditing model, analyzing extracted bidding documents and data and obtaining the effectiveness of the current bidding of a bidding enterprise;

the second module is used for constructing a bidding security auditing model, identifying and uniquely authenticating relevant data of the bidding system and obtaining the security of the current bidding of the bidding document;

the third module is used for constructing a bidding matching auditing model, analyzing the catalog of the bidding document and obtaining the matching of the bidding request content and the bidding document content;

the fourth module is used for constructing a bidding authenticity auditing model, recognizing and extracting key information of qualification and certificate pictures in the bidding document by using an OCR (optical character recognition), verifying authenticity and confirming the bidding qualification and the authenticity of the document;

and a fifth module for outputting the result of checking the tendering qualification of the current bidding enterprise according to the analysis result of the step and sending an automatic alarm for the unqualified condition.

9. An automated qualification audit system in electronic bidding according to claim 7, wherein: comprising a processor and a memory for storing program instructions, the processor being adapted to invoke the stored instructions in the memory to perform a method of automated qualification in electronic bidding according to any one of claims 1-6.

10. An automated qualification audit system in electronic bidding according to claim 7, wherein: comprising a readable storage medium having stored thereon a computer program which, when executed, implements a method for automated qualification review in electronic bidding according to any one of claims 1-6.

Images