A bidding document generation method and terminal device

Abstract

A method for generating a bidding document and a terminal device, relating to the field of file processing. In the method, based on a preset weight, historical bidding documents in a historical database whose similarity with the bidding documents exceeds a preset similarity are calculated, and the bidding documents include bidding requirements, laws and regulations; similar contents in the historical bidding documents whose similarity with the bidding documents exceeds a similarity threshold are obtained; historical bidding document contents corresponding to the similar contents are obtained through information retrieval technology; the historical bidding document contents are filled into a preset bidding document template to generate a first target bidding document, and the preset bidding document template is a bidding template corresponding to the bidding documents; based on the first target bidding document, unfilled contents are determined, and unfilled contents are filled through a preset natural language processing algorithm to generate a second target bidding document. The technical solution provided by the present application is implemented to achieve the effect of being able to generate bidding documents that better meet bidding requirements.

Description

Technical Field

[0001] The present application relates to the field of file processing, and in particular to a file generation method and terminal device. Background Art

[0002] In the current bidding document generation process, the industry's common practice is to select historical bids from a library of historical bids that most closely match the current bidding requirements as templates, then modify them based on actual needs. However, the bidding requirements corresponding to these historical bids often differ from the current bidding requirements, making it difficult to directly apply the selected templates. Further complicating matters, with the constant evolution of policies, regulations, and market environments, these discrepancies often widen, causing the generated historical bids to deviate significantly from the current bidding requirements. Consequently, when generating bidding documents, a significant amount of time and effort is required to modify and improve the selected historical bids.

[0003] Therefore, how to generate bidding documents that better meet bidding requirements becomes a problem that needs to be solved. Summary of the Invention

[0004] This application provides a bidding document generation method and terminal device, which can generate bidding documents that better meet bidding requirements.

[0005] In the first aspect, the present application provides a method for generating a bidding document, including: based on a preset weight, calculating the historical bidding materials in the historical database whose similarity with the bidding materials exceeds a preset similarity, the bidding materials including bidding requirements, laws and regulations; obtaining similar content in the historical bidding materials whose similarity with the bidding materials exceeds a similarity threshold; obtaining historical bid content corresponding to the similar content through information retrieval technology; filling the historical bid content into a preset bid template to generate a first target bid, the preset bid template is a bid template corresponding to the bidding materials; determining the unfilled content based on the first target bid, and filling the unfilled content through a preset natural language processing algorithm to generate a second target bid.

[0006] By adopting the above technical solution, historical bidding materials with high similarity in content and structure to the bidding materials and more in line with actual needs are obtained based on preset weights, and then similar content similar to the bidding materials is determined from the screened historical materials, so as to obtain historical bid content that meets the bidding requirements based on the similar content, and fill the historical bid content into the preset bid template. Thereafter, the filled first target bid is analyzed by a preset natural language processing algorithm, and the content that should be filled in the unfilled content is predicted and filled in the unfilled content, so as to obtain a second target bid that meets the requirements.

[0007] Optionally, before calculating the historical bidding materials in the historical database whose similarity with the bidding materials exceeds the preset similarity based on the preset weights, the method also includes: setting initial weights for multiple requirements included in the bidding materials, the multiple requirements including the project background and technical standards included in the bidding requirements; instructing the user to modify the initial weights according to the bidding materials, generate the preset weights, and determine the historical bidding materials based on the preset weights.

[0008] By adopting the above technical solution, by instructing the user to modify the initial weights according to actual needs, the rationality of the weights corresponding to multiple requirements can be further guaranteed, so that the selected historical bidding materials are more suitable for the bidding project needs, thereby improving the similarity of the screened historical bidding materials.

[0009] Optionally, the historical bidding materials include N historical bidding materials, where N is a positive integer greater than 1, and the historical bidding document contents are filled into a preset bidding document template to generate a first target bidding document, including: obtaining partial similar content of each historical bidding document in the N historical bidding materials, whose similarity with the bidding materials exceeds a similarity threshold, to determine M partially similar contents, where the partial similar contents belong to similar contents, and M is a positive integer greater than 1; obtaining N historical bidding document contents corresponding to the M similar contents through information retrieval technology; and filling the N historical bidding document contents into the preset bidding document template through a semantic matching algorithm to generate a first target bidding document.

[0010] By adopting the above technical solution, multiple similar contents are selected from multiple historical bidding documents that have similar overall structures and contents to the bidding documents, and then multiple historical bid contents are obtained based on the multiple similar contents. This can improve the speed of generating bids while ensuring the accuracy and pertinence of the bids.

[0011] Optionally, after determining the unfilled content based on the first target bid document and filling the unfilled content through a preset natural language processing algorithm to generate a second target bid document, the method also includes: obtaining M filling positions of the historical bid document content in the second target bid document; displaying a material viewing link at the M filling positions, the material viewing link is used to respond to the user's material viewing instruction, and display similar materials corresponding to the filled positions to the user, the similar materials including similar bid documents and similar bidding content; determining the content in the similar materials that overlaps with the similar content, and highlighting the overlapping content.

[0012] By adopting the above technical solution and setting a data viewing link, users can understand and analyze the filled content through the data viewing link.

[0013] Optionally, after determining the unfilled content based on the first target bid document and filling the unfilled content through a preset natural language processing algorithm to generate a second target bid document, the method also includes: checking the content to be reviewed in the second target bid document through a preset verification model, instructing the user to review the content to be reviewed; displaying the content to be reviewed to the user; instructing the user to modify the second target bid document based on the material viewing link and bidding materials to generate a third target bid document.

[0014] By adopting the above technical solution, the second target bid document is inspected through a preset inspection model, which can quickly detect the content that needs to be reviewed, and then instruct the user to modify the second target bid document according to the information and bidding requirements corresponding to the filled content, thereby improving the user's modification efficiency.

[0015] Optionally, before verifying the content to be reviewed in the second target bid through a preset verification model and instructing the user to review the content to be reviewed, the method also includes: obtaining historical bid samples; performing data analysis on each historical bid in the historical bid samples to generate data analysis results, the data analysis including structural analysis, content analysis, and error identification; formulating verification rules based on the data analysis results; generating a preset verification model based on the verification rules, so as to determine the content to be reviewed in the second target bid through the preset verification model.

[0016] By adopting the above technical solution, automated inspection of the generated bidding documents can be achieved, which can not only improve inspection efficiency and reduce manual intervention, but also reduce error rate and improve the quality of bidding documents.

[0017] Optionally, before filling the historical bid content into the preset bid template to generate the first target bid, the method also includes: obtaining historical bid samples, the historical bid samples including training bid samples and test bid samples; parsing the bid structure of each historical bid in the historical bid samples, and the content features of each historical bid, to generate multiple bid structures and multiple content features; classifying the historical bid samples based on the multiple bid structures and multiple content features to determine multiple bid categories; training the initial machine learning model based on the determined multiple bid categories to generate a preset machine learning model; generating multiple preset bid templates through the preset machine learning model, the multiple preset bid templates including the preset bid template.

[0018] By adopting the above technical solution and generating a variety of preset bid templates through a preset machine learning model, it is possible to quickly obtain preset bid templates that meet the bidding requirements, thereby improving the speed of generating bidding documents.

[0019] In a second aspect of the present application, a terminal device is provided, including: a calculation module, which is used to calculate, based on a preset weight, historical bidding materials in a historical database whose similarity with the bidding materials exceeds a preset similarity, and the bidding materials include bidding requirements, laws and regulations; an acquisition module, which is used to obtain similar content in the historical bidding materials whose similarity with the bidding materials exceeds a similarity threshold; a retrieval module, which is used to obtain historical bid content corresponding to the similar content through information retrieval technology; a first generation module, which is used to fill the historical bid content into a preset bid template to generate a first target bid, and the preset bid template is a bid template corresponding to the bidding materials; a second generation module, which is used to determine unfilled content based on the first target bid, and fill the unfilled content through a preset natural language processing algorithm to generate a second target bid.

[0020] In a third aspect of the present application, an electronic device is provided, comprising a processor (401), a memory (405), a user interface (403) and a network interface (404), wherein the memory (405) is used to store instructions, the user interface (403) and the network interface (404) are used to communicate with other devices, and the processor (401) is used to execute the instructions stored in the memory (405) so that the electronic device (400) performs any method as described in the first aspect.

[0021] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores instructions. When the instructions are executed, the method steps of any one of the first aspects are executed.

[0022] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0023] 1. By screening out historical bidding materials that are similar to the bidding materials, and then determining similar contents that are similar to the bidding materials from the screened historical materials, we can obtain historical bid contents that are more in line with the bidding requirements through similar contents that meet the overall bidding requirements, and generate bidding documents that meet the bidding requirements, thereby avoiding using historical bidding documents that are only partially similar but do not meet the overall bidding requirements as filler content.

[0024] 2. By selecting multiple similar contents from multiple historical bidding documents that have similar overall structures and contents to the bidding documents, and then obtaining multiple historical bid documents based on the multiple similar contents, the speed of generating bid documents can be improved while ensuring the accuracy and pertinence of the bid documents. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic diagram of the application scenario provided by the embodiment of the present application.

[0026] Figure 2It is a flowchart of the bidding document generation method disclosed in the embodiment of this application.

[0027] Figure 3 This is a structural diagram of a terminal device disclosed in an embodiment of the present application.

[0028] Figure 4 This is a structural diagram of an electronic device disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0029] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0030] In the description of the embodiments of this application, words such as "for example" or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "for example" or "for instance" in the embodiments of this application should not be construed as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "for example" or "for instance" is intended to present the relevant concepts in a concrete manner.

[0031] In the description of the embodiments of the present application, the term "multiple" means two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0032] Generating bidding documents is a crucial task in the current bidding process. To improve the efficiency of generating bidding documents, the industry's common practice is to select multiple historical bids corresponding to the project type from existing historical bids. Then, the bid that most closely matches the project requirements is selected from these historical bids, used as a bid template, and finally modified based on the actual needs of the current project to generate the target bid. However, the historical bidding requirements corresponding to these historical bids often differ from the current bidding requirements. These discrepancies can arise from factors such as project scale, engineering content, technical requirements, and budget range. These discrepancies can easily make the selected templates difficult to directly apply. Further complicating matters, these discrepancies often widen with the constant evolution of policies, regulations, and market environments. These changes can affect core project elements, such as technical requirements and budget ranges, as well as the overall project framework and process. Consequently, these changes can lead to significant deviations between the historical bids generated based on historical bidding requirements and the current bidding requirements. Consequently, the generation of bidding documents requires significant time and effort to modify and improve the selected historical bids.

[0033] The bidding document generation method provided in this application can be applied to the scenario of generating bidding documents. Figure 1 , Figure 1 This is a schematic diagram of an application scenario provided by the embodiment of this application. Figure 1 In the application scenario shown, after logging in, the user calculates the similarity between all historical materials in the historical database and the content of the bidding materials based on the preset weights, and filters out historical bidding materials with a similarity greater than the preset similarity, that is, comparing the various requirements of the historical projects with the various requirements of the bidding projects, and obtaining historical bidding materials corresponding to historical projects similar to the bidding requirements, so that the filtered historical bidding materials match the overall structure and overall content of the bidding materials; then filter out some similar materials that are similar to the bidding materials from the historical bidding materials, that is, similar content; thereby, the historical bid content with a high degree of similarity to the bidding requirements can be determined through similar content that is similar to the overall structure of the project to which it belongs and has a high degree of content similarity; the historical bid content is filled into the preset bid template to generate a first target bid, and then filled with the filled content according to the preset algorithm to generate a second target bid, that is, a bidding document that better meets the bidding requirements.

[0034] based on Figure 1 Please continue to refer to the application scenarios shown in Figure 2 , Figure 2 This is a process 200 of a method for generating a bidding document provided in an embodiment of the present application. The method for generating a bidding document can be applied to any terminal device capable of human-computer interaction.

[0035] The following, combined Figure 2 A method for generating bidding documents in an embodiment of the present application is described in detail.

[0036] Step S201: Based on a preset weight, calculate the historical bidding documents in the historical database whose similarity with the bidding documents exceeds a preset similarity, where the bidding documents include bidding requirements, laws and regulations.

[0037] Bidding documents refer to the collection of relevant documents and information necessary to clarify the specific requirements and conditions of a bidding project during a bidding activity. For example, bidding documents include bidding requirements, laws and regulations, market research reports, bid rejection criteria, and contract terms. For example, bidding requirements may include project background, scope, technical standards, construction period, and budget. It should be understood that the relevant information included in bidding documents is subject to the specific requirements of the bidding project and is not specifically limited in this application.

[0038] Preset weights refer to the pre-set weights for each document included in the tender documents. For example, the tender requirements might have a weight of 0.1 for project background, 0.1 for scope, 0.1 for technical standards, 0.1 for work schedule, and 0.1 for budget.

[0039] In an optional embodiment, before calculating the historical bidding materials in the historical database whose similarity with the bidding materials exceeds the preset similarity based on the preset weights, the method also includes: setting initial weights for multiple requirements included in the bidding materials, the multiple requirements including the project background and technical standards included in the bidding requirements; instructing the user to modify the initial weights according to the bidding materials, generate the preset weights, and determine the historical bidding materials based on the preset weights.

[0040] Multiple requirements refer to every requirement included in the bidding documents, such as project background, technical standards, local laws and policies.

[0041] The initial weight refers to a default weight value used to calculate the similarity with the tender documents. In the embodiment of the present application, the initial weight value is used to calculate the similarity of the tender documents corresponding to any tender project.

[0042] In combination with the above-mentioned bidding requirements, and taking the preset similarity of 0.6 as an example, the initial weights are set for the multiple requirements included in the bidding documents as [project background weight: 0.1, scope weight: 0.1, technical standard weight: 0.1, work period weight: 0.1, budget weight: 0.1]. After reviewing the bidding documents, the user believes that the preset weights do not meet the bidding project requirements, and modifies the preset weights to obtain the preset weights [project background weight: 0.05, scope weight: 0.05, technical standard weight: 0.2, work period weight: 0.1, budget weight: 0.1]. Optionally, to facilitate user review, various weights can be displayed in a hierarchical manner, such as [Tender Requirements Weight: 0.5, Laws and Regulations Weight: 0.1, Market Research Report Weight: 0.2, Bid Rejection Criteria Weight: 0.1, Contract Clause Weight: 0.1]. The weights for the various requirements included in the tender requirements are [Project Background Weight: 0.05, Scope Weight: 0.05, Technical Standards Weight: 0.2, Work Period Weight: 0.1, Budget Weight: 0.1]. A similarity calculation method (such as cosine similarity) is then used to calculate the similarity between the feature vector of each historical tender document and the feature vector of the current tender document, thereby selecting historical tender documents with similarities exceeding a preset similarity threshold. It is easy to understand that the current step is intended to select historical tender documents that are consistent with the tender document in content and structure, so the preset similarity threshold is set low. In actual applications, the preset similarity threshold can be adjusted based on actual needs.

[0043] It should be understood that if the historical bidding documents include fewer data types than the preset data types, the initial weights corresponding to the missing data types will be set to 0. For example, if the historical bidding documents only include project background and technical standards, the initial weights corresponding to scope, duration, and budget will be set to 0. If the user does not modify the initial weights, the similarity with the bidding documents will be calculated based solely on the project background and technical standards. Optionally, if the historical bidding documents include fewer data types than the preset data types, resulting in the calculated similarities being lower than the preset similarity, all historical bidding documents in the historical database will be sorted in descending order based on the similarity values, and the first arbitrary number (e.g., 3) of historical bidding documents will be considered as historical bidding documents whose similarity exceeds the preset similarity.

[0044] In this embodiment, by instructing the user to modify the initial weights according to actual needs, the rationality of the weights corresponding to multiple requirements can be further guaranteed, so that the selected historical bidding materials are more suitable for the bidding project requirements, thereby improving the similarity of the screened historical bidding materials.

[0045] Step S202: obtaining similar content in historical bidding documents whose similarity to the bidding document exceeds a similarity threshold.

[0046] Similar content is used to indicate content in historical bidding documents that is similar to the bidding documents and exceeds a similarity threshold. Similar content can be project background in historical bidding documents or project background and technical standards, which is not specifically limited in this application.

[0047] Taking a similarity threshold of 0.8 as an example, if both the bidding documents and historical bidding documents include project background and technical standards, then the scope, duration, and budget are calculated using a text similarity algorithm to have similarities of 0.81, 0.9, 0.7, 0.75, and 0.6, respectively, then the similar content is project background and technical standards. Alternatively, other similarity calculation methods (such as feature matching algorithms) can be used to calculate similarity to determine similar content, but this application does not specifically limit this.

[0048] Step S203: Obtain historical bidding document contents corresponding to similar contents through information retrieval technology.

[0049] Historical bid content refers to the portion of the bid that corresponds to similar content. For example, if the similar content is the project background, the historical bid content would be the introduction in the bid.

[0050] It should be understood that before conducting information retrieval based on similar content, all historical bids should be pre-processed by tokenizing, removing stop words, and creating indexes to build a database of historical bids. This allows query statements (such as Boolean operators) to be constructed and a retrieval system to be generated. This allows the search to identify historical bids corresponding to similar content.

[0051] Step S204: Fill the historical bid document content into the preset bid document template to generate a first target bid document. The preset bid document template is a bidding template corresponding to the bidding information.

[0052] In an optional embodiment, before filling the historical bid content into the preset bid template to generate the first target bid, the method also includes: obtaining historical bid samples, the historical bid samples including training bid samples and test bid samples; parsing the bid structure of each historical bid in the historical bid samples, and the content features of each historical bid, to generate multiple bid structures and multiple content features; classifying the historical bid samples based on the multiple bid structures and multiple content features to determine multiple bid categories; training the initial machine learning model based on the determined multiple bid categories to generate a preset machine learning model; generating multiple preset bid templates through the preset machine learning model, the multiple preset bid templates including the preset bid template.

[0053] Exemplarily, historical bid samples are obtained, which include training bid samples and test bid samples. The structure of each historical bid is parsed through a document parsing library (such as python-docx, lxml, etc.), and key structural information such as titles, paragraphs, and lists are extracted. Regular expressions are used to extract key content features from each historical bid, such as project background, technical standards, delivery deadlines, etc. Based on the extracted bid structure and content features, a clustering algorithm is used to divide the bids into multiple different categories. According to the classified bid samples, an initial machine learning model (such as a random forest) is trained to obtain a preset machine learning model. The trained machine learning model is then combined with the historical bid samples to generate a variety of preset bid templates, wherein the multiple preset bid templates include a preset bid template.

[0054] In this embodiment, by presetting a machine learning model and generating a variety of preset bid templates, preset bid templates that meet the bidding requirements can be quickly obtained, thereby improving the speed of generating bidding documents.

[0055] In an optional embodiment, the historical bidding materials include N historical bidding materials, where N is a positive integer greater than 1, and the historical bidding document contents are filled into a preset bidding document template to generate a first target bidding document, including: obtaining partial similar content of each historical bidding document in the N historical bidding materials, whose similarity with the bidding materials exceeds a similarity threshold, to determine M partially similar contents, where the partial similar contents belong to similar contents, and M is a positive integer greater than 1; obtaining N historical bidding document contents corresponding to the M similar contents through information retrieval technology; and filling the N historical bidding document contents into the preset bidding document template through a semantic matching algorithm to generate a first target bidding document.

[0056] Combined with the above example, the N historical bidding documents are historical bidding document A and historical bidding document B. The similar contents in historical bidding document A that exceed the similarity threshold with the bidding documents are project background and technical standards, and the similar contents in historical bidding document B are local regulations and policies, that is, M similar contents are 3 similar contents. Then, the historical bidding document content A corresponding to the project background and technical standards, and the historical bidding document content B corresponding to local regulations and policies are obtained through information retrieval technology. Then, the historical bidding document content A and the historical bidding document content B are analyzed through the semantic matching algorithm, and analyzed and matched with the contents that need to be filled in the preset bidding document template. Finally, the introduction and technical requirements in the historical bidding document content A are added to the introduction and technical requirements in the preset bidding document template, and the bidder instructions in the historical bidding document content B are added to the bidder instructions in the preset bidding document template to generate the first target bid.

[0057] It should be understood that if the M similar contents respectively correspond to historical bid contents and also correspond to the same part in the preset bid template, the historical bid contents corresponding to the similar contents with the highest similarity will be filled into the same part.

[0058] It should also be understood that there is a one-to-one correspondence between the N historical bidding materials and the N historical bid document contents.

[0059] In this embodiment, multiple similar contents are selected from multiple historical bidding documents that have similar overall structures and contents to the bidding documents, and then multiple historical bid contents are obtained based on the multiple similar contents. This can improve the speed of generating bids while ensuring the accuracy and pertinence of the bids.

[0060] In another optional embodiment, after obtaining the partial similar content of each historical bidding document in N historical bidding documents and the similarity between the bidding documents exceeds the similarity threshold, the number of partial similar contents corresponding to each historical bidding document is compared, and the historical bid document corresponding to the historical bidding document with the most partial similar content is determined, and then the historical bid document content corresponding to the historical bid document is obtained through information retrieval technology, and the historical bid document content is filled into the preset bid document template to generate the first target bid document.

[0061] Step S205: determining unfilled content based on the first target bid document, and filling the unfilled content using a preset natural language processing algorithm to generate a second target bid document.

[0062] The preset natural language processing algorithm is based on the historical bidding information stored in the historical database. The initial natural language processing algorithm is trained to generate corresponding text to fill the unfilled parts.

[0063] Exemplarily, a preset natural language processing algorithm is used to analyze the structural features of the first target bid document, identify the unfilled content of the first target bid document, and then use the unfilled content as input. Based on the input context, template structure, and the filled content, corresponding text is generated to fill the unfilled portion. Optionally, other algorithms (such as template comparison methods) can also be used to identify the unfilled content, which is not specifically limited in this application.

[0064] The bidding document generation method provided in the embodiment of the present application obtains historical bidding materials that are highly similar to the content and structure of the bidding materials and are more in line with actual needs based on preset weights, and then determines similar content similar to the bidding materials from the screened historical materials, so as to obtain historical bid content that meets the bidding requirements based on the similar content, fill the historical bid content into a preset bid template, and then analyze the filled first target bid through a preset natural language processing algorithm, predict the content that should be filled in the unfilled content, and fill it in the unfilled content, so as to obtain a second target bid that meets the requirements.

[0065] In an optional embodiment, after determining unfilled content based on the first target bid document and filling the unfilled content through a preset natural language processing algorithm to generate a second target bid document, the method also includes: obtaining M filling positions of the historical bid document content in the second target bid document; displaying a material viewing link at the M filling positions, the material viewing link is used to respond to the user's material viewing instruction, and display similar materials corresponding to the filled positions to the user, the similar materials including similar bid documents and similar bidding content; determining the content in the similar materials that overlaps with the similar content, and highlighting the overlapping content.

[0066] Similar bids are the bid contents filled in the corresponding fill-in fields. Similar tender contents are the historical tender documents corresponding to the similar bids in the historical tender documents. Similar materials include similar bids and similar tender contents corresponding to the fill-in fields. Optionally, similar materials may include more content, which is not specifically limited in this application.

[0067] In this embodiment, by setting a data viewing link, the user can understand and analyze the filled content through the data viewing link.

[0068] In an optional embodiment, after determining the unfilled content based on the first target bid document and filling the unfilled content through a preset natural language processing algorithm to generate a second target bid document, the method also includes: checking the content to be reviewed in the second target bid document through a preset verification model, instructing the user to review the content to be reviewed; displaying the content to be reviewed to the user; instructing the user to modify the second target bid document based on the material viewing link and bidding materials to generate a third target bid document.

[0069] Optionally, before verifying the content to be reviewed in the second target bid through a preset verification model and instructing the user to review the content to be reviewed, the method also includes: obtaining historical bid samples; performing data analysis on each historical bid in the historical bid samples to generate data analysis results, the data analysis including structural analysis, content analysis, and error identification; formulating verification rules based on the data analysis results; generating a preset verification model based on the verification rules, so as to determine the content to be reviewed in the second target bid through the preset verification model.

[0070] In some embodiments, historical bid samples covering different project types, industry standards, and format requirements are retrieved from a relevant database. Each historical bid is then analyzed for format, structure, content, and error identification. Based on the analysis results of the historical bids, verification rules applicable to the current project are formulated, each rule is assigned a corresponding weight, and the formulated verification rules are applied to a selected automated verification tool (e.g., a rule engine) to generate a pre-set verification model. In this embodiment, automated verification of generated bidding documents can be achieved, which not only improves verification efficiency and reduces manual intervention, but also reduces error rates and improves the quality of bidding documents.

[0071] For example, the second target bid document is inspected using a preset inspection model, and potential problems are analyzed, such as sections that do not meet the bidding requirements, violations of laws and regulations, logical or grammatical errors, etc. The inspected problems are displayed to the user as content that should be reviewed. Then, by comparing the preset bid template with the first target bid document, M filling positions are determined, and data viewing links are set at each of the M filling positions, instructing the user to obtain similar data through the data viewing links. The second target bid document is then modified in accordance with the bidding document requirements to generate a third target bid document. In this example, the M filling positions can be identified using a tracking algorithm, and this application does not specifically limit this.

[0072] In this embodiment, the second target bid document is checked by a preset inspection model, and the content that needs to be reviewed can be quickly detected. The user is then instructed to modify the second target bid document according to the information corresponding to the filled content and the bidding requirements, which can improve the user's modification efficiency.

[0073] It is understandable that the terminal device is required to achieve Figure 2 The functions described include hardware and / or software modules that perform the corresponding functions. In combination with the steps of the various examples described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application in combination with the embodiments, but such implementation should not be considered to be beyond the scope of this application.

[0074] In this embodiment, the terminal device can be divided into functional modules according to the above-mentioned method example. For example, different functional modules can be divided according to different functions, or two or more functions can be integrated into a single processing module. The above-mentioned integrated modules can be implemented in the form of hardware. It should be noted that the module division in this embodiment is illustrative and only represents a logical functional division. In actual implementation, other division methods may be used.

[0075] In the case of dividing each functional module into corresponding functional modules, Figure 3 A possible schematic diagram of the terminal device 300 involved in the above embodiment is shown, and the terminal device 300 includes: a calculation module 301, which is used to calculate the historical bidding materials in the historical database whose similarity with the bidding materials exceeds a preset similarity based on a preset weight, and the bidding materials include bidding requirements, laws and regulations; an acquisition module 302, which is used to obtain similar content in the historical bidding materials whose similarity with the bidding materials exceeds a similarity threshold; a retrieval module 303, which is used to obtain historical bid content corresponding to the similar content through information retrieval technology; a first generation module 304, which is used to fill the historical bid content into a preset bid template to generate a first target bid, and the preset bid template is a bid template corresponding to the bidding materials; a second generation module 305, which is used to determine unfilled content based on the first target bid, and fill the unfilled content through a preset natural language processing algorithm to generate a second target bid.

[0076] In an optional implementation of an embodiment of the present application, the calculation module 301 is also used to set initial weights for multiple requirements included in the bidding documents, including the project background and technical standards included in the bidding requirements; instruct the user to modify the initial weights according to the bidding documents, generate preset weights, and determine the historical bidding documents based on the preset weights.

[0077] In an optional implementation method of the embodiment of the present application, the first generation module 304 is also used to obtain partial similar content of each of the N historical bidding documents, whose similarity with the bidding documents exceeds a similarity threshold, to determine M partially similar contents, where the partial similar contents belong to similar contents, and M is a positive integer greater than 1; obtain N historical bid contents corresponding to the M similar contents through information retrieval technology; and fill the N historical bid contents into a preset bid template through a semantic matching algorithm to generate a first target bid.

[0078] In an optional implementation method of an embodiment of the present application, the first generation module 304 is also used to obtain M filling positions of the historical bid content in the second target bid; display a material viewing link at the M filling positions, the material viewing link is used to respond to the user's material viewing instruction, and display similar materials corresponding to the filling positions to the user, the similar materials include similar bids and similar bidding contents; determine the content in the similar materials that overlaps with the similar content, and highlight the overlapping content.

[0079] In an optional implementation method of the embodiment of the present application, the first generation module 304 is also used to verify the content that needs to be reviewed in the second target bid through a preset verification model, instruct the user to review the content that needs to be reviewed; display the content that needs to be reviewed to the user; instruct the user to modify the second target bid based on the data viewing link and bidding materials to generate a third target bid.

[0080] In an optional implementation method of the embodiment of the present application, the second generation module 304 is also used to obtain historical bid samples; perform data analysis on each historical bid in the historical bid samples to generate data analysis results, the data analysis including structural analysis, content analysis, and error identification; formulate inspection rules based on the data analysis results; generate a preset inspection model based on the inspection rules, so as to determine the content to be reviewed in the second target bid through the preset inspection model.

[0081] In an optional implementation method of an embodiment of the present application, the first generation module 304 is also used to obtain historical bid samples, which include training bid samples and test bid samples; parse the bid structure of each historical bid in the historical bid samples, as well as the content features of each historical bid, to generate multiple bid structures and multiple content features; classify the historical bid samples based on the multiple bid structures and multiple content features, and determine multiple bid categories; train the initial machine learning model based on the determined multiple bid categories to generate a preset machine learning model; generate multiple preset bid templates through the preset machine learning model, and the multiple preset bid templates include a preset bid template.

[0082] This application also discloses an electronic device. Figure 4 , Figure 4 The electronic device 400 may include: at least one processor 401 , at least one network interface 404 , a user interface 403 , a memory 405 , and at least one communication bus 402 .

[0083] The communication bus 402 is used to implement the connection and communication between these components.

[0084] The user interface 403 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 403 may also include a standard wired interface and a wireless interface.

[0085] The network interface 404 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0086] Processor 401 may include one or more processing cores. Using various interfaces and circuits, processor 401 connects to various components within the server. It executes instructions, programs, code sets, or instruction sets stored in memory 405, as well as accesses data stored in memory 405, to perform various server functions and process data. Optionally, processor 401 may be implemented using at least one of the following hardware forms: a digital signal processing (DSP), a field-programmable gate array (FPGA), or a programmable logic array (PLA). Processor 401 may integrate one or a combination of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. The CPU primarily processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing content displayed on the display screen; and the modem handles wireless communications. It is understood that the modem may also be implemented independently of the processor 401 and implemented as a separate chip.

[0087] Among them, the memory 405 may include a random access memory (RAM) or a read-only memory (Read-Only Memory). Optionally, the memory 405 includes a non-transitory computer-readable storage medium. The memory 405 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 405 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 405 may also optionally be at least one storage device located away from the aforementioned processor 401. Refer to Figure 4 , the memory 405 as a computer storage medium may include an operating system, a network communication module, a user interface module and an application program of a bidding document generation method.

[0088] exist Figure 4In the electronic device 400 shown, the user interface 403 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 401 can be used to call an application program storing a bidding document generation method in the memory 405. When executed by one or more processors 401, the electronic device 400 executes one or more of the methods described in the above embodiments. It should be noted that for the aforementioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should know that this application is not limited by the described order of actions, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for this application.

[0089] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0090] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are merely schematic, such as the division of units, which is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interface, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0091] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0092] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0093] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of this application, or the portion that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the method of this application. The aforementioned memory includes various media that can store program code, such as USB flash drives, mobile hard drives, magnetic disks, or optical disks.

[0094] The foregoing is merely an exemplary embodiment of the present disclosure and is not intended to limit the scope of the present disclosure. In other words, any equivalent variations and modifications made in accordance with the teachings of the present disclosure are still within the scope of the present disclosure. Those skilled in the art will readily conceive of other embodiments of the present disclosure after considering the disclosure and the practical implications thereof.

[0095] This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered merely as exemplary, and the scope and spirit of the present disclosure are to be determined by the claims.

Claims

1. A method for generating bidding documents, characterized in that: include: Based on the preset weights, the similarity between all historical bidding documents and the content of the bidding documents in the historical database is calculated, and the historical bidding documents with similarities exceeding the preset similarity are screened out, wherein the bidding documents include bidding requirements, laws and regulations; Acquire similar contents in the historical bidding documents whose similarity with the bidding documents exceeds a similarity threshold; Acquire historical bidding document contents corresponding to the similar contents through information retrieval technology; Filling the historical bidding document content into a preset bidding document template to generate a first target bidding document, wherein the preset bidding document template is a bidding template corresponding to the bidding information; Determine unfilled content based on the first target bid document, and fill the unfilled content by using a preset natural language processing algorithm to generate a second target bid document; The historical bidding documents include N historical bidding documents, where N is a positive integer greater than 1. Filling the historical bidding document contents into a preset bidding document template to generate a first target bidding document includes: Obtaining partial similar contents of each of the N historical bidding documents, the similarity of which to the bidding document exceeds the similarity threshold, to determine M partial similar contents, the partial similar contents belonging to the similar contents, where M is a positive integer greater than 1; Obtaining N historical bidding document contents corresponding to the M partially similar contents through information retrieval technology; The N historical bid documents are filled into the preset bid document template through a semantic matching algorithm to generate the first target bid document.

2. The method according to claim 1, characterized in that: Before calculating the similarity between all historical bidding documents and the content of the bidding documents in the historical database based on the preset weights and screening out the historical bidding documents whose similarity exceeds the preset similarity, the method further includes: Setting initial weights for multiple requirements included in the bidding documents, the multiple requirements including the project background and technical standards included in the bidding requirements; The user is instructed to modify the initial weight according to the bidding data, generate a preset weight, and determine the historical bidding data based on the preset weight.

3. The method according to claim 1, characterized in that: After determining the unfilled content based on the first target bid document and filling the unfilled content by using a preset natural language processing algorithm to generate a second target bid document, the method further includes: Obtaining M filling positions of the historical bid content in the second target bid; Displaying a document viewing link at the M filling positions, wherein the document viewing link is used to respond to a user document viewing instruction and display similar documents corresponding to the filling positions to the user, wherein the similar documents include similar tender documents and similar bidding contents; The content in the similar information that overlaps with the similar content is determined, and the overlapping content is highlighted.

4. The method according to claim 3, characterized in that After determining the unfilled content based on the first target bid document and filling the unfilled content by using a preset natural language processing algorithm to generate a second target bid document, the method further includes: Check the content to be reviewed in the second target bid by using a preset checking model, and instruct the user to check the content to be reviewed; Displaying the content to be reviewed to the user; The user is instructed to modify the second target bid based on the document viewing link and the bidding document to generate a third target bid.

5. The method according to claim 4, characterized in that Before checking the content to be reviewed in the second target bid document by using a preset checking model and instructing the user to check the content to be reviewed, the method further includes: Get samples of historical bids; Performing data analysis on each historical bid in the historical bid sample to generate data analysis results, wherein the data analysis includes structure analysis, content analysis, and error identification; Formulate inspection rules based on the data analysis results; A preset inspection model is generated according to the inspection rule, so as to determine the content to be reviewed in the second target bid through the preset inspection model.

6. The method according to claim 1, characterized in that Before filling the historical bid content into the preset bid template to generate the first target bid, the method further includes: Acquire historical bid samples, wherein the historical bid samples include training bid samples and test bid samples; Analyzing the bid structure of each historical bid in the historical bid sample and the content features of each historical bid to generate multiple bid structures and multiple content features; Classifying the historical tender document samples based on the multiple tender document structures and the multiple content features to determine multiple tender document categories; Training the initial machine learning model based on the determined multiple bid document categories to generate a preset machine learning model; A plurality of preset bid templates are generated by the preset machine learning model, and the plurality of preset bid templates include the preset bid template.

7. A terminal device, characterized in that: The terminal device comprises: A calculation module, used to calculate the similarity between all historical bidding materials and the content of bidding materials in the historical database based on preset weights, and screen out historical bidding materials whose similarity exceeds a preset similarity, wherein the bidding materials include bidding requirements, laws and regulations; An acquisition module, used for acquiring similar contents in the historical bidding documents whose similarity with the bidding documents exceeds a similarity threshold; A retrieval module, used to obtain historical bidding document contents corresponding to the similar contents through information retrieval technology; A first generating module, used for filling the historical bidding document content into a preset bidding document template to generate a first target bidding document, wherein the preset bidding document template is a bidding template corresponding to the bidding information; A second generating module is used to determine unfilled content based on the first target bid document, and fill the unfilled content by using a preset natural language processing algorithm to generate a second target bid document; The historical bidding documents include N historical bidding documents, where N is a positive integer greater than 1. Filling the historical bidding document contents into a preset bidding document template to generate a first target bidding document includes: Obtaining partial similar contents of each of the N historical bidding documents, the similarity of which to the bidding document exceeds the similarity threshold, to determine M partial similar contents, the partial similar contents belonging to the similar contents, where M is a positive integer greater than 1; Obtaining N historical bidding document contents corresponding to the M partially similar contents through information retrieval technology; The N historical bid documents are filled into the preset bid document template through a semantic matching algorithm to generate the first target bid document.

8. An electronic device, characterized in that: It comprises a processor, a memory, a user interface and a network interface, wherein the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method steps according to any one of claims 1 to 6 are executed.

Images