Bidding document generation method, electronic equipment, storage medium and program product

Abstract

The invention discloses a bidding document generation method, electronic equipment, a storage medium and a program product, belongs to customization and personalization, and aims to realize customization and personalization of bidding documents. The method comprises the following steps: according to initial content input by a user, generating a language feature tag representing language habits of the user; obtaining a target bidding document corpus matched with the initial content from a corpus database; and generating bidding document content according to the language feature tag and the target bidding document corpus.

Description

Technical Field

[0001] The present application belongs to the field of natural language processing technology, and specifically relates to a method for generating a tender document, an electronic device, a storage medium, and a program product. Background Art

[0002] A tender document is a core document in the bidding process. It is a bid document, meticulously prepared and submitted to the tendering entity based on the detailed terms and requirements of the tender document. It is often referred to as a "bid document." Preparation of a bid document is not simply a collection of documents. Instead, it requires the bidder to thoroughly study the tender documents, conduct on-site research and investigation, and, based on their own strengths and resources, proactively respond to and commit to all requirements of the tender notice. The bid document clearly outlines the specific bid price and other relevant details in order to secure the bid. The bid document serves as the bridge between parties A and B in the bidding process. Therefore, it must be rigorous and logically structured, avoiding inconsistencies or ambiguous statements. Concise and concise bid documents not only facilitate rapid communication but also demonstrate the bidder's commitment to the tender process.

[0003] The traditional approach to generating bid documents involves creating a bid profile for the tender document, then converting the profile into instructional statements. Using these instructions, a pre-built large language model is used to generate the bid catalog. Finally, based on the catalog, the large language model is used to generate the bid document. This traditional approach to generating bid documents limits the quality and accuracy of the generated bid content. Furthermore, the generation process fails to fully consider the individual characteristics and habits commonly used by users when writing bids. This can lead to deviations in the expression and style of the generated bid content from user expectations, making it difficult to meet their actual needs. Summary of the Invention

[0004] The embodiments of the present application provide a method for generating a tender document, an electronic device, a storage medium, and a program product, which can solve the problem that the expression and style of the generated tender document content deviate from the user's expectations and are difficult to meet the user's actual needs.

[0005] In a first aspect, an embodiment of the present application provides a method for generating a bid document, the method comprising: generating a language feature tag representing the user's language habits based on initial content input by the user; obtaining a target bid document corpus matching the initial content from a corpus database; and generating bid document content based on the language feature tag and the target bid document corpus.

[0006] In the second aspect, an embodiment of the present application provides a bid generation device, which includes: a first generation module, used to generate a language feature tag representing the user's language habits based on the initial content input by the user; an acquisition module, used to obtain a target bid corpus matching the initial content from a corpus database; and a second generation module, used to generate bid content based on the language feature tag and the target bid corpus.

[0007] In a third aspect, an embodiment of the present application provides an electronic device comprising a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the method described in the first aspect.

[0008] In a fourth aspect, an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented.

[0009] In a fifth aspect, an embodiment of the present application provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer performs the steps of the method described in the first aspect.

[0010] In an embodiment of the present application, a language feature tag representing the user's language habits is generated based on the initial content input by the user; a target bid document corpus matching the initial content is obtained from a corpus database; and bid content is generated based on the language feature tag and the target bid document corpus, so that the generated bid content conforms to the user's language characteristics and the content is more accurate, thereby achieving customization and personalization of the bid and ensuring the flexibility and accuracy of bid generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 This is a flowchart of a method for generating a tender document provided in an embodiment of the present application; Figure 2 This is a fusion diagram of an integrated model provided in an embodiment of the present application; Figure 3 This is a schematic diagram of a large model prompt word provided in an embodiment of the present application; Figure 4 This is a schematic diagram of the structure of a corpus database provided in an embodiment of the present application; Figure 5 This is a schematic diagram of the composition of a tender document generation system provided in an embodiment of the present application; Figure 6 This is a schematic diagram of the structure of a tender document generation device provided in an embodiment of the present application; Figure 7 This is a structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0012] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0013] The terms "first," "second," and the like in the specification and claims of this application are used to distinguish similar objects, and are not used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, so that the embodiments of this application can be implemented in an order other than that illustrated or described herein, and that the objects distinguished by "first," "second," and the like are generally of the same type, and do not limit the number of objects; for example, the first object can be one or more. In addition, the term "and / or" in the specification and claims refers to at least one of the connected objects, and the character " / " generally indicates that the objects connected are in an "or" relationship.

[0014] The following, in conjunction with the accompanying drawings, describes in detail the bid generation method, electronic device, storage medium and program product provided in the embodiments of the present application through specific embodiments and their application scenarios.

[0015] Figure 1 An embodiment of the present application provides a method for generating a bid document. The method can be performed by an electronic device, such as a computer terminal including an online editing system. In other words, the method can be performed by software or hardware installed on the electronic device, and includes the following steps: Step S102: Generate a language feature tag representing the user's language habits based on the initial content input by the user.

[0016] When writing a bid, the user will enter the initial content, which can be text, characters (strings), etc. The user can set the initial content as the title or body of the bid. Assume that the initial content entered by the user is a Chinese character string. ,but ,in Represents the first Chinese characters, In the embodiment of the present application, a language feature tag representing the user's language habits can be generated based on the initial content input by the user.

[0017] In one implementation, generating a language feature tag representing the user's language habits based on the initial content input by the user includes: encoding the initial content to obtain encoded characters; and analyzing the encoded characters to generate the language feature tag.

[0018] In an embodiment of the present application, after the user enters the initial content, the online editing system of the electronic device can automatically determine the length of the initial content entered by the user, and then use the encoding function to encode the initial content entered by the user to obtain encoded characters. The embodiment of the present application uses a large language model BERT Encoder to perform preliminary encoding of the initial content, as shown below: (1) in, For coded characters, The character threshold is preset. When the cumulative length of the initial input content exceeds the character threshold , the initial content can be encoded.

[0019] In an embodiment of the present application, a large language model BERT Decoder model can be called to analyze the encoded characters in real time to generate a language feature label representing the user's language habits. The language feature label can be a description of the language and can include lexical features such as the user's commonly used words and phrases, for example: the user's preferred conjunction is XX, the user's preferred bid term is XX. It can also include grammatical features such as sentence complexity, for example: the user prefers to use short sentences, the user prefers to use verb-object structures, etc. As the user continues to input, the language feature label can be updated in real time, so that the language feature label can be more consistent with the user's characteristics, thereby maintaining the consistency of the bid content style.

[0020] The embodiment of the present application first uses a large model to generate language feature tags in real time based on the initial content input by the user to describe the user's language features, including lexical features such as the user's commonly used words and phrases, grammatical features such as sentence complexity, etc. Then, when generating the bid content, the language feature tags can be filled in the large model prompt words to maintain the consistency of the bid language style.

[0021] Step S104: Acquire target bid document corpus matching the initial content from a corpus database.

[0022] In the embodiment of the present application, a corpus database including historical bid document corpora is preset. After the user inputs the initial content, the target bid document corpus matching the initial content can be obtained from the corpus database.

[0023] In one implementation, obtaining the target bid document corpus that matches the initial content from the corpus database includes: obtaining a target chapter title that matches a reference chapter title of the initial content from historical chapter titles in the corpus database; determining the historical content tag under the target chapter title as a candidate content tag; generating a reference content tag based on the initial content; and obtaining the target bid document corpus from the corpus database based on the candidate content tag and the reference content tag.

[0024] In an embodiment of the present application, the corpus database may include historical chapter titles and historical content tags under the historical chapter titles, and each historical content tag also has corresponding historical corpus.

[0025] When a user enters initial content, they first enter a reference title for the initial content. In embodiments of the present application, a hybrid similarity matching algorithm (e.g., cosine distance and character overlap calculation) can be designed based on the reference chapter title entered by the user to perform a matching query, thereby obtaining a target chapter title from the historical chapter titles that matches the reference chapter title of the initial content.

[0026] In one implementation, obtaining a target chapter title that matches a reference chapter title of the initial content from the historical chapter titles of the corpus database includes: calculating a first matching score between the reference chapter title and each of the historical chapter titles; and determining the historical chapter title corresponding to the first matching score that is greater than a first score threshold as the target chapter title.

[0027] In an embodiment of the present application, a target chapter title that matches the reference chapter title of the initial content can be obtained from the historical chapter titles in the corpus database using a matching formula. The matching calculation formula is as follows: (2) Among them, is the reference section title entered, is the historical chapter title in the corpus database, The score of the first match of the reference section title to the history section title.

[0028] In this embodiment of the present application, a historical chapter title corresponding to a first matching score greater than a first threshold score may be determined as a target chapter title. Specifically, when the first matching score between a reference chapter title input by the user and certain historical chapter titles in the corpus database exceeds the first threshold score, the content intended to be written by the user may be deemed similar to that historical chapter title (target chapter title).

[0029] In an embodiment of the present application, all historical content tags corresponding to the target chapter title can be extracted, and the historical content tags under the target chapter title can be determined as candidate content tags. This process realizes sentence-level text matching.

[0030] In the embodiment of the present application, reference content tags can be generated based on the initial content input by the user. Specifically, this process directly determines whether the large model can generate bid content that meets the theme. In order to improve the accuracy of generating reference content tags, the system is designed as follows: Figure 2 The integrated model shown in the figure is used to generate reference content labels. The integrated model combines Jiutian13.9B, Qwen2.5-32B, Microsoft / DeBERTa-v3-large and MoritzLaurer / DeBERTa-v3-large. As shown in Formula 1, whenever the initial content input by the user reaches the threshold When generating reference content labels for the initial content, the integrated model can be used. Through a carefully designed fusion mechanism, the outputs of different large language models are integrated and weight parameters are designed to ensure the generation of diverse, personalized, and high-quality tender documents. After obtaining the reference content labels for the initial content, the target tender document corpus can be retrieved from the corpus database based on the candidate content labels and the reference content labels.

[0031] In one implementation, obtaining the target bid document corpus from the corpus database based on the candidate content tags and the reference content tags includes: obtaining a target content tag that matches the reference content tag from the candidate content tags; and determining the historical corpus corresponding to the target content tag under the target chapter title as the target bid document corpus.

[0032] In an embodiment of the present application, a target content tag matching the reference content tag may be obtained from the candidate content tags, and then the historical corpus corresponding to the target content tag under the target chapter title may be determined as the target bid corpus.

[0033] Step S106: Generate the tender content according to the language feature tags and the target tender corpus.

[0034] like Figure 3 As shown, the large model prompt words can be written based on the real-time generated language feature tags and target bid corpus, and Figure 3 The integrated model in the

[15] generates the bid content and displays it on the online editing front-end page. Since the prompt words include language feature tags, the generated bid content is consistent with the language style of the user-written content. Manual addition, deletion, and modification operations can be performed and finally inserted into the bid.

[0035] The bid generation method provided in the embodiment of the present application generates a language feature tag representing the user's language habits based on the initial content input by the user; obtains the target bid corpus matching the initial content from the corpus database; generates the bid content based on the language feature tag and the target bid corpus. It can analyze the user's language features and the content he wants to express in real time, and by generating the language feature tag and combining it with the target bid corpus in the corpus database, generates the bid content that conforms to the user's language features and has more accurate content, thereby realizing the customization and personalized generation of the bid and ensuring the flexibility and accuracy of the bid generation.

[0036] The embodiment of the present application integrates the user's language characteristics and unique writing style into the bid generation process, and at the same time combines it with a rich corpus database to achieve personalized and precise generation of bid content. This personalized generation method ensures that the bid not only meets industry standards but also accurately reflects the user's unique perspective and expression, thereby greatly improving the user experience and the quality of the bid. In addition, the system also significantly reduces the user's workload and improves the efficiency of bid generation through automated and intelligent operations. In addition, the system is highly flexible and scalable, and can adapt to the special needs of different industries and users.

[0037] In one implementation, obtaining a target content tag that matches the reference content tag from the candidate content tags includes: calculating a second matching score between each candidate content tag and the reference content tag; and determining the candidate content tag corresponding to the second matching score that is greater than a second score threshold as the target content tag.

[0038] In an embodiment of the present application, when obtaining a target content tag that matches a reference content tag from candidate content tags, a second matching score between each candidate content tag and the reference content tag can be calculated based on a hybrid matching strategy, and a candidate content tag corresponding to a second matching score greater than a second score threshold is determined as the target content tag. That is, based on the hybrid matching strategy, a second match is searched in the candidate content in the corpus database to obtain a target content tag greater than the second score threshold. The hybrid matching strategy is shown in Formula 3, which implements passage-level text matching: (3) in, Indicates the reference content label of the input, represents candidate content labels, Represents the second matching score.

[0039] In the embodiments of the present application, different text matching algorithms are designed based on sentence-level text and paragraph-level text, which can ensure the matching accuracy of the initial input by the user and the target content label in the corpus database, which makes information retrieval more efficient and accurate, and constructs more accurate prompt words for large model output.

[0040] In one implementation, before generating a language feature tag representing the user's language habits based on the initial content input by the user, the method further includes: splitting the collected historical bids into corpora to obtain historical corpora and historical chapter titles of the historical corpora; annotating the historical corpora to obtain historical content tags of the historical corpora; and constructing the corpus database based on the historical chapter titles, the historical corpora, and the historical content tags.

[0041] In an embodiment of the present application, a corpus database can be constructed in advance, and historical bids can be collected first, and the historical bids can be pre-processed to standardize the format. According to elements such as chapters and paragraphs, the historical bids can be split into corpora using regular expressions to obtain historical corpora and historical chapter titles of the historical corpora. In order to improve the availability of the corpus database, the BERT model can be used to annotate the differential historical corpus to generate historical content labels (such as technical requirements, business terms, etc.). Annotation can be done manually or in combination with automatic annotation to improve the accuracy and efficiency of annotation. In an embodiment of the present application, a relational database can be constructed of historical chapter titles, historical corpora, and historical content labels to achieve mixed storage of historical corpora and historical content labels. The storage relationship is as follows: Figure 4 shown.

[0042] Figure 5 The diagram shows the composition of a tender document generation system provided by an embodiment of the present application. The system includes an input module, an online editing module, a large language model, a corpus database, a content recommendation module, and a manual selection module. The system is combined with an online editing function. During the process of a user writing a tender document, the system can extract the initial content input by the tender document writer in real time, and dynamically generate its language feature tags and reference content tags through the large language model. The system searches the corpus database according to the reference content tags to match the relevant target tender document corpus, and fills the matched target tender document corpus and language feature tags into the prompt words, thereby automatically generating tender document content that meets the user's language characteristics by combining the current user's language characteristics and the relevant target tender document corpus, and can return the most similar tender document content. The user can choose from a variety of tender document contents, thereby realizing customized generation of tender documents and maintaining the consistency of language features before and after the tender documents.

[0043] It should be noted that the bid generation method provided in the embodiments of the present application can be executed by a bid generation device, or a control module in the bid generation device that is used to execute the bid generation method. In the embodiments of the present application, the bid generation device provided in the embodiments of the present application is described by taking the bid generation method executed by the bid generation device as an example.

[0044] Figure 6 FIG is a schematic diagram of the structure of a device for generating a tender document according to an embodiment of the present application. Figure 6 As shown, the tender document generating device 600 includes: a first generating module 610, an acquiring module 620 and a second generating module.

[0045] The first generation module 610 is used to generate a language feature tag representing the user's language habits based on the initial content input by the user; the acquisition module 620 is used to obtain the target bid document corpus matching the initial content from the corpus database; the second generation module 630 is used to generate the bid document content based on the language feature tag and the target bid document corpus.

[0046] In one implementation, the first generating module 610 is configured to encode the initial content to obtain encoded characters; and analyze the encoded characters to generate the language feature tag.

[0047] In one implementation, the acquisition module 620 is used to obtain a target chapter title that matches the reference chapter title of the initial content from the historical chapter titles of the corpus database; determine the historical content tag under the target chapter title as a candidate content tag; generate a reference content tag based on the initial content; and obtain the target bid corpus from the corpus database based on the candidate content tag and the reference content tag.

[0048] In one implementation, the acquisition module 620 is configured to calculate a first matching score between the reference chapter title and each of the historical chapter titles; and determine the historical chapter title corresponding to the first matching score greater than a first score threshold as the target chapter title.

[0049] In one implementation, the acquisition module 620 is configured to acquire a target content tag matching the reference content tag from the candidate content tags; and determine the historical corpus corresponding to the target content tag under the target chapter title as the target bid document corpus.

[0050] In one implementation, the acquisition module 620 is configured to calculate a second matching score between each candidate content tag and the reference content tag; and determine the candidate content tag corresponding to the second matching score greater than a second score threshold as the target content tag.

[0051] In one implementation, the first generation module 610 is further configured to perform corpus splitting on the collected historical bids to obtain historical corpus and historical chapter titles of the historical corpus; annotate the historical corpus to obtain historical content tags of the historical corpus; and construct the corpus database based on the historical chapter titles, the historical corpus, and the historical content tags.

[0052] The bid generation device in the embodiments of the present application can be a device, or a component, integrated circuit, or chip in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. For example, the mobile electronic device can be a mobile phone, tablet computer, laptop computer, PDA, in-vehicle electronic device, wearable device, ultra-mobile personal computer (UMPC), netbook, or personal digital assistant (PDA), while the non-mobile electronic device can be a server, network attached storage (NAS), personal computer (PC), television, ATM, or self-service machine, etc., without specific limitation in the embodiments of the present application.

[0053] The bid generation device in the embodiment of the present application may be a device having an operating system. The operating system may be an Android operating system, an iOS operating system, or other possible operating systems, which are not specifically limited in the embodiment of the present application.

[0054] The bid generation device provided in the embodiment of the present application can achieve Figures 1 to 5 To avoid repetition, the various processes implemented in the method embodiment are not described here.

[0055] like Figure 7 As shown, an embodiment of the present application further provides an electronic device 700, including a processor 701 and a memory 702, wherein the memory 702 stores a program or instruction that can be run on the processor 701, and when the program or instruction is executed by the processor 701, it implements: generating a language feature tag representing the user's language habits based on the initial content input by the user; obtaining a target bid document corpus matching the initial content from a corpus database; and generating bid document content based on the language feature tag and the target bid document corpus.

[0056] In one implementation, the original content is encoded to obtain encoded characters; and the encoded characters are analyzed to generate the language feature tag.

[0057] In one implementation, a target chapter title that matches a reference chapter title of the initial content is obtained from historical chapter titles in the corpus database; historical content tags under the target chapter title are determined as candidate content tags; reference content tags are generated based on the initial content; and the target bid document corpus is obtained from the corpus database based on the candidate content tags and the reference content tags.

[0058] In one implementation, a first matching score between the reference chapter title and each of the historical chapter titles is calculated; and the historical chapter title corresponding to the first matching score greater than a first score threshold is determined as the target chapter title.

[0059] In one implementation, a target content tag matching the reference content tag is obtained from the candidate content tags; and the historical corpus corresponding to the target content tag under the target chapter title is determined as the target bid document corpus.

[0060] In one implementation, a second matching score between each candidate content tag and the reference content tag is calculated; and the candidate content tag corresponding to the second matching score greater than a second score threshold is determined as the target content tag.

[0061] In one implementation, before generating language feature tags representing the user's language habits based on the initial content input by the user, the collected historical bid documents are split into corpora to obtain historical corpora and historical chapter titles of the historical corpora; the historical corpora are annotated to obtain historical content tags of the historical corpora; and the corpus database is constructed based on the historical chapter titles, the historical corpora, and the historical content tags.

[0062] The specific execution steps can refer to the various steps of the above-mentioned bid generation method embodiment, and can achieve the same technical effect. To avoid repetition, they will not be described here.

[0063] It should be noted that the electronic devices in the embodiments of the present application include: servers, terminals, or other devices other than terminals.

[0064] The above electronic device structure does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. For example, the input unit may include a graphics processing unit (GPU) and a microphone, and the display unit may be configured as a display panel in the form of a liquid crystal display, an organic light-emitting diode, etc. The user input unit includes at least one of a touch panel and other input devices. A touch panel is also called a touch screen. Other input devices may include, but are not limited to, a physical keyboard, function keys (such as volume control buttons, power buttons, etc.), a trackball, a mouse, and a joystick, which will not be detailed here.

[0065] The memory can be used to store software programs and various data. The memory may mainly include a first storage area for storing programs or instructions and a second storage area for storing data. The first storage area may store an operating system, applications or instructions required for at least one function (such as a sound playback function, an image playback function, etc.). In addition, the memory may include volatile memory or non-volatile memory, or the memory may include both volatile and non-volatile memory. The non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM) and direct rambus random access memory (DRRAM).

[0066] The processor may include one or more processing units; optionally, the processor may integrate an application processor and a modem processor, wherein the application processor primarily handles operations related to the operating system, user interface, and application programs, and the modem processor primarily processes wireless communication signals, such as a baseband processor. It is understood that the modem processor may not be integrated into the processor.

[0067] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the various processes of the above-mentioned bid generation method embodiment are implemented and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0068] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer-readable storage medium, such as ROM, RAM, magnetic disk or optical disk.

[0069] An embodiment of the present application also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer executes the various processes of the above-mentioned embodiment of the bid generation and can achieve the same technical effect. To avoid repetition, they will not be repeated here.

[0070] It should be noted that, in this article, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element. In addition, it should be noted that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in the opposite order according to the functions involved. For example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0071] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better embodiment. Based on this understanding, the technical solution of this application, or the part that contributes to the existing technology, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of this application.

[0072] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of this application, ordinary technicians in this field can also make many forms without departing from the purpose of this application and the scope of protection of the claims, all of which are within the protection of this application.

Claims

1. A method for generating a tender document, characterized in that: include: Generate a language feature tag representing the user's language habits based on the initial content input by the user; Acquire target bid document corpus matching the initial content from a corpus database; The tender document content is generated according to the language feature tags and the target tender document corpus.

2. The method according to claim 1, characterized in that Generating a language feature tag representing the user's language habits based on the initial content input by the user includes: Encoding the initial content to obtain encoded characters; The coded characters are analyzed to generate the language feature labels.

3. The method according to claim 1, characterized in that The step of obtaining target bid document corpus matching the initial content from a corpus database includes: Acquire a target chapter title that matches the reference chapter title of the initial content from the historical chapter titles of the corpus database; determining the historical content tags under the target chapter title as candidate content tags; generating a reference content tag based on the initial content; The target bid document corpus is acquired from the corpus database according to the candidate content tags and the reference content tags.

4. The method according to claim 3, characterized in that The step of acquiring a target chapter title that matches a reference chapter title of the initial content from the historical chapter titles of the corpus database includes: Calculating a first matching score between the reference chapter title and each of the historical chapter titles; The historical chapter title corresponding to the first matching score greater than a first score threshold is determined as the target chapter title.

5. The method according to claim 3, characterized in that The step of acquiring the target bid document corpus from the corpus database according to the candidate content tags and the reference content tags includes: Acquire a target content tag that matches the reference content tag from the candidate content tags; The historical corpus corresponding to the target content tag under the target chapter title is determined as the target bid document corpus.

6. The method according to claim 5, characterized in that The acquiring a target content tag matching the reference content tag from the candidate content tags includes: Calculating a second matching score between each of the candidate content tags and the reference content tags; The candidate content tag corresponding to the second matching score greater than a second score threshold is determined as the target content tag.

7. The method according to claim 1, characterized in that Before generating a language feature tag representing the user's language habits based on the initial content input by the user, the method further includes: Splitting the collected historical bid documents into corpora to obtain historical corpora and historical chapter titles of the historical corpora; Annotating the historical corpus to obtain historical content labels for the historical corpus; The corpus database is constructed according to the historical chapter titles, the historical corpus and the historical content tags.

8. An electronic device, characterized in that: The method comprises a processor, a memory, and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the method for generating a tender document according to any one of claims 1 to 7.

9. A readable storage medium, characterized in that: The readable storage medium stores a program or instruction, and when the program or instruction is executed by the processor, the steps of the method for generating a bid document according to any one of claims 1 to 7 are implemented.

10. A computer program product, characterized in that The computer program product includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer is caused to perform the steps of the method for generating a tender document according to any one of claims 1 to 7.

Images