Bidding scheme compilation real-time prompting system and method
By utilizing a real-time prompting system for bid proposal preparation, and leveraging technologies such as a knowledge management platform and large language models, the system addresses the issue of low efficiency in bid document preparation, achieving efficient content prompts and quality assurance, thereby improving the efficiency and quality of bid proposal preparation.
Patent Information
- Application Number
- CN202511371040.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-24
- Publication Date
- 2026-01-20
AI Technical Summary
In the process of preparing tender documents, writers face problems such as low efficiency and difficulty in ensuring the quality of proposals, especially in finding relevant materials and lacking creative ideas.
A real-time prompting system for bid proposal development is adopted, including a knowledge management platform module, an editor module, and a retrieval device. It converts paper documents using an OCR recognition tool, provides content prompts using a large language model and vector database, and retrieves relevant information from the Internet using a web crawler module. It also monitors user input in real time and provides suggestions for similar content.
It improved the efficiency and quality of bid document preparation, reduced the time spent searching for information, enhanced the writer's creative expression ability, and ensured the completeness and innovation of the bid proposal.
Smart Images
Figure CN121365652A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of human-computer interaction, and particularly relates to a bidding scheme compiling real-time prompting system and method. BACKGROUND
[0002] In the current competitive bidding market, the bidding party or the agent institution pays high attention to the innovativeness, rationality and integrity of the bidding scheme in the bidding evaluation stage. As the embodiment of the core competitiveness of the enterprise in the bidding process, the quality of the bidding document directly determines whether the bidding can be finally successful.
[0003] However, the writer encounters some difficulties in the process of compiling the bidding document. First, the writer of the bidding document needs to manually search the bidding scheme related content from various materials scattered everywhere, which consumes a lot of time. For example, the writer often needs to find the materials meeting the requirements of the bidding document from a large number of sales contracts and enterprise qualification certificates as the input of the bidding document in the compiling process. Second, the writer of the bidding document often faces the difficulty of lack of creativity in the process of compiling the bidding scheme technical document. For example, the writer lacks new ideas or expression methods and cannot find new writing ideas, thereby affecting the quality of the bidding scheme technical document. As a result, the writer has low efficiency in compiling the bidding document and the quality is difficult to guarantee.
[0004] Therefore, it is urgent to provide a bidding scheme compiling real-time prompting system and method to solve the problem of low efficiency of the writer in compiling the bidding document. SUMMARY
[0005] The present application provides a bidding scheme compiling real-time prompting system and method to solve the problem of low efficiency of the writer in compiling the bidding document.
[0006] To achieve the above object, the present application adopts the following technical scheme:
[0007] A bidding scheme compiling real-time prompting system comprises a knowledge management platform module, an editor module and a retriever. The knowledge management platform module is used for storing the documents uploaded by the user and comprises a public library and a private library. The editor module comprises a prompt area and an editing area for the user to input the writing content. The editor module is used for transmitting the writing content back to the private library of the user in the knowledge management platform module through the HTTP protocol, simultaneously monitoring the user input and the mouse event, and judging whether the real-time prompting is needed according to the user input. If the real-time prompting is needed, the editor module sends the text information to the retriever. The retriever retrieves the content prompting information through networking or obtains the content prompting information from the public library. The editor module displays the content prompting information in the prompt area after obtaining the content prompting information from the retriever. If the real-time prompting is not needed, the user writes by himself / herself.
[0008] To optimize the above technical solutions, the specific measures taken also include:
[0009] Further, it also includes a data preparation module, which converts paper or picture format files into text using OCR recognition tools, then obtains the required text information through text analysis, and stores it in the public library of the knowledge management platform module together with the original electronic version of the file.
[0010] Further, it also includes a slice embedding module and a vector database. The slice embedding module is used to divide the text information sent by the public library or the retriever into text blocks according to the slicing strategy. After the content of each text block is embedded and calculated by the large language model built-in the module, a vector representation is generated. The vector database is used to store the vector data generated by the public library through the slice embedding module, or to obtain the n text blocks closest to the vector data in terms of semantics, as well as the document and link to which each text block belongs, according to the vector data generated from the text information, and return the relevant information to the editor module for display in the prompt area of the editor module.
[0011] Further, it also includes a crawler module. The retriever sends text information to the crawler module, which retrieves relevant document information from the Internet according to the text information and returns the HTTP link of the document information to the retriever.
[0012] Further, the prompt area includes a text area for displaying text, a picture area for displaying pictures, and a video area for displaying videos.
[0013] Further, a method for applying the above bid scheme preparation real-time prompting system includes the following steps:
[0014] When the user sets to turn on the receiving real-time writing prompt function of the editor module, the editor module monitors the user's input and mouse events in the editing area, and determines whether real-time prompting is needed according to the user's input. If real-time prompting is needed, the editor module sends text information to the retriever, which retrieves content prompt information from the public library or the Internet according to the text information. After the editor module obtains the content prompt from the retriever, it displays it in the prompt area. If real-time prompting is not needed, the user writes it himself.
[0015] Further, the determination of whether real-time prompting is needed according to the user's input includes the following steps: the editor module determines whether the user's input triggers real-time prompting by the characters preset by the user. When the editor module's editing area inputs the preset characters, it confirms that real-time prompting is needed, otherwise it is not.
[0016] Further, the characters include a " > " character, a " similar case requirement: " character, a ". " character and a ", " character.
[0017] If the editor module monitors that the user inputs the " > " character, the editor module searches forward for the " < " character at the current input position of the user, and then sends the file name between the " < " and " > " characters to the retriever, which retrieves the corresponding file from the knowledge management platform module;
[0018] If the editor module monitors that the user inputs the " similar case requirement: " character, the editor module searches backward for the ". " character at the current input position of the user, and then sends the requirement content between the " similar case requirement: " and ". " characters to the retriever, which retrieves the corresponding related contract case from the knowledge management platform module and returns the editor module to prompt the " contract project name, project type, project construction content, contract signing time, project construction unit and project signing amount " field information and the contract case link in the corresponding public library;
[0019] If the editor module monitors that the user inputs the ", " or ". " character, the editor module simultaneously retrieves two types of texts, first, searches forward for the ". " at the current input position of the user, and then obtains the sentence content being input or completed by the user between the searched character position and the current position; second, searches forward for the carriage return character at the current input position of the user, and then obtains the paragraph content being input or completed by the user; the editor module simultaneously sends several paragraph texts to the retriever, the retriever retrieves and returns the obtained content as prompt information to the editor module, which is displayed in the prompt area of the editor module.
[0020] Further, the step of judging whether real-time prompting is needed according to the input of the user includes the following steps: the editor module judges whether the user is in an editing state by analyzing whether the user is inputting characters, and if the user is in the editing state, it is confirmed that real-time prompting is needed, otherwise, it is not needed.
[0021] Further, the step of judging whether real-time prompting is needed according to the input of the user includes the following steps: the editor module judges by a preset typing interval time, and if the typing interval of the user exceeds the preset typing interval time, it is confirmed that real-time prompting is needed, otherwise, it is not needed.
[0022] The beneficial effects of the present application are:
[0023] The application can find the similar paragraphs and sentences in the knowledge management platform module or the public internet according to the information of the paragraph and sentence that the writer is inputting, and prompt the writer for timely reference, so as to improve the writing efficiency and quality of the writer; meanwhile, the materials meeting the requirements of the bidding document can be found from the large number of sales contracts and enterprise qualification certificates in the knowledge management platform module, so as to reduce the time of the writer for searching the materials. BRIEF DESCRIPTION OF DRAWINGS
[0024] Figure 1 The figure is a schematic diagram of the architecture of the bidding scheme preparation real-time prompting system proposed by the application;
[0025] Figure 2 The figure is a schematic diagram of the interface of the editor module of the bidding scheme preparation real-time prompting system proposed by the application a;
[0026] Figure 3 The figure is a schematic diagram of the interface of the editor module of the bidding scheme preparation real-time prompting system proposed by the application b;
[0027] Figure 4 The figure is a schematic diagram of the flow of the data preparation module of the bidding scheme preparation real-time prompting system proposed by the application;
[0028] Figure 5 The figure is a schematic diagram of the flow of the retriever of the method proposed by the application a;
[0029] Figure 6 The figure is a schematic diagram of the flow of the retriever of the method proposed by the application b;
[0030] Figure 7 The figure is a schematic diagram of the flow of the retriever of the method proposed by the application c. DETAILED DESCRIPTION
[0031] The technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all the other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the application.
[0032] As shown in the accompanying Figure 1 and the accompanying Figure 2As shown, the bidding scheme preparation real-time prompting system of the embodiment of the application comprises a knowledge management platform module, an editor module and a retriever. The knowledge management platform module is used to store documents uploaded by users, including a public library and a private library. The editor module comprises a prompt area and an editing area for user input of writing content. The editor module is used to save the writing content in the private library of the user in the knowledge management platform module through HTTP protocol, simultaneously monitor user input and mouse events, and determine whether real-time prompting is needed according to the user input. If real-time prompting is needed, the editor module sends text information to the retriever. The retriever retrieves content prompting information from the public library or through network according to the text information. The editor module displays the content prompting information obtained from the retriever in the prompt area. If real-time prompting is not needed, the user writes by himself.
[0033] The content of the public library is open to all users of the knowledge management platform. All users can access the content of the stored documents, which include enterprise contracts, enterprise qualifications, technical data and the like.
[0034] The private library uses object storage technology to store and manage private library data of each user. The data of each user is stored in different buckets, and a role-based access control (RBAC) strategy is implemented to ensure that only users and system administrators with corresponding permissions can access the private library. All access requests are authenticated and checked for permissions to ensure that data cannot be accessed by other users without permission. The data in the private library can be encrypted to ensure that even in the case of data leakage, unauthorized third parties cannot read or understand it. At the network level, HTTPS and other protocols are used between the editor and the private library to ensure the security of data transmission.
[0035] The editor module is implemented based on HTML, CSS and JavaScript technology. The user's browser can download it from the system through HTTP and run it on the client-side device, such as a PC, a tablet computer and a smart phone, etc. The user can set whether to receive real-time writing prompts in the editor module.
[0036] As shown in the specific embodiment based on the above, Figure 4 The data preparation module converts paper or picture format contract files and qualification certificates of the enterprise into text using an OCR recognition tool, and then obtains the required text information such as the project name, project type, project construction content, contract signing time, project construction unit, project signing amount, certificate name, etc. through text analysis, and stores the original contract files and electronic versions of the qualification certificates in the public library of the knowledge management platform module at the same time.
[0037] In another specific embodiment based on the above, a slice embedding module and a vector database are further included, the slice embedding module is used to divide the text information sent by the public library document or retriever into text blocks according to a slicing strategy, and each text block generates a vector representation after the content of the text block is calculated by the embedding of the large language model built-in the module; the vector database is used to store the vector data generated by the public library document through the slice embedding module, or to obtain n text blocks that are closest in semantics to the vector data according to the vector data generated from the text information, and the document and link to which each text block belongs, and return the related information to the editor module for display in the prompt area of the editor module.
[0038] In the present scheme, the slicing strategy described above can be fixed-length slicing, chapter slicing, etc. The link of each slice to the document to which it belongs in the knowledge management platform module is stored together. The embedding calculation described above is a common term in the field of deep learning, which means representing an object with vectors. These vectors are low-dimensional and dense, which is more convenient for deep learning algorithms to process. The key feature of the vector database described above is to quickly find other vectors that are most similar to a given vector, which is achieved by calculating the similarity between vectors using the Euclidean similarity or cosine similarity algorithm.
[0039] Specifically, if the editor transmits a qualification certificate name, the retriever sends the qualification certificate name sent by the editor to the public knowledge base, and retrieves the corresponding file content from the public knowledge base. If the editor transmits a similar case requirement, the retriever retrieves the corresponding relevant contract case from the public library of the knowledge management platform, and returns the data to the editor module for display in the prompt area of the editor. The content includes: “contract project name, project type, project construction content, contract signing time, project construction unit, project signing amount” and other field information and the link to the contract electronic version in the public library of the knowledge management platform. The user can click to view or directly copy the contract case. If the editor transmits the name of a policy document issued by a relevant department of a country, the retriever obtains the content of the corresponding file through a crawler. If the editor transmits a segment of text of other content, the retriever sends the segment of text to the slice embedding module, obtains the vector representation corresponding to the segment of text through the large model built-in the module, and then sends the vector representation to the vector database to obtain n text blocks closest in semantics to the vector, and the document and link to which each text block belongs, and returns these information to the editor module for display in the prompt area of the editor.
[0040] In another specific embodiment based on the above, a crawler module is further included, and the retriever sends text information to the crawler module, and the crawler module retrieves relevant document information from the Internet according to the text information and returns an HTTP link of the document information to the retriever.
[0041] As shown in the accompanyingFigure 3 In another specific embodiment based on the above, the prompt area includes a text area for displaying text, a picture area for displaying pictures, and a video area for displaying videos.
[0042] Specifically, in use, if the user inputs writing content in the editor module's editing area after starting to receive real-time writing prompts, the editor module transmits the content back to the knowledge management platform module for saving in the user's private library via the HTTP protocol. The editor module also monitors user input and mouse events to determine whether real-time prompts are needed based on the user's input; if real-time prompts are needed, the editor module sends the necessary text information to the retriever, which searches for or calculates real-time content prompt information based on the text information. After the editor module obtains the content prompts from the retriever, it displays them one by one in the prompt area, including text, picture, video, and other multimedia forms of prompts for the user to browse and copy. The editor module interface can use clear icons and colors to distinguish between different prompt types while maintaining the simplicity of the interface.
[0043] A method for a bidding scheme preparation real-time prompt system, comprising the following steps:
[0044] When the user sets to start the editor module's real-time writing prompt receiving function, the editor module monitors user input and mouse events in the editing area and determines whether real-time prompts are needed based on the user's input. If real-time prompts are needed, the editor module sends text information to the retriever, which searches for or obtains content prompt information from the public library based on the text information. After the editor module obtains the content prompts from the retriever, it displays them in the prompt area. If real-time prompts are not needed, the user writes them himself.
[0045] The determination of whether real-time prompts are needed based on the user's input includes the following steps: the editor module determines whether real-time prompts are needed based on the user's input by using the characters preset by the user to trigger real-time prompts. When the editor module's editing area inputs the preset characters, it confirms that real-time prompts are needed, otherwise, it does not.
[0046] Specifically, the characters include the " 》 " character, the " similar case requirements: " character, the ". " character, and the ", " character.
[0047] If the editor module monitors the user's input of the " 》 " character, it searches forward from the user's current input position for the " 《 " character, and then sends the file name between the " 《 " and " 》 " characters to the retriever, which searches for the corresponding file from the knowledge management platform module.
[0048] If the editor module monitors that the user inputs the character of "similar case requirement:", the editor module searches the character of "." backward from the current input position of the user, and then sends the requirement content between the characters of "similar case requirement:" and "." to the retriever, which retrieves the corresponding relevant contract case from the knowledge management platform module, and returns the editor module to prompt the field information of "contract project name, project type, project construction content, contract signing time, project construction unit, and project signing amount" and the contract piece link in the corresponding public library.
[0049] If the editor module monitors that the user inputs the character of "," or ".", the editor module simultaneously retrieves two types of text. First, the editor module searches the character of "." forward from the current input position of the user, and obtains the sentence content that the user is inputting or has completed between the character position searched and the current position. Second, the editor module searches the character of carriage return line feed forward from the current input position of the user, and obtains the paragraph content that the user is inputting or has completed between the character position searched and the current position. The editor module sends the several pieces of text to the retriever, which sends the several pieces of text to the slice embedding module to obtain the corresponding vector representations, respectively. Then, the retriever retrieves n pieces of text slices that are respectively most similar in semantics to the several vectors from the vector database as the alternative prompt information. Then, the retriever filters out the most relevant content according to the semantic similarity between the alternative prompt information and the user input, and returns the obtained content to the editor module as the prompt information for display in the prompt area of the editor module. Specifically, if it is two pieces of text, the editor module sends the two pieces of text to the slice embedding module to obtain the corresponding vector representations of the two pieces of text, respectively. Then, the retriever retrieves 2n pieces of text slices that are respectively most similar in semantics to the two vectors from the vector database as the alternative prompt information. Then, the retriever filters out the most relevant content according to the semantic similarity between the alternative prompt information and the user input.
[0050] In the above, whether the user needs to be prompted in real time according to the user input also includes the following steps: the editor module intelligently determines whether the user needs to be prompted in real time according to the user input by combining the context environment input by the user. The editor module determines whether the user is in an editing state by analyzing whether the user is inputting text, or determines whether the user is in a document browsing state by collecting and analyzing the position change of the mouse or the events triggered by the scroll wheel through JavaScript. If the user is in the editing state, it is determined that real-time prompting is needed, otherwise, real-time prompting is not needed.
[0051] The above-mentioned determination of whether to provide real-time prompts to the user based on the user's input also includes the following steps: The editor module makes a judgment based on the preset typing interval time. When the user's typing interval exceeds the preset typing interval time, such as the default 3 seconds, or when content is deleted multiple times in a row, it confirms that real-time prompts are required; otherwise, no prompts are provided.
[0052] One specific embodiment of the present invention is as follows:
[0053] The administrator or maintenance personnel scan the paper contract documents and qualification certificates into image format, and then send the image format contract documents and qualification certificates to the data preparation module.
[0054] The data preparation module uses tools such as OCR recognition to convert image-formatted contract documents and qualification certificate content into text. Subsequently, the data preparation module parses the text to obtain information such as the project name, project type, project construction content, contract signing time, project construction unit, project contract amount, and certificate name. This information, along with the original electronic versions of the contract and qualification certificate, is stored in the public library of the knowledge management platform module.
[0055] As attached Figure 5 The following is the qualification certificate retrieval process:
[0056] When in use, if the editor module detects the user inputting the ">" character, it searches backwards for the "<" character from the current input position and then sends the qualification document name between the "<" and ">" characters to the search engine. The search engine sends the file name sent by the editor module to the public database and retrieves the corresponding file content from the database. If found, it returns the link to the electronic version of the certificate to the search engine. Upon receiving the result, the search engine returns it to the editor module, which displays the certificate name in the prompt area for the user to click and copy.
[0057] As attached Figure 6 The following is a sample search process for contract cases:
[0058] During use, the editor module detects the user's input of the characters "Similar Case Requirements:", and then searches for the character "." from the current input position. The information between "Similar Case Requirements:" and "." represents the case requirements. The editor module sends the requirements between "Similar Case Requirements:" and "." to the search engine. The search engine retrieves relevant contract cases from a public database based on the case requirements. The search engine returns the information to the editor. The returned information includes fields such as "Project Type, Contract Document Name, Project Construction Content, Contract Signing Time, Project Contract Amount, and Project Construction Unit," along with the corresponding contract document links. The editor displays the pushed content in the editor's tooltip area.
[0059] As shown in the accompanying drawings, the retrieval process of the document slice is as follows: Figure 7
[0060] If the editor module monitors that the user inputs a "," or "." character, the editor module needs to retrieve two types of text at the same time. First, the editor module searches forward from the current input position of the user to the "." or carriage return character, and the characters between the searched character position and the current position, to obtain the sentence content that the user is inputting or has completed. Second, the editor module searches forward from the current input position of the user to the carriage return character, and the characters between the searched character position and the current position, to obtain the paragraph content that the user is inputting or has completed. The editor module sends the two pieces of text to the retriever at the same time. The retriever sends the two pieces of text to the slice embedding module, and the slice embedding module performs embedding calculation on the two pieces of text respectively to obtain the vector representations corresponding to the two pieces of text respectively. The retriever sends the two vectors to the vector database. The vector database searches for the first n pieces of text slices that are most similar in semantics to the two vectors respectively, that is, a total of 2n pieces of text slices, as the candidate prompt information, and returns to the retriever. The size of n is configurable.
[0061] The retriever screens out the most relevant content according to the similarity in semantics between the candidate prompt information and the text input by the user. The retriever can set a threshold w of the similarity, and select k pieces of slices with a vector similarity less than w from the 2n slices returned by the vector database as the prompt information and return to the editor module. The sizes of w and k are configurable. The similarity between vectors can be calculated using methods such as cosine similarity and Euclidean similarity.
[0062] For example, the cosine similarity between two vectors A and B is calculated as follows: The Euclidean similarity between two vectors A and B can also be calculated, where d(A, B) is the Euclidean similarity between vectors A and B, a i and b i are the values of the elements in the corresponding dimensions of A and B respectively:
[0063] The editor module displays the content of the k pieces of slices returned by the retriever in the prompt area for the user to refer to when writing, and saves the prompts viewed and used by the user in the private library for the user to review and use again when needed.
[0064] The present application can find the paragraphs and sentences that are similar in semantics according to the information such as the paragraph and sentence being input by the writer, from the knowledge management platform module or the public Internet, and prompt the writer for timely reference, thereby improving the efficiency and quality of the writer in writing the tender document; meanwhile, the materials meeting the requirements of the tender document are found from a large number of sales contracts and enterprise qualification certificates in the knowledge management platform module, thereby reducing the time for the writer to find the materials.
[0065] The editor module of the present application can monitor the characters input by the user in real time, and trigger the editor module to retrieve the file name input by the user, the current input sentence and paragraph information by monitoring characters such as “,” “.” and “>”, and then send them to the retriever to trigger the content prompt content acquisition process.
[0066] The retriever of the present application can obtain a plurality of semantically most relevant text slices from the vector database, and set a threshold value, only return the slices corresponding to the vectors with a similarity less than the threshold value to the editor as prompt content to be displayed by the editor. In this way, irrelevant prompt content is avoided, thereby avoiding interference with the user's creation process.
[0067] The present application can quickly obtain relevant information of qualifications and certificates for reference.
[0068] Those skilled in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed in the present application can be realized by electronic hardware or a combination of electronic hardware and computer software. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.
[0069] The above is only the preferred embodiment of the present application, and the protection scope of the present application is not limited to the above-mentioned embodiments. Any technical solution falling within the concept of the present application shall fall within the protection scope of the present application. It should be noted that for ordinary skilled in the art, some improvements and refinements without departing from the principles of the present application shall be considered within the protection scope of the present application.
Claims
1. A real-time bidding proposal development prompting system, characterized by: The knowledge management platform module is used for storing documents uploaded by users, including a public library and a private library, the editor module includes a prompt area and an editing area for user input of writing content, the editor module is used for returning the writing content to the private library of the user in the knowledge management platform module through an HTTP protocol, while monitoring user input and mouse events, and judging whether real-time prompting is needed according to the user input, if real-time prompting is needed, the editor module sends text information to the retriever, the retriever retrieves content prompt information from the public library or through network searching according to the text information, the editor module displays the content prompt information obtained from the retriever in the prompt area, and if real-time prompting is not needed, the user writes by himself.
2. A real-time bid scenario development prompting system according to claim 1, wherein: The data preparation module is further included, the data preparation module converts paper or picture format files into texts by using an OCR recognition tool, and then obtains required text information through text analysis, and stores the original file electronic version and the text information into the public library of the knowledge management platform module at the same time.
3. A real-time bid scenario development prompting system as claimed in claim 1 wherein: The slice embedding module is used for dividing the text information of the public library or the retriever into text blocks according to a slice strategy, each text block is subjected to embedding calculation by a large language model built in the module to generate vector representation, and the vector database is used for storing vector data of the public library generated by the slice embedding module, or obtaining n text blocks closest to the vector data in terms of semantics according to the vector data generated according to the text information, and a document and a link to which each text block belongs, and returning relevant information to the editor module for display in the prompt area of the editor module.
4. The real-time bid proposal preparation prompting system according to claim 1, wherein: The crawler module is further included, the retriever sends text information to the crawler module, the crawler module retrieves relevant document information from the Internet according to the text information, and returns an HTTP link of the document information to the retriever.
5. The real-time bid proposal preparation prompting system according to claim 1, wherein: The prompt area includes a text area for displaying texts, a picture area for displaying pictures and a video area for displaying videos.
6. A method for applying to the real-time prompting system for preparing a bid proposal according to any one of claims 1 to 5, characterized in that, The method includes the following steps: When the user sets to start the receiving real-time writing prompt function of the editor module, the editor module monitors user input and mouse events in the editing area, and judges whether real-time prompting is needed according to the user input, if real-time prompting is needed, the editor module sends text information to the retriever, the retriever retrieves content prompt information from the public library or through network searching according to the text information, the editor module displays the content prompt information obtained from the retriever in the prompt area, and if real-time prompting is not needed, the user writes by himself.
7. A method according to claim 6, wherein, The judgment whether real-time prompting is needed according to the user input includes the following steps: the editor module judges by using a character preset by the user for triggering real-time prompting, when the preset character is input in the editing area of the editor module, it is confirmed that real-time prompting is needed, otherwise, real-time prompting is not performed.
8. A method according to claim 7, characterised in that: The character includes a " 》 " character, a " similar case requirement: " character, a ". " character and a ", " character. If the editor module monitors the user inputting the " 》 " character, the editor module searches forward for the " 《 " character at the current input position of the user, and then sends the file name between the " 《 " and " 》 " characters to the retriever, which retrieves the corresponding file from the knowledge management platform module; If the editor module monitors the user inputting the " similar case requirements: " character, the editor module searches backward for the "." character at the current input position of the user, and then sends the requirement content between the " similar case requirements: " and "." characters to the retriever, which retrieves the corresponding relevant contract case from the knowledge management platform module and returns the editor module with the prompt " contract project name, project type, project construction content, contract signing time, project construction unit, and project signing amount " field information and the contract piece link in the corresponding public library; If the editor module monitors the user inputting the "," or "." character, the editor module simultaneously retrieves two types of text, first, searches forward for the "." character at the current input position of the user, and then obtains the sentence content that the user is inputting or has completed between the searched character position and the current position; second, searches forward for the carriage return character at the current input position of the user, and obtains the paragraph content that the user is inputting or has completed; the editor module simultaneously sends several pieces of text to the retriever, the retriever retrieves and returns the obtained content to the editor module as prompt information, which is displayed in the prompt area of the editor module.
9. A method according to claim 6, wherein, The step of determining whether to give real-time prompts to the user according to the user's input includes the following steps: the editor module determines whether the user is in an editing state by analyzing whether the user is inputting text, and if so, confirms that real-time prompts are needed, otherwise, no real-time prompts are given.
10. The method of claim 6, wherein, The step of determining whether to give real-time prompts to the user according to the user's input includes the following steps: the editor module determines whether the user is in an editing state by analyzing whether the user is inputting text, and if so, confirms that real-time prompts are needed, otherwise, no real-time prompts are given. The step of determining whether to give real-time prompts to the user according to the user's input includes the following steps: the editor module determines whether the user is in an editing state by analyzing whether the user is inputting text, and if so, confirms that real-time prompts are needed, otherwise, no real-time prompts are given.