Website construction method and device based on large model and electronic equipment

Abstract

The invention provides a website construction method and device based on a large model, electronic equipment and a readable storage medium, and relates to the technical field of artificial intelligence such as natural language processing, deep learning and large models. The website construction method based on the large model comprises the steps that user request information is input into the large model, and construction demand information is obtained; the construction demand information is input into the large model, a front-end planning information set is obtained, and the front-end planning information set comprises front-end planning information corresponding to each website page; the front-end planning information set is input into the large model, a DSL description information set is obtained, and the DSL description information set comprises DSL description information corresponding to each website page; and performing page rendering according to the DSL description information set, and obtaining a target website corresponding to the construction demand information according to a rendering result. The website construction steps can be simplified, and the website construction efficiency can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technologies, and particularly to artificial intelligence technologies such as natural language processing, deep learning, and large models. A website construction method, apparatus, electronic device, and readable storage medium based on a large model are provided. Background Art

[0002] When constructing a website in the prior art, professional developers are required to plan website development tasks and write relevant codes according to the planning results. There are problems such as high knowledge costs required for construction, cumbersome construction steps, and low construction efficiency. Therefore, there is an urgent need to provide a website construction method that can simplify construction steps and improve construction efficiency. Summary of the Invention

[0003] According to a first aspect of the present disclosure, a website construction method based on a large model is provided, including: inputting user request information into the large model to obtain construction requirement information; inputting the construction requirement information into the large model to obtain a set of front-end planning information, where the set of front-end planning information includes front-end planning information corresponding to each website page; inputting the set of front-end planning information into the large model to obtain a set of domain-specific language (DSL) description information, where the set of DSL description information includes DSL description information corresponding to each website page; performing page rendering according to the set of DSL description information, and obtaining a target website corresponding to the construction requirement information according to the rendering result.

[0004] According to a second aspect of the present disclosure, a website construction apparatus based on a large model is provided, including: a processing unit configured to input user request information into the large model to obtain construction requirement information; a planning unit configured to input the construction requirement information into the large model to obtain a set of front-end planning information, where the set of front-end planning information includes front-end planning information corresponding to each website page; a first generation unit configured to input the set of front-end planning information into the large model to obtain a set of domain-specific language (DSL) description information, where the set of DSL description information includes DSL description information corresponding to each website page; and a construction unit configured to perform page rendering according to the set of DSL description information, and obtain a target website corresponding to the construction requirement information according to the rendering result.

[0005] According to a third aspect of the present disclosure, an electronic device is provided, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor is enabled to execute the method as described above.

[0006] According to a fourth aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause the computer to execute the method as described above.

[0007] According to a fifth aspect of the present disclosure, there is provided a computer program product including a computer program, which implements the method as described above when executed by a processor.

[0008] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The drawings are used to better understand the solution and do not constitute a limitation to the present disclosure. Among them:

[0010] Figure 1 is a schematic diagram according to the first embodiment of the present disclosure;

[0011] Figure 2 is a schematic diagram according to the second embodiment of the present disclosure;

[0012] Figure 3 is a schematic diagram according to the third embodiment of the present disclosure;

[0013] Figure 4 is a schematic diagram according to the fourth embodiment of the present disclosure;

[0014] Figure 5 is a schematic diagram according to the fifth embodiment of the present disclosure;

[0015] Figure 6 is a schematic diagram according to the sixth embodiment of the present disclosure;

[0016] Figure 7 is a block diagram of an electronic device for implementing the method for building a website based on a large model according to the embodiments of the present disclosure. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] The following makes an explanation of the exemplary embodiments of the present disclosure in conjunction with the drawings. Various details of the embodiments of the present disclosure are included to facilitate understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, the description of well-known functions and mechanisms is omitted below.

[0018] Figure 1 is a schematic diagram according to the first embodiment of the present disclosure. As Figure 1As shown in the figure, the method for building a website based on a large model in this embodiment specifically includes the following steps:

[0019] S101. Input the user request information into the large model to obtain the construction requirement information;

[0020] S102. Input the construction requirement information into the large model to obtain a set of front-end planning information, where the set of front-end planning information includes the front-end planning information corresponding to each website page;

[0021] S103. Input the set of front-end planning information into the large model to obtain a set of domain-specific language (DSL) description information, where the set of DSL description information includes the DSL description information corresponding to each website page;

[0022] S104. Perform page rendering according to the set of DSL description information, and obtain the target website corresponding to the construction requirement information according to the rendering result.

[0023] The method for building a website based on a large model in this embodiment, by virtue of the powerful generation ability of the large model, can achieve the purpose of building a website only according to the request information input by the user, without the user having to plan the website development tasks and write code by themselves when building a website, which can simplify the website construction steps and improve the website construction efficiency.

[0024] In this embodiment, the large model is a deep learning model trained using a large amount of data, which can be used to process various language tasks, such as text summary generation, question answering, translation, code generation, and other tasks; the large model in this embodiment can be a large language model (LLM) or a multimodal large model.

[0025] When this embodiment executes S101, it can obtain the query (i.e., query) input by the user through the input end as the user request information; it can be understood that the user request information in this embodiment can be the requirement information corresponding to building a website (i.e., the user intention of the user request information is the website construction intention), or other types of requirement information (i.e., the user intention of the user request information is not the website construction intention), such as chatting, knowledge answering, etc.

[0026] Specifically, when this embodiment executes S101 to input the user request information into the large model to obtain the construction requirement information, the implementation method that can be adopted is: input the user request information into the large model, and the large model performs intention recognition to obtain the intention recognition result; in response to determining that the obtained intention recognition result is the website construction intention, the large model extracts key information from the user request information; according to the extracted key information, the construction requirement information is obtained.

[0027] In this embodiment, the construction requirement information includes the user's requirements for the website type of the website to be constructed (i.e., what type of website to construct), the type requirements for the website pages (i.e., what types of pages the website needs to have), the functional requirements for the website pages (i.e., what functions the website pages need to provide), and the design requirements for the website pages (i.e., what layout, style, etc. the website pages need to have); when executing S101 in this embodiment, the large model can extract information related to type requirements, functional requirements, design requirements, etc. from the user request information as the key information for obtaining the construction requirement information.

[0028] That is to say, in this embodiment, through the large model, intention recognition is performed based on the user request information. When it is determined that the user's intention is a website construction intention, the large model then extracts the key information from the user request information as the construction requirement information, which can avoid the problem of being unable to extract key information from user request information that does not have a website construction intention, and improves the accuracy and efficiency of obtaining the construction requirement information.

[0029] It can be understood that when executing S101 in this embodiment, intention recognition may not be performed, and the large model can directly extract key information based on the user request information; for example, in a current scenario of website construction (such as when the application used by the current user is a website construction application), the query input by the user is usually a query related to website construction.

[0030] In practical applications, there may also be the following scenarios: Although the user intention corresponding to the user request information is a website construction intention, the user request information does not include key information or includes less key information, resulting in the construction requirement information obtained by the large model based on the user request information being too simple and not rich or perfect enough, thus affecting the subsequent website construction process; for example, when the user request information is "Please help me build a website" or the user request information is "Please help me build a library website", the key information included is less, only including the requirement of "building a website" or the website type of "building a library website".

[0031] To ensure that more comprehensive and complete construction requirement information can be obtained, when this embodiment executes S101 to obtain construction requirement information based on the extracted key information, it may further include the following: in response to determining that the extracted key information does not meet the preset requirements, the large model generates at least one interaction question based on the extracted key information, and the generated interaction question is used to obtain more comprehensive and complete requirement information (or requirement information not included in the key information); use the generated at least one interaction question to interact with the user, and obtain construction requirement information according to the reply information input by the user for each interaction question; in addition, this embodiment may also use the reply information input by the user and the previously extracted key information together as the construction requirement information.

[0032] That is to say, this embodiment obtains the construction requirement information for building a website through one or more rounds of conversations between the large model and the user, ensuring that the obtained construction requirement information is more comprehensive and complete, making the website built based on this construction requirement information more in line with the actual needs of the user, and improving the construction accuracy of the website.

[0033] Among them, when this embodiment executes S101, it may determine that the extracted key information does not meet the preset requirements when determining that the number of the extracted key information is small or when determining that the number of words of the extracted key information is small.

[0034] For example, if the extracted key information is "build a library website", the interaction questions generated by the large model in this embodiment based on this key information may be "What pages are included in this library website", "What is the page style of this library website", "What functions does the page of this library website need to have", etc., so as to obtain the type requirements, function requirements and design requirements of the website page according to the reply information of the user for each interaction question.

[0035] In addition, when this embodiment executes S101, it may further include the following: in response to determining that the obtained intent recognition result is not the website construction intent, the large model generates a clarification question, and the clarification question generated by the large model is used for the user to clarify their actual needs; use the generated clarification question to interact with the user, and re-obtain the user request information according to the reply information input by the user for the clarification question.

[0036] That is to say, in the case where the user's intention is not the website construction intention, this embodiment can generate a clarification question through the large model to guide the user to re-enter. For example, the generated clarification question can be "I am a website construction assistant. Please enter information related to the website you hope to build, and I will build the website for you", ensuring that the user intention corresponding to the re-obtained user request information is the website construction intention, so as to facilitate the subsequent continuation of the website construction process.

[0037] It can be understood that when this embodiment executes S101, it can also modify the construction requirement information obtained by the large model according to the modification instruction input by the user, such as modifying the page style, adding or deleting functions required by the page, etc., so as to ensure that the obtained construction requirement information meets the actual needs of the user and improve the flexibility of obtaining the construction requirement information.

[0038] After this embodiment executes S101 to obtain the construction requirement information, it executes S102 to input the construction requirement information into the large model to obtain a set of front-end planning information; the set of front-end planning information obtained in this embodiment includes the front-end planning information corresponding to each website page.

[0039] In this embodiment, the front-end planning information corresponding to the website page may include the type information of the website page (such as the guide page, introduction page, detail page, display page, etc.), the function information of the website page (such as the display function, purchase function, introduction function, guide function, etc.), and the design information of the website page (such as page layout, page style, page header, footer, interaction method, etc.).

[0040] When this embodiment executes S102, it can input the construction requirement information and the corresponding prompt text (i.e., the prompt text) into the large model together, so that the large model plans the website development task according to the construction requirement information, and obtains a set of front-end planning information including at least one front-end planning information.

[0041] In this embodiment, when the large model plans the website development task according to the construction requirement information, in addition to obtaining a set of front-end planning information, it can also obtain a set of back-end planning information; among them, the set of back-end planning information includes the back-end planning information corresponding to each website page.

[0042] In this embodiment, if all the pages included in the website to be built are static pages, there is no need to plan the back-end of the page when planning the website development task; if not all the pages included in the website to be built are static pages, when planning the website development task, it is also necessary to plan the back-end of the page when planning the front-end of the page.

[0043] In this embodiment, the backend planning information corresponding to the website page may include application programming interface information required for the website page to implement a certain function, database information of the website page, business logic information corresponding to the website page, and so on.

[0044] In this embodiment, the set of frontend planning information and the set of backend planning information obtained through the large model can be used as a simplified product requirement document (PRD, Product Requirement Document) for building the website. The product requirement document may further include user requirement information.

[0045] That is to say, in this embodiment, through the large model, the set of frontend planning information is obtained according to the user requirement information, without the need for the user to manually plan the website development tasks (i.e., manually write the product requirement document), which can effectively reduce the planning cost of the website development tasks. Moreover, the large model has strong generation ability and can also improve the planning efficiency and planning accuracy of the website development tasks.

[0046] After obtaining the set of frontend planning information in S102, this embodiment executes S103 to input the set of frontend planning information into the large model to obtain a set of Domain-Specific Language (DSL) description information. The set of DSL description information obtained in this embodiment includes DSL description information corresponding to each website page.

[0047] In this embodiment, a Domain-Specific Language (DSL) is a computer language designed to be used within a specific application domain, which can provide more precise abstractions and more efficient expressions to describe the tasks and logics in that domain.

[0048] In this embodiment, the DSL description information corresponding to each website page can be the visualization description result (i.e., the visualization page) of each website page. Among them, the DSL description information may include the layout visualization result, text visualization result, picture visualization result, block (section) or component visualization result of the website page, and so on.

[0049] When executing S103, this embodiment can input the set of frontend planning information and the corresponding prompt text (i.e., prompt) into the large model, so that the large model generates the corresponding page layout, the blocks or components included in the page, the text in the page, and the pictures in the page according to each frontend planning information in the set of frontend planning information, and then obtains the DSL description information corresponding to each website page according to the above generated content.

[0050] That is to say, in this embodiment, through the large model, according to the front-end planning information corresponding to each website page, the layout, sections, text, and pictures corresponding to each website page are generated, so as to obtain the DSL description information corresponding to each website page based on the generated content, which can improve the generation efficiency and intelligence of the DSL description information.

[0051] After this embodiment executes S103 to obtain the DSL description information set, it may further include the following: displaying each DSL description information in the DSL description information set to the user; and in response to receiving a modification instruction from the user, modifying the corresponding DSL description information.

[0052] Among them, the modification instruction of the user in this embodiment may be a specific modification instruction, such as replacing a certain picture in the DSL description information with a specific picture (such as a picture provided by the user); it may also be a fuzzy modification instruction, such as "replace a certain picture in the DSL description information". In this case, the large model regenerates the picture and uses the regenerated picture to replace the original picture. For example, in the scenario of building a library website, the book picture on the corresponding book display page is generated by the large model. If the user is not satisfied with the book picture, it can be regenerated by the large model.

[0053] That is to say, since the DSL description information obtained in this embodiment can be visualized, visualizing and displaying the DSL description information to the user enables the user to modify the content (such as pictures, text, etc.) in the DSL description information, further improving the matching degree between the obtained DSL description information corresponding to the website page and the actual needs of the user.

[0054] After this embodiment executes S103 to obtain the DSL description information set, it executes S104 to perform page rendering according to the DSL description information set, and obtains the target website corresponding to the construction requirement information based on the rendering result; that is, obtains the target website according to the rendering results of different DSL description information.

[0055] When this embodiment executes S104, if the large model only obtains the front-end planning information set according to the construction requirement information (that is, the website page does not require backend capabilities), the target website can be obtained based on the rendering results obtained from each DSL description information.

[0056] When this embodiment executes S104 to perform page rendering according to the DSL description information set and obtain the target website corresponding to the construction requirement information, the following content may also be included: obtaining a back-end planning information set, where the obtained back-end planning information set includes the back-end planning information corresponding to each website page; performing page rendering according to the DSL description information set to obtain the rendering result of each website page; obtaining the back-end interface information of each website page according to the back-end planning information set; and obtaining the target website corresponding to the construction requirement information according to the rendering result and the back-end interface information of each website page.

[0057] That is to say, when this embodiment plans the website development task according to the construction requirement information by the large model, if the front-end planning information set and the back-end planning information set are obtained at the same time, the target website will be constructed jointly according to the front-end planning information set and the back-end planning information set, so that the website pages in the constructed target website have corresponding back-end capabilities.

[0058] Among them, when this embodiment executes S104 to obtain the back-end interface information of each website page according to the back-end planning information set, for each website page, the real-time generation of an interface (such as an API, an application programming interface) can be performed through the large model according to the back-end planning information corresponding to the website page, and the generated result is used as the back-end port information corresponding to the website page; for each website page, the existing interface (such as a third-party interface) can also be determined according to the back-end planning information corresponding to the website page, and the determined existing interface is used as the back-end port information corresponding to the website page.

[0059] When this embodiment executes S104 to obtain the target website corresponding to the construction requirement information according to the rendering result and the back-end interface information of each website page, for each website page, the rendering result corresponding to the website page can be matched with the back-end interface information, and then the target website can be obtained according to the matching result of each website page.

[0060] It can be understood that this embodiment can also perform quality inspection on the target website before deploying the target website, so that the target website is deployed again in the case of passing the quality inspection.

[0061] Through the website construction method based on the large model provided by this embodiment, the large model is used to sequentially obtain the construction requirement information, the front-end planning information set, and the DSL description information set, so as to complete the construction of the target website based on the obtained DSL description information set. The entire process of website construction realizes intelligence and automation, without the need for users to manually plan the website development task and write code, reduces the knowledge cost required for website construction, and can also improve the accuracy and efficiency of website construction.

[0062] Figure 2 It is a schematic diagram according to the second embodiment of the present disclosure. As Figure 2 shown, when implementing S103 "input the front-end planning information set into the large model to obtain a domain-specific language (DSL) description information set" in this embodiment, the implementation method that can be adopted is:

[0063] S201. For each piece of front-end planning information in the front-end planning information set, the large model retrieves page templates in a preset page template library according to this front-end planning information;

[0064] S202. In response to retrieving a page template corresponding to this front-end planning information, the large model generates page text and / or page pictures corresponding to this page template;

[0065] S203. According to this page template, as well as the page text and / or page pictures corresponding to this page template, obtain the DSL description information corresponding to this front-end planning information, and obtain the DSL description information set according to the DSL description information corresponding to each piece of front-end planning information.

[0066] That is to say, in this embodiment, by combining the page templates in the preset page template library and the generation ability of the large model, the DSL description information corresponding to the front-end planning information or the website page is obtained according to the retrieved page template, and then the DSL description information set is obtained according to the DSL description information corresponding to each piece of front-end planning information, which can improve the generation efficiency and generation accuracy of the DSL description information set.

[0067] In this embodiment, the preset page template library includes multiple page templates, each page template is pre-generated by the large model, and the page templates are stored in the preset page template library in the form of DSL description information.

[0068] When implementing S201 in this embodiment, retrieval can be performed in the preset page template library according to the page type information or page function information in the front-end planning information, so that the retrieved page template corresponds to the corresponding page type or has the corresponding page function.

[0069] In this embodiment, the retrieved page templates include multiple page blocks, and different page blocks can correspond to different display information, different functions, etc.

[0070] When executing S202 in this embodiment, the page text and / or page images generated by the large model actually correspond to different page blocks in the page template, that is, according to different page blocks, as well as construction requirement information or front-end planning information, to generate text and images corresponding to different page blocks; for example, if the retrieved page template corresponds to a display page and the construction requirement information is "construct a library website", then a text introducing books and a book picture can be generated according to the above content.

[0071] When executing S203 in this embodiment, the generated page text and / or page images can be filled into the page template. Specifically, the page text and page images are filled into the corresponding page blocks in the page template, so as to complete the content filling of the blocks and obtain the DSL description information corresponding to the front-end planning information or the website web page.

[0072] When executing S203 in this embodiment, after obtaining the DSL description information, the DSL description information set can be obtained according to the DSL description information corresponding to each front-end planning information.

[0073] Figure 3 It is a schematic diagram according to the third embodiment of the present disclosure. As Figure 3 shown, when executing S103 "input the front-end planning information set into the large model to obtain a domain-specific language DSL description information set" in this embodiment, the implementation method that can be adopted is:

[0074] S301. For each front-end planning information in the front-end planning information set, the large model retrieves the page template in the preset page template library according to the front-end planning information;

[0075] S302. In response to not retrieving the page template corresponding to the front-end planning information, the large model retrieves the block template in the preset block template library according to the front-end planning information;

[0076] S303. In response to retrieving the block template corresponding to the front-end planning information, the large model generates the page text and / or page images corresponding to the block template;

[0077] S304. According to the block template and the page text and / or page images corresponding to the block template, obtain the DSL description information corresponding to the front-end planning information, and obtain the DSL description information set according to the DSL description information corresponding to each front-end planning information.

[0078] That is to say, based on the preset page template library, this embodiment will also use the preset block template library, combine the generation ability of the large model, and obtain the DSL description information corresponding to the front-end planning information or website page according to the retrieved block template. Furthermore, the DSL description information set can be obtained according to the DSL description information corresponding to each front-end planning information, which can improve the generation efficiency and accuracy of the DSL description information set.

[0079] In this embodiment, the preset block template library includes multiple block templates. Each block template is pre-generated by the large model and stored in the preset block template library in the form of DSL description information; for example, form block templates, graphic and text display block templates, business component (lottery, payment, voting, etc.) block templates, etc.

[0080] When this embodiment executes S302, it can retrieve in the preset block template library according to the page type information or page function information in the front-end planning information, so that the retrieved block template corresponds to the corresponding page type or has the corresponding page function.

[0081] When this embodiment executes S303, the large model generates page text and / or page pictures corresponding to the retrieved block template and meeting the construction requirement information or front-end planning information; for example, the text and pictures corresponding to the graphic and text display block template, the text and / or pictures corresponding to the form block template, etc.

[0082] When this embodiment executes S304, it can fill the generated page text and / or page pictures into the block template, thereby completing the content filling of the block and obtaining the DSL description information corresponding to the front-end planning information or website web page.

[0083] It can be understood that in the case where this embodiment does not retrieve the page template corresponding to the front-end planning information, it can obtain the default page template, and then place the block template with the completed content filling into the default page template, thereby obtaining the DSL description information corresponding to the front-end planning information.

[0084] Figure 4 It is a schematic diagram according to the fourth embodiment of the present disclosure. As Figure 4 shown in the figure, the website construction method based on the large model of this embodiment may further include the following content:

[0085] S401. For each front-end planning information in the front-end planning information set, in response to not retrieving the block template corresponding to the front-end planning information, the large model generates the source code corresponding to the front-end planning information according to the front-end planning information;

[0086] S402. Perform page rendering based on the generated source code, and obtain the website page corresponding to the front-end planning information according to the rendering result.

[0087] That is to say, in the case where no template corresponding to the front-end planning information is retrieved in the preset page template library and the preset block template library, the large model cannot generate DSL description information that can be visualized. Then, directly through the large model, generate the source code (the source code cannot be visualized) according to the front-end planning information, and then obtain the corresponding website page according to the generated source code, ensuring that the construction of the target website can be completed in any case.

[0088] It can be understood that when executing S401 in this embodiment, if there is no block template corresponding to a certain front-end planning information, the source code can be generated only for this front-end planning information, or the source code can be generated for all the front-end planning information in the front-end planning information set, that is, the large model generates the source code corresponding to all website pages.

[0089] When executing S401 in this embodiment, the corresponding back-end planning information (i.e., corresponding to the same website page) of the front-end planning information can also be obtained, and then the large model generates the source code corresponding to the website page according to the front-end planning information and the back-end planning information corresponding to the website page.

[0090] When generating the source code in this embodiment, the large model can detect the syntax and dependencies of the source code in sequence, run tests, and detect test tools. Only when the syntax and dependency detection pass, the operation is successful, and the test tool detection passes, will the source code be used for subsequent page rendering.

[0091] That is to say, this embodiment can complete the construction of the target website through a technical path different from the DSL generation path, that is, the source code generation path, ensuring that the construction of the target website can be completed in different situations.

[0092] Figure 5 It is a schematic diagram according to the fifth embodiment of the present disclosure. Figure 5 Shows the framework diagram of the website construction method based on the large model in this embodiment: Figure 5 It includes multiple agents, such as an intent recognition agent, a requirement clarification agent, a requirement information acquisition agent, a planning agent, an instruction modification agent, a DSL generation agent, and a source code generation agent. That is, in this embodiment, the cooperation between multiple agents of the large model is used to complete the construction of the website.

[0093] Specifically, the intent recognition agent is used to identify whether the user's intent is a website construction intent, the requirement clarification agent is used to generate clarification questions, the requirement information acquisition agent is used to generate interaction questions to interact with the user, the planning agent is used to generate a set of planning information, the instruction modification agent is used to modify requirement information, text, pictures, etc. according to the user's instructions, the DSL generation agent is used to generate a set of DSL description information for the corresponding website page by means of a preset page template library or a preset block template library, and the source code generation agent is used to generate the source code of the corresponding website page.

[0094] In this embodiment, different agents correspond to different interfaces. By invoking the corresponding interfaces, the recognition of the user's intent, the generation of requirement information, the generation of planning information, etc. can be realized.

[0095] Figure 6 It is a schematic diagram according to the sixth embodiment of the present disclosure. As Figure 6 shown, the website construction device 600 based on the large model of this embodiment includes:

[0096] A processing unit 601, configured to input user request information into the large model to obtain construction requirement information;

[0097] A planning unit 602, configured to input the construction requirement information into the large model to obtain a set of front-end planning information, where the set of front-end planning information includes front-end planning information corresponding to each website page;

[0098] A first generation unit 603, configured to input the set of front-end planning information into the large model to obtain a set of domain-specific language DSL description information, where the DSL description information set includes DSL description information corresponding to each website page;

[0099] A construction unit 604, configured to perform page rendering according to the DSL description information set, and obtain a target website corresponding to the construction requirement information according to the rendering result.

[0100] The processing unit 601 can obtain the query (i.e., query) input by the user through the input end as the user request information; it can be understood that the user request information in this embodiment can be requirement information corresponding to the constructed website (i.e., the user's intent of the user request information is a website construction intent), or other types of requirement information (i.e., the user's intent of the user request information is not a website construction intent), such as chatting, knowledge answering, etc.

[0101] Specifically, when the processing unit 601 inputs the user request information into the large model to obtain the construction requirement information, the implementation method that can be adopted is as follows: input the user request information into the large model, and the large model performs intent recognition to obtain the intent recognition result; in response to determining that the obtained intent recognition result is a website construction intent, the large model extracts key information from the user request information; according to the extracted key information, the construction requirement information is obtained.

[0102] In this embodiment, the construction requirement information includes the user's requirement for the website type of the website to be constructed (i.e., what type of website to construct), the type requirement of the website page (i.e., what kind of page the website needs to have), the functional requirement of the website page (i.e., what functions the website page needs to provide), and the design requirement of the website page (i.e., what layout, what style, etc. the website page needs to have); the processing unit 601 can use the information related to the type requirement, functional requirement, design requirement, etc. extracted by the large model from the user request information as the key information for obtaining the construction requirement information.

[0103] That is to say, the processing unit 601 performs intent recognition according to the user request information through the large model. When it is determined that the user intent is a website construction intent, the large model then extracts key information from the user request information as the construction requirement information, which can avoid the problem of being unable to extract key information from user request information that does not have a website construction intent, and improves the accuracy and efficiency of obtaining the construction requirement information.

[0104] It can be understood that the processing unit 601 may also not perform intent recognition, and the large model directly extracts key information according to the user request information.

[0105] In actual applications, there may also be the following scenarios: Although the user intent corresponding to the user request information is a website construction intent, the user request information does not include key information or includes less key information, resulting in the construction requirement information obtained by the large model according to the user request information being too simple and not rich and perfect enough, thus affecting the subsequent website construction process.

[0106] To ensure that more rich and perfect construction requirement information can be obtained, when the processing unit 601 obtains the construction requirement information according to the extracted key information, it may also include the following content: in response to determining that the extracted key information does not meet the preset requirements, the large model generates at least one interaction question according to the extracted key information; uses the generated at least one interaction question to interact with the user, and obtains the construction requirement information according to the reply information input by the user for each interaction question;

[0107] That is to say, the processing unit 601 obtains the construction requirement information for building a website through one or more rounds of conversations between the large model and the user, ensuring that the obtained construction requirement information is richer and more perfect, making the website built based on this construction requirement information more in line with the actual needs of the user and improving the construction accuracy of the website.

[0108] Among them, the processing unit 601 can determine that the extracted key information does not meet the preset requirements when the number of the extracted key information is small or when the number of characters of the extracted key information is small.

[0109] In addition, the processing unit 601 can also perform the following: in response to determining that the obtained intent recognition result is not the website construction intent, generate a clarification question by the large model; use the generated clarification question to interact with the user, and re-obtain the user request information according to the reply information input by the user for the clarification question.

[0110] That is to say, when the user intent is not the website construction intent, the processing unit 601 guides the user to re-enter by generating a clarification question through the large model, ensuring that the user intent corresponding to the re-obtained user request information is the website construction intent, so as to facilitate the subsequent continuation of the website construction process.

[0111] It can be understood that the processing unit 601 can also modify the construction requirement information obtained by the large model according to the modification instruction input by the user, so as to ensure that the obtained construction requirement information meets the actual needs of the user.

[0112] After the processing unit 601 obtains the construction requirement information in this embodiment, the planning unit 602 inputs the construction requirement information into the large model to obtain a set of front-end planning information; the set of front-end planning information obtained in this embodiment includes the front-end planning information corresponding to each website page.

[0113] In this embodiment, the front-end planning information corresponding to the website page may include the type information of the website page (such as the guide page, introduction page, detail page, display page, etc.), the function information of the website page (such as the display function, purchase function, introduction function, guide function, etc.), and the design information of the website page (such as the page layout, page style, page header, footer, interaction method, etc.).

[0114] The planning unit 602 can input the construction requirement information and the corresponding prompt text (i.e., the prompt text) into the large model together, so that the large model plans the website development task according to the construction requirement information, and thus obtains a set of front-end planning information.

[0115] In this embodiment, when the large model plans the website development task according to the construction requirement information, in addition to obtaining the front-end planning information set, it can also obtain the back-end planning information set; among them, the back-end planning information set includes the back-end planning information corresponding to each website page.

[0116] In this embodiment, the back-end planning information corresponding to the website page may include application program interface information required when the website page implements a certain function, database information of the website page, business logic information corresponding to the website page, and so on.

[0117] That is to say, the planning unit 602 obtains the front-end planning information set through the large model according to the user requirement information, without the user manually planning the website development task (that is, manually writing the product requirement document), which can effectively reduce the planning cost of the website development task, and the large model has strong generation ability, and can also improve the planning efficiency and planning accuracy of the website development task.

[0118] After this embodiment obtains the front-end planning information set by the planning unit 602, the first generation unit 603 inputs the front-end planning information set into the large model to obtain the domain-specific language DSL description information set; the DSL description information set obtained in this embodiment includes the DSL description information corresponding to each website page.

[0119] In this embodiment, the domain-specific language (DSL) is a computer language designed to be used within a specific application domain, which can provide more precise abstractions and more efficient expression methods to describe the tasks and logics in this domain.

[0120] In this embodiment, the DSL description information corresponding to each website page can be the visualization description result (i.e., the visualization page) of each website page; among them, the DSL description information may include the layout visualization result, text visualization result, picture visualization result, block (section) or component visualization result of the website page, and so on.

[0121] The first generation unit 603 can input the front-end planning information set and the corresponding prompt text (i.e., prompt) into the large model, so that the large model generates the corresponding page layout, the blocks or components included in the page, the text in the page, and the pictures in the page according to each front-end planning information in the front-end planning information set, and then obtains the DSL description information corresponding to each website page according to the above generated content.

[0122] That is to say, the first generation unit 603 uses a large model to generate the layout, sections, text, and images for each website page according to the front-end planning information corresponding to each website page, and thus obtains the DSL description information for each website page based on the generated content, which can improve the generation efficiency and intelligence of the DSL description information.

[0123] After obtaining the DSL description information set, the first generation unit 603 may further include the following: presenting each DSL description information in the DSL description information set to the user; and modifying the corresponding DSL description information in response to receiving a modification instruction from the user.

[0124] That is to say, since the DSL description information obtained in this embodiment can be visualized, presenting the DSL description information to the user visually enables the user to modify the content (such as pictures, text, etc.) in the DSL description information, further improving the matching degree between the obtained DSL description information of the corresponding website page and the actual needs of the user.

[0125] In addition, when the first generation unit 603 inputs the front-end planning information set into the large model to obtain the domain-specific language DSL description information set, the following implementation manner may also be adopted: for each front-end planning information in the front-end planning information set, the large model retrieves a page template in a preset page template library according to the front-end planning information; in response to retrieving a page template corresponding to the front-end planning information, the large model generates page text and / or page images corresponding to the page template; based on the page template and the page text and / or page images corresponding to the page template, the DSL description information corresponding to the front-end planning information is obtained, and the DSL description information set is obtained based on the DSL description information corresponding to each front-end planning information.

[0126] That is to say, the first generation unit 603 combines the page templates in the preset page template library and the generation ability of the large model, and obtains the DSL description information corresponding to the front-end planning information or the website page according to the retrieved page template, and then obtains the DSL description information set based on the DSL description information corresponding to each front-end planning information, which can improve the generation efficiency and accuracy of the DSL description information set.

[0127] In this embodiment, the preset page template library includes multiple page templates, each page template is pre-generated by the large model, and the page templates are stored in the preset page template library in the form of DSL description information.

[0128] When the first generation unit 603 inputs the front-end planning information set into the large model to obtain the domain-specific language (DSL) description information set, the following implementation method can also be adopted: for each front-end planning information in the front-end planning information set, the large model retrieves page templates in the preset page template library according to the front-end planning information; in response to not retrieving a page template corresponding to the front-end planning information, the large model retrieves block templates in the preset block template library according to the front-end planning information; in response to retrieving a block template corresponding to the front-end planning information, the large model generates page text and / or page pictures corresponding to the block template; according to the block template, and the page text and / or page pictures corresponding to the block template, DSL description information corresponding to the front-end planning information is obtained, and a DSL description information set is obtained according to the DSL description information corresponding to each front-end planning information.

[0129] That is to say, based on the preset page template library, the first generation unit 603 also uses the preset block template library, combines the generation ability of the large model, and obtains DSL description information corresponding to the front-end planning information or website page according to the retrieved block template, and then obtains a DSL description information set according to the DSL description information corresponding to each front-end planning information, which can improve the generation efficiency and generation accuracy of the DSL description information set.

[0130] In this embodiment, the preset block template library includes multiple block templates, each block template is pre-generated by the large model, and the block templates are stored in the preset block template library in the form of DSL description information; for example, form block templates, graphic display block templates, business component (lottery, payment, voting, etc.) block templates, etc.

[0131] The website construction device 600 based on the large model in this embodiment may further include a second generation unit 605, which is used to perform the following: for each front-end planning information in the front-end planning information set, in response to not retrieving a block template corresponding to the front-end planning information, the large model generates source code corresponding to the front-end planning information according to the front-end planning information; page rendering is performed according to the generated source code, and a website page corresponding to the front-end planning information is obtained according to the rendering result.

[0132] That is to say, in the case where no template corresponding to the front-end planning information is retrieved in both the preset page template library and the preset block template library, the large model cannot generate visualizable DSL description information, so the second generation unit 605 directly generates source code (the source code cannot be visualized) according to the front-end planning information through the large model, and then obtains the corresponding website page according to the generated source code, ensuring that the target website can be constructed in any case.

[0133] It can be understood that if there is no block template corresponding to a certain front-end planning information, the second generation unit 605 can generate the source code only for this front-end planning information, or the second generation unit 605 can also generate the source code for all the front-end planning information in the front-end planning information set, that is, the large model is used to generate the source code corresponding to all website pages.

[0134] In this embodiment, after the first generation unit 603 obtains the DSL description information set, the construction unit 604 performs page rendering according to the DSL description information set, and obtains the target website corresponding to the construction requirement information according to the rendering result; that is, the target website is obtained according to the rendering results of different DSL description information.

[0135] If the large model only obtains the front-end planning information set according to the construction requirement information (that is, the website page does not require back-end capabilities), the construction unit 604 can obtain the target website according to the rendering results obtained from each DSL description information in the DSL description information set.

[0136] When the construction unit 604 performs page rendering according to the DSL description information set and obtains the target website corresponding to the construction requirement information, it may further include the following: obtaining the back-end planning information set, where the obtained back-end planning information set includes the back-end planning information corresponding to each website page; performing page rendering according to the DSL description information set to obtain the rendering results of each website page; obtaining the back-end interface information of each website page according to the back-end planning information set; and obtaining the target website corresponding to the construction requirement information according to the rendering results and back-end interface information of each website page.

[0137] That is to say, when the large model plans the website development task according to the construction requirement information, if the back-end planning information set is obtained while obtaining the front-end planning information set, the construction unit 604 will jointly complete the construction of the target website according to the front-end planning information set and the back-end planning information set, so that the website pages in the constructed target website have corresponding back-end capabilities.

[0138] Among them, when the construction unit 604 obtains the back-end interface information of each website page according to the back-end planning information set, for each website page, it can perform real-time generation of an interface (such as an API, an application programming interface) according to the back-end planning information corresponding to the website page, and use the generation result as the back-end port information corresponding to the website page; or for each website page, it can determine the existing interface (such as a third-party interface) according to the back-end planning information corresponding to the website page, and use the determined existing interface as the back-end port information corresponding to the website page.

[0139] When the construction unit 604 obtains the target website corresponding to the construction requirement information based on the rendering result of each website page and the backend interface information, for each website page, it can match the rendering result corresponding to the website page with the backend interface information, and then obtain the target website according to the matching result of each website page.

[0140] It can be understood that the construction unit 604 can also perform quality inspection on the target website before deploying the target website, so that when the quality inspection passes, the target website is deployed.

[0141] In the technical solution of the present disclosure, the acquisition, storage, and application of the user's personal information involved all comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0142] In the technical solution of the present disclosure, before obtaining or collecting the user's personal information, the user's authorization or consent is obtained.

[0143] According to the embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0144] As Figure 7 shown, it is a block diagram of an electronic device for a website construction method based on a large model according to an embodiment of the present disclosure. The electronic device is intended to represent various forms of digital computers, such as, a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, a personal digital processor, a cellular phone, a smart phone, a wearable device, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0145] As Figure 7 shown, the device 700 includes a computing unit 701, which can execute various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 702 or the computer program loaded from the storage unit 708 into the random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the device 700 can also be stored. The computing unit 701, the ROM 702, and the RAM 703 are connected to each other through a bus 704. The input / output (I / O) interface 705 is also connected to the bus 704.

[0146] Multiple components in device 700 are connected to I / O interface 705, including: input unit 706, such as a keyboard, mouse, etc.; output unit 707, such as various types of displays, speakers, etc.; storage unit 708, such as a disk, optical disc, etc.; and communication unit 709, such as a network card, modem, wireless communication transceiver, etc. Communication unit 709 allows device 700 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0147] Computing unit 701 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of computing unit 701 include but are not limited to a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Computing unit 701 executes the various methods and processes described above, such as the website construction method based on a large model. For example, in some embodiments, the website construction method based on a large model can be implemented as a computer software program that is tangibly contained in a machine-readable medium, such as storage unit 708.

[0148] In some embodiments, part or all of the computer program can be loaded and / or installed onto device 700 via ROM 702 and / or communication unit 709. When the computer program is loaded into RAM 703 and executed by computing unit 701, one or more steps of the website construction method based on a large model described above can be executed. Alternatively, in other embodiments, computing unit 501 can be configured to execute the website construction method based on a large model by any other suitable means (e.g., by means of firmware).

[0149] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGA), application-specific integrated circuits (ASIC), application-specific standard products (ASSP), system-on-chip systems (SOC), complex programmable logic devices (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0150] The program code for implementing the methods of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable large model-based website construction devices, such that when the program codes are executed by the processor or controller, the functions / operations specified in the flowchart and / or block diagram are implemented. The program codes can be executed entirely on the machine, partially on the machine, executed partially on the machine and partially on a remote machine as an independent software package, or executed entirely on a remote machine or server.

[0151] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0152] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for presenting information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, speech input, or tactile input).

[0153] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend, middleware, or frontend components. The components of the system can be interconnected to each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), and the Internet.

[0154] A computer system can include a client and a server. The client and the server are generally far from each other and typically interact through a communication network. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system, solving the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services ("Virtual Private Server", or simply "VPS"). The server can also be a server of a distributed system, or a server combined with a blockchain.

[0155] It should be understood that various forms of the processes shown above can be used, steps can be reordered, added, or deleted. For example, the steps recited in this disclosure can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this is not limited herein.

[0156] The above specific embodiments do not constitute a limitation on the protection scope of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the protection scope of this disclosure.

Claims

1. A website construction method based on a large model, comprising: Inputting user request information into the large model to obtain construction requirement information; Inputting the construction requirement information into the large model to obtain a set of front-end planning information, where the set of front-end planning information includes front-end planning information corresponding to each website page; Inputting the set of front-end planning information into the large model to obtain a set of domain-specific language (DSL) description information, where the set of DSL description information includes DSL description information corresponding to each website page; Performing page rendering according to the set of DSL description information, and obtaining a target website corresponding to the construction requirement information based on the rendering result.

2. The method according to claim 1, wherein The step of inputting user request information into the large model to obtain construction requirement information includes: Inputting the user request information into the large model, and having the large model perform intent recognition to obtain an intent recognition result; In response to determining that the intent recognition result is a website construction intent, having the large model extract key information from the user request information; Obtaining the construction requirement information based on the key information.

3. The method according to claim 2, wherein The step of obtaining the construction requirement information based on the key information includes: In response to determining that the key information does not meet the preset requirements, having the large model generate at least one interaction question based on the key information; Using the at least one interaction question to interact with the user, and obtaining the construction requirement information based on the reply information input by the user for each interaction question.

4. The method according to claim 2, further comprising In response to determining that the intent recognition result is not a website construction intent, having the large model generate a clarification question; Using the clarification question to interact with the user, and re-obtaining the user request information based on the reply information input by the user for the clarification question.

5. The method according to claim 1, wherein The step of inputting the set of front-end planning information into the large model to obtain a set of domain-specific language (DSL) description information includes: For each front-end planning information in the set of front-end planning information, having the large model retrieve a page template in a preset page template library based on the front-end planning information; In response to retrieving a page template corresponding to the front-end planning information, having the large model generate page text and / or page pictures corresponding to the page template; Obtaining DSL description information corresponding to the front-end planning information based on the page template and the page text and / or page pictures corresponding to the page template, and obtaining the set of DSL description information based on the DSL description information corresponding to each front-end planning information.

6. The method according to claim 1, wherein, The step of inputting the set of front-end planning information into the large model to obtain a set of domain-specific language (DSL) description information includes: For each front-end planning information in the set of front-end planning information, having the large model retrieve a page template in a preset page template library based on the front-end planning information; In response to not retrieving a page template corresponding to the front-end planning information, having the large model retrieve a block template in a preset block template library based on the front-end planning information; In response to retrieving a block template corresponding to the front-end planning information, having the large model generate page text and / or page pictures corresponding to the block template; Obtain the DSL description information corresponding to the front-end planning information based on the block template, as well as the page text and / or page images corresponding to the block template, and obtain the DSL description information set according to the DSL description information corresponding to each front-end planning information.

7. The method according to claim 6 further includes For each front-end planning information in the front-end planning information set, in response to not retrieving a block template corresponding to the front-end planning information, the large model generates source code corresponding to the front-end planning information according to the front-end planning information; Perform page rendering according to the generated source code, and obtain the website page corresponding to the front-end planning information according to the rendering result.

8. The method according to claim 1 further includes After obtaining the DSL description information set, display each DSL description information to the user; In response to receiving a modification instruction from the user, modify the corresponding DSL description information.

9. The method according to claim 1, wherein The performing page rendering according to the DSL description information set and obtaining the target website corresponding to the construction requirement information includes: Obtain a back-end planning information set, where the back-end planning information set includes back-end planning information corresponding to each website page; Perform page rendering according to the DSL description information set to obtain the rendering result of each website page; Obtain the back-end interface information of each website page according to the back-end planning information set; Obtain the target website corresponding to the construction requirement information according to the rendering result of each website page and the back-end interface information.

10. A website construction device based on a large model, including: A processing unit for inputting user request information into the large model to obtain construction requirement information; A planning unit for inputting the construction requirement information into the large model to obtain a front-end planning information set, where the front-end planning information set includes front-end planning information corresponding to each website page; A first generation unit for inputting the front-end planning information set into the large model to obtain a domain-specific language (DSL) description information set, where the DSL description information set includes DSL description information corresponding to each website page; A construction unit for performing page rendering according to the DSL description information set and obtaining the target website corresponding to the construction requirement information according to the rendering result.

11. An electronic device, including: At least one processor; And A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method according to any one of claims 1-9.

12. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to execute the method according to any one of claims 1-9.

13. A computer program product, including a computer program that, when executed by a processor, implements the method according to any one of claims 1-9.

Images