An information delivery method and device, a storage medium, an electronic device, and a product

Abstract

This specification provides an information delivery method, apparatus, storage medium, electronic device, and product. The method includes: acquiring target generation requirements for the information to be delivered; identifying target material information matching the target generation requirements in a pre-built material information library, and generating target prompts based on the target material information and the target generation requirements; inputting the target prompts into an information delivery system based on an artificial intelligence model, so that the information delivery system generates a universal presentation code that conforms to the target generation requirements and target operation instructions for the universal presentation code; and invoking an automated delivery tool through the target operation instructions, so that the automated delivery tool renders and delivers the information to be delivered on different information delivery platforms according to the universal presentation code.

Description

Technical Field

[0001] This specification relates to one or more embodiments in the field of computer technology, and more particularly to an information delivery method, apparatus, storage medium, electronic device and product. Background Technology

[0002] With the rapid development of internet technology, information dissemination methods have become increasingly diversified. Social media, advertising platforms, knowledge-sharing communities, and other information delivery platforms have become core carriers for users to obtain information and for businesses to conduct operations. To improve the efficiency and accuracy of information dissemination, information delivery technology is widely used in scenarios such as content distribution in social media operations, precise targeting in advertising, and resource delivery in knowledge management.

[0003] However, existing information delivery methods rely heavily on manual operation. Whether it is the process of selecting and planning the topics and creating the text and graphics content for the recommended information, or the process of publishing the recommended information on different information delivery platforms, all of these require manual operation. This method not only consumes a lot of manpower and has low overall efficiency, but also makes it difficult to guarantee the quality of the recommended information due to the uncertainty of human operation experience. Summary of the Invention

[0004] In view of the above, one or more embodiments of this specification provide the following technical solutions: According to a first aspect of one or more embodiments of this specification, an information delivery method is proposed, comprising: Obtain the target generation requirements for the information to be delivered; Target material information that matches the target generation requirements is identified from a pre-built material information library, and target prompt words are generated based on the target material information and the target generation requirements; The target prompt words are input into an information delivery system based on an artificial intelligence model, so that the information delivery system generates a general presentation code that meets the target generation requirements and target operation instructions for the general presentation code; wherein, the general presentation code is generated based on the target material information and meets the target generation requirements, and the general presentation code is used to present the graphic and text content of the information to be delivered on different information delivery platforms. The target operation command invokes the automated delivery tool, which then renders and delivers the information to be delivered on different information delivery platforms based on the general presentation code.

[0005] According to a second aspect of one or more embodiments of this specification, an information delivery device is provided, comprising: The acquisition unit is used to acquire the target generation requirements for the information to be delivered. The matching unit is used to determine the target material information that matches the target generation requirement in the pre-built material information library, and generate target prompt words based on the target material information and the target generation requirement; The output unit is used to input the target prompt words into an information delivery system based on an artificial intelligence model, so that the information delivery system generates a general presentation code that meets the target generation requirements and a target operation instruction for the general presentation code; wherein, the general presentation code is generated based on the target material information and meets the target generation requirements, and the general presentation code is used to present the graphic content of the information to be delivered on different information delivery platforms; The delivery unit is used to invoke the automated delivery tool through the target operation instruction, so that the automated delivery tool renders and delivers the information to be delivered on the different information delivery platforms according to the general presentation code.

[0006] According to a third aspect of one or more embodiments of this specification, an electronic device is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor implements the steps of the method described above by executing the executable instructions.

[0007] According to a fourth aspect of one or more embodiments of this specification, a computer-readable storage medium is provided that stores computer instructions thereon, which, when executed by a processor, implement the steps of the method described above.

[0008] According to a fifth aspect of one or more embodiments of this specification, a computer program product is provided, including a computer program / instructions that, when executed by a processor, implement the steps of the method described above.

[0009] As can be seen from the above embodiments, this specification matches target materials from a material library and generates prompt words. It then uses an information delivery system to produce universal presentation code and operation instructions for multiple information delivery platforms, ultimately completing the automated rendering and delivery of the information to be delivered on the target platform. This process significantly reduces manual intervention, lowering labor costs while improving overall efficiency. Furthermore, relying on the generation capabilities of artificial intelligence models, it overcomes the limitations of manual operation, effectively guaranteeing the quality of recommended information. Moreover, based on universal presentation code adapted to different platforms, it eliminates the tedious process of repeated manual publishing across multiple platforms, achieving efficient cross-platform presentation and comprehensively solving the problems of existing methods that rely on manual labor, are costly, inefficient, and have difficulty guaranteeing quality. Attached Figure Description

[0010] Figure 1 This is a schematic diagram of the architecture of a service system for information delivery provided in an exemplary embodiment; Figure 2This is a flowchart illustrating an exemplary embodiment of an information delivery method. Figure 3 This is a schematic diagram illustrating the parsing process of target generation requirements provided in an exemplary embodiment; Figure 4 This is a schematic diagram illustrating a process for determining target material information, provided in an exemplary embodiment. Figure 5 This is a schematic diagram of an information generation process provided in an exemplary embodiment; Figure 6 This is a schematic diagram of a login process based on an automated delivery tool, provided as an exemplary embodiment. Figure 7 This is a schematic diagram of an information delivery system provided in an exemplary embodiment; Figure 8 This is a schematic structural diagram of a device provided in an exemplary embodiment; Figure 9 This is a block diagram of an information delivery device provided in an exemplary embodiment. Detailed Implementation

[0011] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0012] In various scenarios such as social media operations, advertising, and knowledge management, the demand for diverse information delivery methods, including dynamic cards, graphic and textual information, and video trailers, is increasing daily. These information formats, with their intuitive, vivid, and easy-to-disseminate characteristics, have become powerful tools for attracting user attention and conveying key information.

[0013] However, the current information delivery process has significant shortcomings, with the entire process heavily reliant on manual operation: topic selection and planning require identifying trending topics and catering to the target audience, which is time-consuming and labor-intensive; graphic and text creation requires conceiving copy, matching images, and designing layouts, which demands a high level of professional skills; cross-platform delivery is extremely inefficient due to the different formats, sizes, and rules of each platform; and manually generated information is difficult to be compatible with the characteristics of different platforms, resulting in poor display effects, failure to fully realize the dissemination value, waste of delivery resources, and poor delivery results.

[0014] Based on this, this specification provides an information delivery method that, based on the generation capabilities of an artificial intelligence model, generates universal presentation code and operation instructions for multiple information delivery platforms according to the target generation requirements. This enables the automated rendering and delivery of recommended information on different information recommendation platforms, effectively improving the quality and efficiency of information delivery.

[0015] Figure 1 This is a schematic diagram of the architecture of a service system for information delivery, provided as an exemplary embodiment. Figure 1 As shown, the system may include a server 11, a network 12, and several electronic devices, such as a PC (Personal Computer) 13, a mobile phone 14, etc.

[0016] Server 11 can be a physical server containing an independent host, or it can be a virtual server hosted in a host cluster. During operation, server 11 can run server-side programs for a certain application to implement the relevant functions of that application. For example, when server 11 runs an information delivery service program, it can function as a corresponding information delivery service platform.

[0017] PC13 and mobile phone14 are just some of the types of electronic devices that users can use. In reality, users can obviously also use electronic devices such as tablets, laptops, PDAs (Personal Digital Assistants), wearable devices (such as smart glasses, smartwatches, etc.), etc., and one or more embodiments in this specification do not limit this. During operation, the electronic device can run a client-side program of an application to achieve the relevant functions of that application. For example, when the electronic device runs an information delivery service program, it can act as a client for that information delivery service. The aforementioned information delivery service client application can be launched and run on the electronic device. This client-side program can be a native application installed on the electronic device, or it can be a mini-program, quick app, or other similar form. Of course, when using web technologies such as HTML5 or similar, the relevant functions can be achieved through a page displayed by a browser. This browser can be a standalone browser application or a browser module embedded in some applications.

[0018] As for the network 12 that enables interaction between electronic devices such as PC13 and mobile phone 14 and server 11, communication can be achieved using either wired or wireless networks, depending on the communication methods supported by the respective electronic devices. This specification does not impose any restrictions on this. For example, PC13 can support both wired and wireless communication, so it can use either wired or wireless networks as needed. Mobile phone 14 typically only supports wireless communication, so it can use a wireless network for communication.

[0019] Based on a service system architecture, this specification provides an information delivery method, such as... Figure 2 As shown.

[0020] Figure 2 This is a flowchart illustrating an exemplary embodiment of an information delivery method, including the following steps: S200: Obtain the target generation requirements for the information to be delivered.

[0021] In this specification, the executing entity used to perform the information delivery method can be a designated device such as a server. Of course, it can also be a terminal device such as a mobile phone, tablet computer, desktop computer, or laptop computer, or a client installed on these terminal devices. For ease of description, the following will use a server as the executing entity to illustrate the information delivery method provided in this specification.

[0022] In practical applications, users (such as operators delivering information) can input descriptive text for the information to be delivered. This descriptive text can be used to describe the requirements for the information to be delivered, including one or more of the following: design requirements for images (such as the size, color, and layout of dynamic cards), requirements for content text (such as the language style, keywords, and word count of the design copy), and requirements for topic text (such as the field or category of the topic or title). Typically, these requirements are presented in the form of unstructured information in the descriptive text. Therefore, the server can parse these descriptive texts to obtain the structured target generation requirements for the information to be delivered.

[0023] In this specification, the server can utilize the adversarial mechanism of the multi-analysis model and combine it with the evaluation model to select the optimal parsing result.

[0024] To facilitate understanding, this specification provides a schematic diagram illustrating the method for parsing target generation requirements, such as... Figure 3 As shown.

[0025] Figure 3 This is a schematic diagram of a target generation requirement parsing process provided in an exemplary embodiment, including the following steps: S300: Obtain the description text for the information to be delivered; S302: Input the description text into at least two parsing models; S304: The description text is parsed by each parsing model to obtain the candidate generation requirements output by each parsing model; wherein, the model parameters corresponding to different parsing models are different; S306: Based on the preset evaluation model, evaluate each candidate generation requirement and obtain the comprehensive performance score corresponding to each candidate generation requirement; S308: Based on the comprehensive performance score, determine the target generation requirement from among the candidate generation requirements.

[0026] The server can pre-start multiple instances of the parsing model. These instances can be different versions of the same model with different parameter configurations, or they can be multiple completely different models. Additionally, the server can preprocess the input descriptive text, such as removing special characters and automatically detecting the language type.

[0027] The processed descriptive text is then input into these parsing models. The final parsed structured target generation requirements can include: image design requirements C (i.e., image design requirements), content text requirements T (i.e., content text requirements), and topic text requirements S (i.e., topic text requirements). Therefore, for each parsing model, the corresponding output can be represented as: Where X represents the original descriptive text, Representation Model The corresponding model parameters, This represents the total number of analytical models.

[0028] Each analytical model can output corresponding candidate generation requirements based on a preset parameterized template, which can be represented as: {"card_requirements": { "quantity": "int", "style": ["minimalist", "vibrant", "professional"], "color_palette": "hex_codes[]"}, "copy_requirements": { "tone": ["formal", "casual", "persuasive"], "key_points": "string[]"}, "topic_requirements": {"target_audience": "string", "novelty_score": "float[0-1]"}} For each parsing model, the output instructions can be guaranteed by constructing a prompt word. The prompt word for any parsing model can be: Please strictly follow the parameterized template format when outputting content; Image design requirements should include [quantity, style requirements, color scheme]; The content text must include [tone and style, key information points]; The subject text requirements must include [target audience, novelty requirements]; ……".

[0029] The server can then input the candidate generation requirements of each parsing model's output into the evaluation model.

[0030] The evaluation model can determine the relevance score of each candidate generation request based on the semantic correlation between the candidate generation request and the description text. Based on the comprehensiveness of the candidate generation requirement in covering the content of the descriptive text, determine the completeness score corresponding to the candidate generation requirement. Based on the novelty of the candidate generation requirement, determine the innovation score corresponding to the candidate generation requirement. Among them, the degree of association is positively correlated with the relevance score, the degree of comprehensiveness is positively correlated with the completeness score, and the degree of novelty is positively correlated with the innovativeness score.

[0031] The evaluation model can perform a weighted sum based on relevance score, completeness score, innovation score, and the weights corresponding to each score to obtain the comprehensive performance score corresponding to the candidate generation requirement. This comprehensive performance score can be expressed as: The weights corresponding to each score can be set according to the actual situation. Preferably, the weights corresponding to each score can be: .

[0032] The server can then use the candidate generation requirement with the highest overall performance score as the target generation requirement and store it in the parameter database.

[0033] Furthermore, after information is delivered, the server can collect user interaction data on different types of information to be delivered (e.g., collecting interaction data weekly). This interaction data includes users' browsing and clicking patterns on historical recommended information (such as page views, click-through rates, and likes). Based on this interactive data, the server can then determine the preferred recommended information from among the various types of information to be delivered and define the sample generation requirements for generating this preferred recommended information. For example, the historical recommended information with the highest number of likes, page views, or click-through rates can be used as the preferred recommended information.

[0034] For each analytical model, the server can adjust the model parameters of the analytical model with the optimization objective of minimizing the deviation between the sample generation requirement and the target generation requirement of the analytical model for the same descriptive information output. At the same time, the server can determine the target performance score corresponding to the sample generation requirement based on the interactive data, and adjust the model parameters of the evaluation model with the optimization objective of minimizing the deviation between the target performance score and the comprehensive performance score output by the evaluation model for the sample generation requirement.

[0035] In practical applications, there are several ways to determine the target performance score. For example, it can be determined based on the specific values ​​of page views, click-through rate (CTR), and likes. Alternatively, these three metrics can be weighted and summed, such as assigning a weight of 40% to CTR, 30% to page views, and 30% to likes before calculating the sum. Of course, it can also be combined with the number of shares and comments, with each share and comment receiving a weight of 20% in the weighted calculation, together with page views, CTR, and likes, to form the target performance score. In addition, for scenarios with conversion requirements, adding the conversion rate metric and assigning it a corresponding weight can also be used to determine the target performance score.

[0036] In addition, the server can determine the first loss value based on the deviation between the sample generation requirements and the target generation requirements of the analytical model for the same descriptive information, and determine the second loss value based on the deviation between the target performance score and the comprehensive performance score output by the evaluation model for the sample generation requirements. With minimizing the first and second loss values ​​as the optimization objective, the server can adjust the model parameters of the analytical model and the evaluation model respectively, thereby obtaining a more accurate analytical model and evaluation model.

[0037] S202: Determine the target material information that matches the target generation requirement from the pre-built material information library, and generate target prompt words based on the target material information and the target generation requirement.

[0038] Once the target generation requirements are determined, the server can identify matching target material information from a pre-built material information library. This target material information includes the theme text matching the target generation requirements and the content text matching the theme text. Of course, in practical applications, it may also include other target material information such as dynamic effect templates, icon elements, font libraries, background textures, data visualization components, and images; this manual does not specifically limit this.

[0039] Each material information can be stored in the material information database in the form of vectors. In determining the target material information, the server first converts the structured target generation requirements into corresponding semantic vectors (e.g., extracting requirement features through a natural language processing model to generate fixed-dimensional vectors). Then, it calculates the similarity between the requirement vector and each material vector in the material information database (e.g., cosine similarity, Euclidean distance, etc.). Materials with similarity higher than a preset threshold are then selected as candidate materials. Finally, a second round of screening is conducted using preset priority conditions (e.g., "prioritize materials with high reuse rates in the last 3 months" and "exclude materials already used in similar campaigns") to determine the target material information that meets the requirements. Based on the target material information and the target generation requirements, target prompts are then generated. The aforementioned semantic vectors can be 512-dimensional semantic vectors encoded using the bge-small model.

[0040] Because information delivery scenarios have high requirements for the timeliness of materials (for example, e-commerce promotional materials need to match the current event node, hot topic delivery needs to be combined with recent social hot topics, and industry policy materials need to be synchronized with the latest policy content), relying solely on semantic matching degree may filter out materials that are outdated, which will lead to a decrease in delivery effectiveness. Therefore, when determining target material information, in addition to referring to the semantic matching degree (i.e., vector similarity) between the target generation requirements and material information, the generation time of the candidate material information can also be considered.

[0041] For each candidate material in the material information database, the server can determine the semantic matching degree between the candidate material and the target generation requirement. Then, based on the semantic matching degree and the generation time of the candidate material, the server determines the recommendation score of the candidate material. Each candidate material can be set with an ISO format timestamp. The closer the generation time is to the current time, the higher the recommendation score. The older the generation time is than the current time, the lower the recommendation score.

[0042] The server can then determine the target material information from among the candidate material information based on the recommendation score, and then generate target prompt words based on the target material information and target generation requirements.

[0043] For example, the server can select one or more candidate materials with a recommendation score greater than a preset score as target materials; or, for another example, the server can sort the candidate materials in descending order of recommendation score and select the candidate materials ranked before the preset position as target materials.

[0044] To facilitate understanding, this specification provides a schematic diagram illustrating the process for determining target material information, such as... Figure 4 As shown.

[0045] The server can input the target generation requirement into a time-weighted retrieval system via a corresponding input interface. The time-weighted retrieval system then weights the semantic matching degree between each candidate material and the target generation requirement based on the generation time of each candidate material, thus obtaining a recommendation score for each candidate material. Based on the recommendation score, the target material is identified from the candidate material information, and target prompts are generated based on the target material and the target generation requirement. The server can then input the target prompts into an information delivery system based on an artificial intelligence model and output the content generated by the information delivery system through a corresponding output interface (the content generated by the information delivery system will be described in detail below and will not be elaborated upon here).

[0046] Specifically, the time-weighted search engine above sets the search quantity to the final target material quantity. This is three times the amount of material available, to avoid insufficient selection due to subsequent filtering. An example of the code logic is as follows: #Semantic similarity retrieval: Obtain three times the target number of candidate materials; candidates=vec_db.kNN(query_vector=demand_vector,k=top_n×3).

[0047] For time weighting, an exponential decay model can be used to calculate the time weight of creative materials. The core logic is that "the closer the creative material's creation time is to the current time, the higher the weight; as time goes on, the weight decreases exponentially." The "half-life" parameter flexibly controls the rate of timeliness decay (the default is 30 days, which can be adjusted according to industry characteristics; for example, for trending campaigns, half-life can be set to 7 days, and for long-term brand creative materials, it can be set to 90 days). The code implementation is as follows: deftime_weight(create_time, half_life=30): #Calculate the number of days between the time the material was generated and the current time (create_time is the ISO format timestamp of the material); delta_days=(datetime.now()-datetime.fromisoformat(create_time)).days; #Exponential decay formula: The more days the interval, the smaller the time weight, and the minimum value approaches 0; return2(-delta_days / half_life).

[0048] The final recommendation score (final_score) of candidate materials is calculated by combining "semantic matching score" and "time weight". A weighting coefficient α (default 0.7) balances the priority of "semantic relevance" and "timeliness"—a higher α value indicates a higher priority for matching semantically relevant materials; a lower α value indicates a greater focus on recently generated materials. The formula and code logic are as follows: # Hybrid search scoring formula: α is the weight coefficient of semantic matching degree, and (1-α) is the weight coefficient of time weight; final_score=α×semantic_score+(1-α)×time_weight; #Code implementation: Sort in descending order by final recommendation score (α defaults to 0.7); candidates.sort(key=lambdax:0.7×x["semantic_score"]+0.3×time_weight(x["create_time"]),reverse=True).

[0049] After sorting, the server determines the target material information from the candidate materials according to preset rules. The final filtering logic code is as follows: #Select the top _n candidate materials as the target material information; target_materials=candidates[:top_n].

[0050] Once the target material information is determined, the server can generate target prompt words based on the "target generation requirements (S)" and "target material information (context_str)" and combine them with the large model decision prompt template, providing clear guidance for subsequent calls to the artificial intelligence model to generate content for delivery.

[0051] The template must clearly define three key elements: "task objective, input information, and output format," ensuring that the large model accurately understands the requirements and generates structured results. For example, the template for "generating alternative topics" is as follows: Based on the following requirements and raw materials, generate three alternative topics and indicate their recommendation level:\n Requirement: {S}\n; Raw material: {context_str}\n; Output format: 1. [Topic] (Recommendation percentage) \n Reasons...\n"

[0052] After the server fills the "Target Generation Requirements" and "Target Material Information (context_str)" in the above template, it calls the corresponding artificial intelligence model in the information delivery system in a specified format. An example of the call parameters is as follows: {"model":"xxx", "messages": [{ "role": "system", "content": "You are a professional content planning assistant"}, {"role": "user", "content": prompt_template.format(S, context_str)}]}.

[0053] Thus, the server can obtain the target prompt, which may include image design requirement C, content text requirement T, theme text requirement S, theme text, and content text.

[0054] It should be noted that the theme text in the target material information can be determined based on the theme text requirement S, and the content text can be directly determined from the multiple content texts under the theme text.

[0055] S204: Input the target prompt word into an information delivery system based on an artificial intelligence model, so that the information delivery system generates a general presentation code that meets the target generation requirements and a target operation instruction for the general presentation code; wherein, the general presentation code is generated based on the target material information and meets the target generation requirements, and the general presentation code is used to present the graphic content of the information to be delivered on different information delivery platforms.

[0056] In this specification, the information delivery system may include multiple artificial intelligence models (such as Large Language Modeling, LLM). These models may include a scheme planning model for generating planning proposals, an information generation model for generating presentation codes and target operation instructions, and a content review model for reviewing the generated content. For ease of understanding, this specification provides a schematic diagram of the information generation process, such as... Figure 5 As shown.

[0057] The aforementioned target prompts may include: planned prompts and generated prompts.

[0058] The following are examples of prompts for inputting a solution planning model: You are a content planning expert specializing in the sharing field, skilled in designing highly engaging expression frameworks and conveying ideas.

[0059] ##Responsibilities: Based on the specific theme and other requirements provided by the user, create an attractive card format for the online community. Ensure the content is engaging, the entire card set is coherent and complete, and the content is truthful and accurate.

[0060] ## Basic Requirements: 1. The creative concept must align with the sharing theme, the overall flow of the card should be smooth, the content should be engaging, and it should conform to the expression logic of the chosen topic.

[0061] 2. Produce the most reliable expression framework. Based on the theme and content, plan the sub-theme, content, and expression requirements of each card to ensure that the content of each card is complete and full.

[0062] 3. The content displayed on each card cannot be just directional hints; all the necessary content for that card must be provided. If the presentation format is a table, all the data required for the table must be provided completely. The first card is the cover card by default, and its content can be: "Show only the main title".

[0063] 4. Card design requirements need to be constrained from the following dimensions: Color scheme guidelines (accurate to color codes) ensure that the color system of all cards is consistent, for example, primary color: #XXX, secondary color: #XXX, #XXX; Layout requirements: For example, paragraphs need to be separated by dividing lines to ensure visual separation; Decorative elements (borders / backgrounds / dividers), etc., to ensure aesthetics and harmony. For example, adding semi-transparent XXemojis to the lower right corner of the card can enhance its sophistication. Information display methods: such as tables, paragraph text, etc., determined according to the content. For example, "use tables to display parallel data in the content" or "use text, with important information highlighted".

[0064] Note that the first card is the cover card by default. The requirement for the default cover card is preceded by the phrase: "Based on the requirements of the fineness and design of commercial posters:", and the cover card is given the design height constraints of commercial posters.

[0065] If there are images that can be used, all cards must not reuse images. If one image is used, it should not be reused on other cards. The content of the image must be used by the current card or serve an explanatory purpose. Each card should not use more than one image to avoid making the card too long.

[0066] ## Output: Please strictly follow the following JSON template to organize your solution: '''json [{"theme": "Theme of the first card"; "content": "The content that the first card should display"; "images": "The first card contains image links and descriptions. Images must have descriptions, such as (image link: the content of this image is XXXXX). You can only select from the available images below. If there are no images available, leave it blank." "require": "The first card design requirements must ensure consistency across several dimensions and constraints."}; {"theme": "The theme of the second card";} "content": "The content that the second card needs to display"; "images": "The image links and descriptions required for the second card. Images must have descriptions and can only be selected from the available images below; otherwise, leave it blank." "require": "The design requirements for the second card must ensure the consistency of the constraints and design across several dimensions."

[0067] / / ....]''' Based on the above requirements, and using the following themes and the latest available materials, please provide suggestions for creating card content that meets the requirements: ## Subject: $(theme); ## Latest content available: $(content); ## Requirement: $(require); Please strictly follow the above structure in the JSON file to return the design proposal. No other description is needed; just return this JSON file.

[0068] The server can input the aforementioned planning prompts into the scheme planning model, enabling the model to determine a planning scheme for generating the information to be deployed. This planning scheme is generated based on the target material information and the target generation requirements. The planning scheme serves to build a mapping bridge between "target generation requirements – target material information – presentation code," such as clarifying the development basis and constraints of the presentation code for the information to be deployed. Specifically, this includes: code development specifications (e.g., front-end framework selection for dynamic cards, code implementation standards for interactive logic), material calling rules (e.g., path referencing format for image / icon resources, code import method for font libraries), code definition of visual parameters (e.g., hexadecimal encoding of color values, responsive code formulas for size adaptation), and content rendering logic (e.g., code output structure for text paragraphs, dynamic replacement variable settings for theme text). This provides precise requirement transformation and technical guidance for generating directly executable presentation code for the information to be deployed.

[0069] Furthermore, after determining the planning scheme, the server can input it into the information generation model to match the information generation model with the planning scheme using universal presentation code and target operation instructions. This universal presentation code can be HTML / SVG encoding that conforms to the Model Context Protocol (MCP) corresponding delivery standard. This universal presentation code is generated based on the target material information and meets the target generation requirements, and is used to present the graphic and textual content of the information to be delivered on different information delivery platforms.

[0070] For example, the planning prompts for the input solution planning model can be: Based on the content to be expressed in the card below, please create a card that accurately matches the theme and content. Use HTML code to draw a beautiful and harmonious static card. The card design should be so sophisticated that it can be directly printed as a small card for posting.

[0071] ## Basic generation requirements: Please generate a complete HTML code snippet for me. The code should be concise and efficient, without excessive code stacking, and should be encoded in UTF-8.

[0072] ##Card Requirements: 1. Output an HTML code and draw a content card. The color scheme and style should match the theme, and the content should not exceed the topic and requirements below.

[0073] 2. The default aspect ratio of the card is 450px:600px. It cannot be shorter than this ratio. It must not overflow or be truncated (the appearance of a slider also counts as truncation). If the content is too long, the card must be stretched to fit the content and prevent the content from being truncated. The layout should be distributed with appropriate white space. The overall content should be full and balanced. Do not make it too crowded or have large blank areas.

[0074] 3. The cards must be aesthetically pleasing and exquisite, full of design sense, and the content must be complete and accurate. Elements such as date and time must be provided strictly according to the content. The year, month, and day must not be tampered with. The current time is $(system_CURR_DATE), ensuring accuracy.

[0075] 4. Ensure layout stability, full display of content without obstruction or truncation, and do not use additional JS packages.

[0076] ## Additional requirements that must be met: ## Expressing the theme and text content: Cards cannot contain content that is not part of the theme or the displayed content.

[0077] Card theme: $(theme); Card display content: $(content); Replenish: ##Images that can be used: You may insert images as needed into the card. If the image is not in the list below, it should not be used. If the list is empty, please ignore this step. Cards can reference images and their descriptions: $(image); Please only return the card codes you created, excluding other useless descriptions.

[0078] In addition, when the planning scheme contains multiple dynamic cards that need to be generated, the information generation model can process the generation requirements and material information corresponding to each dynamic card in parallel, thereby generating the presentation code of multiple dynamic cards simultaneously, thus improving the efficiency of information generation.

[0079] Furthermore, the server can input the general rendering code into the content review model, so that the content review model can review the information content embedded in the general rendering code based on a preset abnormal word library, so as to render and deliver the information to be delivered while ensuring that it does not contain abnormal words.

[0080] S206: The automated delivery tool is invoked through the target operation instruction, so that the automated delivery tool renders and delivers the information to be delivered on the different information delivery platforms according to the general presentation code.

[0081] After generating the presentation code, the information generation model can generate corresponding operation instructions based on the presentation code. The content of the operation instructions can include two parts: one is the instruction to render the information to be delivered using the presentation code, and the other is the instruction to deliver the rendered information to various information recommendation platforms.

[0082] In practical applications, information delivery systems can be equipped with corresponding automated delivery tools (such as Playwright). The information generation model can first generate initial operation instructions for general presentation codes, and then, based on the MCP set in the information delivery system, convert the instruction format of the initial operation instructions into an instruction format that the automated delivery tool can recognize, thereby obtaining the target operation instructions.

[0083] The information delivery system can invoke automated delivery tools through target operation commands, enabling these tools to call the rendering engines corresponding to each information delivery platform. Based on the common rendering code, the system renders and automatically delivers the information to be delivered on different information delivery platforms.

[0084] Furthermore, before delivering the information, the information recommendation system can use the aforementioned automated delivery tools to obtain the target document object model (DOM) of the login page corresponding to the information delivery platform. Then, based on the login page structure features and login operation logic reflected by the target DOM, the system executes the login operation for the information delivery platform.

[0085] To facilitate understanding, this manual provides a schematic diagram of a login process based on an automated deployment tool, such as... Figure 6 As shown.

[0086] The process of logging into the information delivery platform through automated delivery tools may include the following steps: S600: The automated delivery tool can load cookies to perform login, initiate the login process, and attempt to complete the login operation using existing cookie information.

[0087] During this process, the login result of loading cookies in S600 can be verified. If it is determined to be "yes", it means that you are already logged in. Therefore, you can directly jump to S612 to perform the cookie saving operation. If it is determined to be "no", proceed to the next step.

[0088] S602: Clean up invalid cookies. Clear invalid cookie data that may interfere with the normal login process and prepare for the new login step.

[0089] S604: Locate the mobile phone input box. Find the interactive element used to enter the mobile phone number on the page and determine the input location.

[0090] S606: Enter your mobile phone number. In the input box located in S604, enter the mobile phone number you need to log in to.

[0091] "Whether to provide a verification code": This decision determines the subsequent operation path based on the login scenario requirements. If the decision is "yes", proceed to S608a; if the decision is "no", proceed to S608b.

[0092] S608a: Enter the verification code. In the corresponding verification code input area, fill in the obtained valid verification code.

[0093] S608b: Sends a verification code, triggering the verification code sending mechanism, and sends an SMS / message for verification to the entered mobile phone number.

[0094] S610: Verification code backfill environment variables. The verification code information obtained in S608b is backfilled into the preset environment variable location to prepare for login verification.

[0095] The system verifies the input information (including mobile phone number, verification code, etc.) and the verification process. If the result is "yes", proceed to step S612; if the result is "no", return to the appropriate step (such as re-entering information, re-sending verification code, etc., the return path is not detailed in the figure) and try to log in again.

[0096] S612: Save cookies. Store valid cookies generated or updated after successful login for reuse in subsequent operations.

[0097] S614: Login complete. Login process finished. Successfully entered the information delivery platform.

[0098] It should be further noted that the processes S200 to S206 above can be completed through an information delivery system. For ease of understanding, this specification provides a schematic diagram of such an information delivery system, as shown below. Figure 7 As shown.

[0099] The information delivery system includes: a demand extraction module 700 (corresponding to step S200), a material information extraction module 702 (corresponding to step S202), an information generation module 704 (corresponding to step S704), and an automated delivery module 706 (corresponding to step S206).

[0100] The requirement extraction module 700 is equipped with multiple parsing and evaluation models to extract target generation requirements from the original description text.

[0101] The material information matching module 702 is equipped with a RAG engine and a material information database, so that the RAG engine can retrieve the target material information from the material information database.

[0102] The information generation module 704 includes a scheme planning model, an information generation model, and a content review model. The scheme planning model is used to generate a planning scheme based on the planning prompts. The information generation model is used to generate general presentation codes and target operation instructions based on the generation prompts and the planning scheme. The content review model is used to review the content of the general presentation codes.

[0103] The automated delivery module 706 is used to convert the instruction format based on the built-in MCP protocol and call the automated delivery tool to render and deliver the graphic content of the information to be delivered.

[0104] In addition, the information delivery system can collect interactive data on the delivered information and use the returned data to train analytical and evaluation models.

[0105] It should be added that, in addition to the text and image content corresponding to the dynamic card, the information to be delivered in this solution can also include video images, which can be rendered using the aforementioned general presentation code.

[0106] As can be seen from the above methods, the automated information delivery system of this application has significant benefits in multiple dimensions: Firstly, it significantly reduces labor costs and operational barriers by using natural language to break down structured parameters, eliminating the need for manual requirement analysis. It automatically acquires the latest industry materials using RAG time decay retrieval, reducing the time spent on manual topic selection. The automated delivery tool Playwright achieves automated delivery and cross-platform compatibility, eliminating the need for manual batch operations or adaptation to different platforms. It also automatically adapts to differences in the DOM across different platforms, avoiding manual code modifications and form adaptation.

[0107] Secondly, the parallel generation of text and visual content significantly improves the efficiency of creation and distribution, and promotes creation simultaneously to shorten the cycle. The dynamic coding generation function outputs structured style descriptions and automatically generates SVG / HTML code during the card planning stage, eliminating the need for manual coding.

[0108] Third, it achieves efficient cross-platform adaptation. The same content can be automatically adapted to different information recommendation platforms without manual code modification. The generated universal presentation code can be adapted to different platform rendering engines, meeting the needs of multi-platform delivery and improving the automation level and practicality of the entire content creation and delivery process.

[0109] Figure 8 This is a schematic structural diagram of a device provided in an exemplary embodiment. For example... Figure 8As shown, device 800 mainly consists of a communication interface 802, a user interface 804, a processor 806, and a data storage 808. These components are interconnected and communicate with each other via a system bus, network, or other connection mechanism 810. The communication interface 802 enables device 800 to communicate with other devices, access networks, and transmission networks via analog or digital modulation. For example, the communication interface 802 may include a chipset and antenna for wireless communication with a radio access network or access point. Furthermore, the communication interface 802 can be a wired interface such as Ethernet, Token Ring, or a USB port, or a wireless interface such as Wi-Fi, Bluetooth, Global Positioning System (GPS), or a wide-area wireless interface (e.g., WiMAX or LTE). Of course, the communication interface 802 can also support other forms of physical layer interfaces and standard or proprietary communication protocols. The communication interface 802 may also include multiple physical communication interfaces, such as Wi-Fi, Bluetooth, and wide-area wireless interfaces.

[0110] User interface 804 includes receiving user input and providing output to the user. Therefore, user interface 804 may include input components such as a keypad, keyboard, touch-sensitive or presence-sensitive panel, computer mouse, trackball, joystick, microphone, still camera, and video camera, and output components such as a display screen (which may be combined with a touch-sensitive panel), CRT, LCD, LED, display using DLP technology, printer, and other similar devices known or developed in the future. User interface 804 may also generate auditory output via speakers, speaker jacks, audio output ports, audio output devices, headphones, and other similar devices known or developed in the future. In some embodiments, user interface 804 may include software, circuitry, or other forms of logic capable of transmitting and receiving data from external user input / output devices. Additionally or alternatively, device 800 may support remote access from other devices via communication interface 802 or another physical interface (not shown). User interface 804 may be configured to receive user input, the position and movement of which may be indicated by indicators or cursors described herein. User interface 804 may also be configured as a display device for rendering or displaying text fragments.

[0111] The processor 806 may contain one or more general-purpose processors and / or special-purpose processors.

[0112] Data storage 808 may include one or more volatile and / or non-volatile storage components and may be integrated wholly or partially with processor 806. Data storage 808 may include removable and non-removable components.

[0113] Processor 806 is capable of executing program instructions 818 (e.g., compiled or uncompiled program logic and / or machine code) stored in data storage 808 to perform the various functions described herein. Data storage 808 may contain a non-transitory computer-readable medium on which program instructions are stored, which, when executed by device 800, enable device 800 to perform any methods, processes, or functions disclosed in this specification and / or the accompanying drawings. Execution of program instructions 818 by processor 806 may result in processor 806 using data 812.

[0114] For example, program instructions 818 may include an operating system 822 (e.g., an operating system kernel, device drivers, and / or other modules) installed on device 800 and one or more application programs 820 (e.g., a browser, social application, or game application). Similarly, data 812 may include operating system data 816 and application data 814. Operating system data 816 is primarily accessible to the operating system 822, while application data 814 is primarily accessible to one or more application programs 820. Application data 814 may reside in a file system visible or hidden from the user of device 800.

[0115] Application 820 can communicate with operating system 822 through one or more application programming interfaces (APIs). These APIs help application 820 read and / or write application data 814, transmit or receive information via communication interface 802, receive or display information on user interface 804, etc.

[0116] In some terminology, application 820 may be simply referred to as "app". Furthermore, application 820 can be downloaded to device 800 through one or more online app stores or app markets. However, applications can also be installed on device 800 in other ways, such as through a web browser or a physical interface on device 800 (e.g., a USB port).

[0117] Please refer to Figure 9 Information delivery devices can be applied to, for example Figure 8 The device shown implements the technical solution described in this specification. The information delivery device may include: The acquisition unit 900 is used to acquire the target generation requirements for the information to be delivered. The matching unit 902 is used to determine the target material information that matches the target generation requirement in the pre-built material information library, and generate target prompt words based on the target material information and the target generation requirement; Output unit 904 is used to input the target prompt word into an information delivery system based on an artificial intelligence model, so that the information delivery system generates a general presentation code that meets the target generation requirements and a target operation instruction for the general presentation code; wherein, the general presentation code is generated based on the target material information and meets the target generation requirements, and the general presentation code is used to present the graphic content of the information to be delivered on different information delivery platforms; The delivery unit 906 is used to invoke the automated delivery tool through the target operation instruction, so that the automated delivery tool renders and delivers the information to be delivered on the different information delivery platforms according to the general presentation code.

[0118] Optionally, the acquisition unit 900 is specifically used to: acquire descriptive text for the information to be delivered; input the descriptive text into at least two parsing models, so that the descriptive text is parsed by each parsing model to obtain candidate generation requirements output by each parsing model; wherein the model parameters corresponding to different parsing models are different; evaluate each candidate generation requirement based on a preset evaluation model to obtain a comprehensive performance score corresponding to each candidate generation requirement; and determine the target generation requirement from among the candidate generation requirements based on the comprehensive performance score.

[0119] Optionally, the acquisition unit 900 is specifically configured to, for each candidate generation requirement, determine a relevance score corresponding to the candidate generation requirement based on the semantic correlation between the candidate generation requirement and the description text; determine a completeness score corresponding to the candidate generation requirement based on the comprehensiveness of the content covered by the candidate generation requirement in the description text; determine an innovation score corresponding to the candidate generation requirement based on the novelty of the candidate generation requirement; and perform a weighted summation based on the relevance score, the completeness score, the innovation score, and the weights corresponding to each score to obtain a comprehensive performance score corresponding to the candidate generation requirement.

[0120] Optionally, the acquisition unit 900 is further configured to: collect user interaction data on different information to be delivered; wherein the interaction data is used to characterize the user's browsing and clicking behavior on historical recommended information; determine preferred recommended information from each information to be delivered based on the interaction data, and determine the sample generation requirements for generating the preferred recommended information; adjust the model parameters of each analytical model with the optimization objective of minimizing the deviation between the sample generation requirements and the target generation requirements output by the analytical model; and determine the target performance score corresponding to the sample generation requirements based on the interaction data, and adjust the model parameters of the evaluation model with the optimization objective of minimizing the deviation between the target performance score and the comprehensive performance score output by the evaluation model.

[0121] Optionally, the matching unit 902 is specifically used to: determine the semantic matching degree between each candidate material information in the material information database and the target generation requirement; determine the recommendation score corresponding to the candidate material information based on the semantic matching degree and the generation time corresponding to the candidate material information; wherein, the closer the generation time is to the current time, the greater the recommendation score; and determine the target material information from each candidate material information according to the recommendation score.

[0122] Optionally, the target generation requirements include at least one of the following: design requirements for images, requirements for content text, and requirements for theme text, and the target material information includes: theme text that matches the target generation requirements and content text that matches the theme text.

[0123] The optional information delivery system at the end of the year includes: a scheme planning model and an information generation model; The target prompts include: planned prompts and generated prompts; The output unit 904 is specifically used for: inputting the planning prompts into the scheme planning model, so that the scheme planning model determines a planning scheme for generating the information to be deployed; wherein the planning scheme is generated based on the target material information and the target generation requirements; and inputting the generation prompts and the planning scheme into the information generation model, so that the information generation model generates a general presentation code that matches the planning scheme.

[0124] Optionally, the information delivery system includes: a content review model; Before invoking the automated delivery tool through the target operation instruction, the output unit 904 is further configured to: input the general presentation code into the content review model, so that the content review model reviews the information content embedded in the general presentation code.

[0125] Optionally, the output unit 904 is specifically used to: generate an initial operation instruction for the general presentation code; and convert the instruction format of the initial operation instruction into an instruction format recognizable by the automated delivery tool based on the Model Context Protocol (MCP) set in the information delivery system to obtain the target operation instruction.

[0126] Optionally, before rendering and delivering the information to be delivered on the different information delivery platforms, the delivery unit 906 is further configured to: obtain the target document object model (DOM) of the login page corresponding to the information delivery platform through the automated delivery tool; and execute a login operation for the information delivery platform based on the login page structure features and login operation logic reflected by the target DOM.

[0127] For ease of description, the above devices are described by dividing them into various modules or units based on their functions. Of course, when implementing one or more of these specifications, the functions of each module or unit can be implemented in the same or different software and / or hardware, or a module that performs the same function can be implemented by a combination of multiple sub-modules or sub-units, etc. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0128] Based on the same concept as the methods described above, this specification also provides an electronic device, including: a processor; a memory for storing processor-executable instructions; wherein the processor performs the steps of the method as described in any of the above embodiments by executing the executable instructions.

[0129] Based on the same concept as the methods described above, this specification also provides a computer-readable storage medium having computer instructions stored thereon that, when executed by a processor, implement the steps of the methods as described in any of the above embodiments.

[0130] Based on the same concept as the methods described above, this specification also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the methods as described in any of the above embodiments.

[0131] What those skilled in the art will understand is: In this specification, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, product, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, product, or apparatus. Without further limitation, the presence of additional identical or equivalent elements in a process, method, product, or apparatus that includes said elements is not excluded.

[0132] In this specification, “a,” “an,” and “the” do not specifically refer to the singular, but may also include the plural.

[0133] In this specification, ordinal numbers such as "first," "second," etc., do not necessarily indicate order; they are often used to distinguish between objects. For example, "first server" and "second server" usually refer to two servers. To differentiate between these two servers, they are described as "first server" and "second server." Of course, sometimes these two servers may be the same server.

[0134] In this specification, unless explicitly stated otherwise, "receiving and sending data" does not necessarily mean direct receiving and sending; it can also mean indirect receiving and sending. For example, A receiving data sent by B can be understood as A directly receiving the data sent by B, or it can be understood as A indirectly receiving the data sent by B through other entities such as C. Similarly, B sending data to A can be understood as B sending the data directly to A, or it can be understood as B indirectly sending the data to A through other entities such as C. Here, C can be one entity, or it can be two or more entities.

[0135] In this specification, unless explicitly stated otherwise, the relationships between structures can be direct or indirect. For example, when describing "A is connected to B," unless it is explicitly stated that A and B are directly connected, it should be understood that A can be directly connected to B or indirectly connected to B. Similarly, when describing "A is on top of B," unless it is explicitly stated that A is directly above B (AB is adjacent and A is above B), it should be understood that A can be directly above B or indirectly above B (AB is separated by other elements, and A is above B). And so on.

[0136] This specification uses specific terms to describe embodiments thereof. Terms such as "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Furthermore, those skilled in the art can combine and integrate the different embodiments or examples described herein, as well as the features of those different embodiments or examples, without contradiction.

[0137] Although one or more embodiments of this specification provide method steps as described in the embodiments or flowcharts, it is understood that the order of steps listed in the embodiments or flowcharts is only one of many possible execution orders and does not represent the only execution order. Therefore, when the claims involve method steps, any changes or adjustments to the order of such steps, or the parallelism between steps, are also within the scope of protection of the claims.

Claims

1. An information delivery method, comprising: Obtain the target generation requirements for the information to be delivered; Target material information that matches the target generation requirements is identified from a pre-built material information library, and target prompt words are generated based on the target material information and the target generation requirements; The target prompt words are input into an information delivery system based on an artificial intelligence model, so that the information delivery system generates a general presentation code that meets the target generation requirements and target operation instructions for the general presentation code; wherein, the general presentation code is generated based on the target material information and meets the target generation requirements, and the general presentation code is used to present the graphic and text content of the information to be delivered on different information delivery platforms. The target operation command invokes the automated delivery tool, which then renders and delivers the information to be delivered on different information delivery platforms based on the general presentation code.

2. The method as described in claim 1, wherein obtaining the target generation requirements for the information to be delivered specifically includes: Obtain the description text for the information to be delivered; The description text is input into at least two parsing models, and each parsing model parses the description text to obtain the candidate generation requirements output by each parsing model; wherein, the model parameters corresponding to different parsing models are different; Based on the preset evaluation model, each candidate generation requirement is evaluated to obtain the comprehensive performance score corresponding to each candidate generation requirement. Based on the comprehensive performance score, the target generation requirement is determined from among the candidate generation requirements.

3. The method as described in claim 2, wherein the evaluation of the requirements for generating each candidate specifically includes: For each candidate generation requirement, a relevance score is determined based on the semantic correlation between the candidate generation requirement and the description text; a completeness score is determined based on the comprehensiveness of the content covered by the candidate generation requirement in the description text; and an innovativeness score is determined based on the novelty of the candidate generation requirement. The comprehensive performance score corresponding to the candidate generation requirement is obtained by weighted summation based on the relevance score, the completeness score, the innovation score, and the weights corresponding to each score.

4. The method of claim 2, further comprising: Collect user interaction data on different information to be delivered; wherein, the interaction data is used to characterize the user's browsing and clicking behavior on historical recommended information; Based on the interaction data, the preferred recommendation information is determined from each piece of information to be delivered, and the sample generation requirements for generating the preferred recommendation information are determined. For each analytical model, the model parameters of the analytical model are adjusted with the optimization objective of minimizing the deviation between the sample generation requirement and the target generation requirement output by the analytical model; and, based on the interaction data, the target performance score corresponding to the sample generation requirement is determined, and the model parameters of the evaluation model are adjusted with the optimization objective of minimizing the deviation between the target performance score and the comprehensive performance score output by the evaluation model.

5. The method as described in claim 1, wherein determining target material information matching the target generation requirements from a pre-built material information database specifically includes: For each candidate material in the material information database, determine the semantic matching degree between the candidate material and the target generation requirement; Based on the semantic matching degree and the generation time corresponding to the candidate material information, the recommendation score corresponding to the candidate material information is determined; wherein, the closer the generation time is to the current time, the greater the recommendation score. Based on the recommendation score, the target material information is determined from the candidate material information.

6. The method of claim 1, wherein the target generation requirement includes: The target material information includes at least one of the following: design requirements for images, requirements for content text, and requirements for theme text. The target material information includes: theme text that matches the target generation requirements and content text that matches the theme text.

7. The method as described in claim 1, wherein the information delivery system comprises: Solution planning model and information generation model; The target prompts include: planned prompts and generated prompts; Generate generic rendering code that meets the target generation requirements, specifically including: The planning prompts are input into the scheme planning model so that the scheme planning model can determine a planning scheme for generating the information to be deployed; wherein, the planning scheme is generated based on the target material information and the target generation requirements; The generated prompt words and the planning scheme are input into the information generation model so that the information generation model generates a general presentation code that matches the planning scheme.

8. The method as described in claim 1, wherein the information delivery system comprises: Content moderation model; Before invoking the automated delivery tool via the target operation command, the method further includes: The general presentation code is input into the content review model so that the content review model can review the information content embedded in the general presentation code.

9. The method of claim 1, wherein generating target operation instructions for the general rendering code specifically includes: Generate initial operation instructions for the general rendering code; Based on the Model Context Protocol (MCP) set in the information delivery system, the instruction format of the initial operation instruction is converted into an instruction format that the automated delivery tool can recognize, thereby obtaining the target operation instruction.

10. The method of claim 1, wherein before rendering and delivering the information to be delivered on the different information delivery platforms, the method further comprises: The automated delivery tool is used to obtain the target document object model (DOM) of the login page corresponding to the information delivery platform. Based on the login page structure features reflected by the target DOM and the login operation logic, a login operation is performed for the information delivery platform.

11. An information delivery device, comprising: The acquisition unit is used to acquire the target generation requirements for the information to be delivered. The matching unit is used to determine the target material information that matches the target generation requirement in the pre-built material information library, and generate target prompt words based on the target material information and the target generation requirement; The output unit is used to input the target prompt words into an information delivery system based on an artificial intelligence model, so that the information delivery system generates a general presentation code that meets the target generation requirements and a target operation instruction for the general presentation code; wherein, the general presentation code is generated based on the target material information and meets the target generation requirements, and the general presentation code is used to present the graphic content of the information to be delivered on different information delivery platforms; The delivery unit is used to invoke the automated delivery tool through the target operation instruction, so that the automated delivery tool renders and delivers the information to be delivered on the different information delivery platforms according to the general presentation code.

12. An electronic device, comprising: processor; A memory for storing processor-executable instructions; wherein the processor implements the steps of the method as described in any one of claims 1-10 by executing the executable instructions.

13. A computer-readable storage medium having stored thereon computer instructions that, when executed by a processor, implement the steps of the method as claimed in any one of claims 1-10.

14. A computer program product comprising a computer program / instructions that, when executed by a processor, implement the steps of the method as claimed in any one of claims 1-10.

Images