Page building and delivery information processing method and system, and electronic device
By using generative models to analyze localized data and generate marketing campaign suggestions, combined with AI-powered product selection and page building, the problem of lagging market insights in cross-border export systems has been solved, enabling rapid response and efficient deployment, and improving the relevance and efficiency of marketing campaigns.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HANGZHOU ALIBABA INT INTERNET IND CO LTD
- Filing Date
- 2026-03-20
- Publication Date
- 2026-07-31
AI Technical Summary
In cross-border export commodity information service systems, market insights often lag behind market changes in overseas countries/regions, resulting in marketing campaigns failing to respond in a timely manner, missing window opportunities, and impacting traffic acquisition and conversion efficiency.
By analyzing localized data from target countries/regions using generative models, marketing campaign suggestions are generated, including campaign themes and strategy descriptions. Automated campaign page deployment is achieved through AI-powered product selection, creative content generation, and page building.
It enables rapid response to market changes, improves the responsiveness and efficiency of marketing campaigns, eliminates perceived time differences caused by geographical location, and ensures that marketing campaign themes match the needs of local users.
Smart Images
Figure CN122492301A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of information processing technology, and in particular to methods, systems and electronic devices for page construction and information delivery. Background Technology
[0002] In product information service systems (also known as "e-commerce platforms," which primarily provide services such as product publishing, display, and transactions), the theme and product selection of marketing campaigns directly impact system traffic, conversion rates, and other metrics. Specific marketing campaigns are typically conducted based on market insights, which include identifying trending events, purchasing trends, and so on. Market insights are usually the starting point for e-commerce operations; they are not only background research before developing campaign strategies but also the underlying navigation system driving subsequent decisions.
[0003] In existing technologies, market insights are typically conducted by operations personnel, using methods such as browsing user community platforms, attending trade shows, and reading industry media. However, in cross-border commodity information service systems, especially in cross-border export models, sellers (factories, brands, individuals, etc.) within a single country (e.g., country A) sell their goods to consumers in multiple overseas countries / regions. In this model, market insights usually need to be conducted targeting overseas countries / regions. However, system operations personnel typically live and work in country A, leading to delays in obtaining timely market insights from overseas countries / regions. This can result in delayed marketing activities in those countries, or even missed marketing windows. Summary of the Invention
[0004] This application provides methods, systems, and electronic devices for page construction and information processing, which can improve the responsiveness to market changes and the efficiency of advertising.
[0005] This application provides the following solution: A method for page creation and information delivery processing, comprising: Collect localized data relevant to the target country / region; By analyzing the localized data through a generative model, hot topic information related to the operation of the product information service system is identified. Based on the hot topic information and the localized knowledge information corresponding to the target country / region, selectable marketing activity suggestions are generated. The marketing activity suggestions include marketing activity themes and marketing strategy descriptions. The marketing strategy descriptions include scenario descriptions, product selection strategy descriptions, and placement strategy descriptions. After receiving a request to adopt the marketing campaign suggestion information and create a marketing campaign, the marketing strategy description is semantically understood through a generative model, and a set of products that match the product selection strategy description is selected. Material content corresponding to the scenario description is generated, and the corresponding marketing campaign page is built by filling the product set and the material content into the corresponding area in the preset page frame structure template. After the page is built, a generative model is used to perform semantic understanding on the content describing the delivery strategy, and page delivery about the marketing campaign page is delivered to users based on the semantic understanding results.
[0006] This also includes: After generating marketing campaign suggestion information, a generative model is used to generate evaluation content on the marketing campaign theme and marketing strategy description across multiple evaluation dimensions. The evaluation content includes evaluation text and evaluation scores to help determine whether to adopt the marketing campaign theme and marketing strategy description.
[0007] The generation of material content corresponding to the scene description information includes: Generate the theme text content to be displayed on the marketing campaign page, as well as the theme image materials related to the marketing campaign page.
[0008] This also includes: Based on the image material requirements of the header area of the marketing campaign page, the theme image material is processed so that it can be displayed in the header area of the marketing campaign page; Based on the image material requirements of the resource slots on the distribution channel page, the theme image material is processed so that when providing the access link to the marketing campaign page in the resource slots on the distribution channel page, the same theme image material as the header area of the marketing campaign page is used for processing and display.
[0009] Among them, the hot topics identified for the target countries / regions include multiple events, and multiple marketing campaigns are created accordingly; The step of delivering the marketing campaign page to the user based on the semantic understanding results includes: When it is necessary to display marketing campaign page information to target users, the generative model determines the target marketing campaign suitable for the target users based on semantic understanding results and the target users' behavioral paths and / or interest tag information, so as to display the marketing campaign page corresponding to the target marketing campaign to the target users.
[0010] The step of delivering the marketing campaign page to users based on semantic understanding results also includes: During the display of the advertising channel page, the access address information of the marketing activity page corresponding to the target marketing activity is provided to the advertising channel page so that an access entry link to the marketing activity page is provided in the resource slot of the advertising channel page. After the access entry link is triggered, the marketing activity page corresponding to the target marketing activity is displayed.
[0011] This also includes: The experiment process is initiated, in which a generative model is used to perform semantic understanding on the description of the delivery strategy. Based on the semantic understanding results, the page about the marketing campaign is delivered to the user. Then, the generative model is used to statistically analyze the delivery performance data corresponding to multiple different marketing campaigns, so as to determine whether to take a certain marketing campaign offline or to allocate different traffic to different marketing campaigns based on the delivery performance data.
[0012] A page creation and delivery information processing system, comprising: The data collection module is used to collect localized data related to the target country / region; The perception task intelligent engine is used to analyze the localized data through a generative model, determine the hot event information related to the operation of the commodity information service system, and generate selectable marketing activity suggestions based on the hot event information and the localized knowledge information corresponding to the target country / region. The marketing activity suggestions include the marketing activity theme and the marketing strategy description. The marketing strategy description includes scenario description, product selection strategy description and placement strategy description. A task intelligence engine is built to perform semantic understanding of the marketing strategy description content through a generative model after receiving a request to adopt the marketing activity suggestion information and create a marketing activity. Based on the semantic understanding results, a set of products that match the product selection strategy description content is selected, material content corresponding to the scenario description information is generated, and the corresponding marketing activity page is built by filling the product set and the material content into the corresponding area of the preset page frame structure template. The intelligent engine for campaign delivery is used to perform semantic understanding of the content describing the delivery strategy through a generative model after the page is built, and to deliver the page about the marketing campaign to users based on the semantic understanding results.
[0013] A computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of any of the preceding methods.
[0014] An electronic device, comprising: One or more processors; and A memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of any of the preceding methods.
[0015] A computer program product includes a computer program / computer executable instructions that, when executed by a processor in an electronic device, implement the steps of any of the preceding methods.
[0016] According to the specific embodiments provided in this application, the following technical effects are disclosed: This application embodiment allows for the collection of localized data related to a target country / region. Then, a generative model is used to analyze this localized data, identifying trending events relevant to the operation of the product information service system. Based on these trending events and localized knowledge information corresponding to the target country / region, marketing activity suggestions are generated. These suggestions may include a marketing activity theme and a marketing strategy description, which includes scenario descriptions, product selection strategy descriptions, and placement strategy descriptions. Upon receiving a request to adopt the marketing activity suggestions and create a marketing activity, a generative model can be used to perform semantic understanding on the marketing strategy description, selecting a set of products that match the product selection strategy description and generating content materials corresponding to the scenario description. The product set and the content materials can then be filled into the corresponding areas of a pre-set page framework template to build the corresponding marketing activity page. After page construction, a generative model can also be used to perform semantic understanding on the placement strategy description, and the marketing activity page can be displayed to users based on the semantic understanding results. This approach eliminates the reliance on passively receiving information manually, instead proactively capturing strategies and generating strategy suggestions based on trending topics. This eliminates perceived time differences caused by geographical location, ensuring that the event theme better matches the needs of local users. Furthermore, AI-powered product selection results, AI-generated copywriting, and content can be dynamically assembled, enabling interface adjustments without manual intervention. Complete event page generation is completed in seconds, supporting rapid deployment and automated page placement. Clearly, this solution connects all stages—product selection, content generation, page construction, and page placement—around strategies generated from automated market insights, achieving an automated "insight-based deployment" system. This facilitates a shift from a "passive follow-up" to a "proactive prediction" operational model, improving responsiveness to market changes and increasing deployment efficiency.
[0017] Of course, any product implementing this application does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description
[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0019] Figure 1 This is a schematic diagram of the system architecture provided in the embodiments of this application; Figure 2 This is a flowchart of the method provided in the embodiments of this application; Figure 3 This is a schematic diagram of the interface provided in an embodiment of this application; Figure 4 This is a schematic diagram of the system provided in the embodiments of this application; Figure 5 This is a schematic diagram of the electronic device provided in the embodiments of this application. Detailed Implementation
[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.
[0021] First, it's important to clarify that, in existing technologies, the general operational process for marketing activities in a product information service system is as follows: Operations personnel conduct market research through backend data analysis to identify trends or trending events, and determine the activity theme accordingly. Then, products are selected around the theme; for example, high-selling, highly rated, and cost-effective products can be chosen to attract user clicks and generate traffic. After product selection, the activity page is built. This involves designing the page's framework, which may include a header image area, an atmosphere area, a product guide area, and a product display area. Additionally, the design of relevant image materials may be involved. The selected products and related materials are then filled into the corresponding areas of the framework, completing the page construction. Finally, the completed page and selected products are promoted to target users through various channels.
[0022] As mentioned in the background section, market insight is the starting point for operations. However, existing cross-border export commodity information service systems suffer from blind spots in localized hot topic information perception and delayed response, severely restricting traffic acquisition and conversion efficiency. For example, suppose country B is experiencing a catastrophic weather event. At this time, users may need emergency supplies more urgently. However, if the commodity information service system fails to detect this event in time (mainly because marketers usually work and live in country A, making it difficult to perceive sudden hot topics in other countries), it may continue to recommend daily necessities to local users, failing to meet their more urgent needs.
[0023] To address the aforementioned issues, this application embodiment leverages AI (Artificial Intelligence) capabilities to analyze localized data from multiple different countries / regions, enabling the perception of trending events related to the marketing of the product information service system. Furthermore, it can automatically generate corresponding marketing campaign themes and marketing strategy descriptions. For example, specific marketing strategy descriptions may include scenario descriptions, product selection strategy descriptions, and placement strategy descriptions. Subsequently, after operations personnel select and adopt a theme, based on semantic understanding of the marketing strategy descriptions, automated product selection, content generation, and page construction can be completed. After page construction, semantic understanding of the placement strategy descriptions can be used to complete page placement targeting consumer users. Thus, this approach provides an automated "insight-based placement" system. By using AI technology to connect the entire chain—opportunity identification, inventory matching, content production, page construction, and precise distribution—it achieves a transformation from a "passive follow-up" to a "proactive prediction" operational model, improving responsiveness to market changes and increasing the efficiency of placement efforts.
[0024] In practical implementation, multiple agents can collaborate to complete the entire process from market insight to page deployment. Within a single agent, a generative model can be used in conjunction with the engineering workflow. The so-called intelligent agent, referred to as an "intelligent engine" in this embodiment, is a system that autonomously executes tasks by designing workflows using available tools. It understands needs, plans goals, and executes tasks through generative models, possessing the ability to autonomously understand, perceive, plan, remember, and use tools, and can automatically complete complex tasks. Specifically, the generative model can be a model that, given certain implicit parameters or conditions, learns the distribution patterns of real data and can randomly generate observable data, thereby "creating" new data that does not appear in the original data but conforms to similar characteristics. Specifically, the intelligent system provided in this embodiment can first include a data collection module, and then include a perception task agent, a page building task agent (which may also include a product selection agent, a material generation agent, etc.), and a deployment task agent.
[0025] The data collection module is primarily used to collect localized data related to a specific target country / region. This includes search-related data within the specific product information service system, and may also include public network data from external systems such as search engines, social networks, and content platforms. Regarding in-site search-related data, this embodiment focuses primarily on the dynamic changes in search activity, or search fluctuations, including whether there has been a recent surge in search volume for a particular search term, etc.
[0026] The Perception Task Agent is primarily used to analyze the localized data using generative models, identify trending events relevant to the operation of the product information service system, and generate marketing campaign themes and marketing strategy descriptions based on these trending events and localized knowledge of the target country / region. These marketing strategy descriptions include scenario descriptions, product selection strategy descriptions, and campaign placement strategy descriptions. In other words, the output of the Perception Task Agent can directly be the marketing campaign theme and marketing strategy descriptions; that is, the specific theme of the marketing campaign to be created, and the specific strategies to be used in product selection and placement, can all be directly generated by the Perception Task Agent. This generated result can be provided to operations personnel, who do not need to be aware of the localized data behind the generated result; they can directly decide whether to adopt the Agent's generated result to create the corresponding marketing campaign.
[0027] If the operations team adopts a theme and corresponding strategy generated by the perception task agent, an automated page building and deployment process can be triggered. During page building, the product selection agent first performs semantic understanding on the product selection strategy description generated by the perception task agent. Based on the semantic understanding results, it matches products with similarity criteria in the product information database, forming the set of products selected for the marketing campaign. Of course, in addition to the semantic understanding of the product selection agent, some product selection algorithm logic can also be combined during the product selection process. Furthermore, the material generation agent can perform semantic understanding on the scene description information to generate material content corresponding to the scene description information. This includes the theme text content displayed on the corresponding event page, and image material content displayed in areas such as the event page header. The generated image material content can also be processed according to business rule templates to meet the display rules requirements in areas such as the page header, including image size. In addition, considering that specific marketing event pages may provide access through some channel pages (such as traffic distribution pages such as client homepages), in order to achieve visual continuity between the access point and the page after the jump, the same image material as the header area of the event page can also be displayed at the access point of the specific channel page. Of course, since the access point resource positions provided by the channel pages will have different requirements for the size of the image material, the image material generated by the material generation agent can also be processed based on the requirements of the specific channel page resource positions for image materials.
[0028] After completing product selection and creative content generation, the corresponding marketing campaign pages can be built based on the specific product set, creative content, and pre-set page framework templates. The page framework templates can be pre-configured and fixed, or multiple templates can be pre-configured, and one can be randomly selected during page building, or the template can be chosen based on its suitability for the specific marketing campaign, and so on.
[0029] After the page setup is complete, the marketing campaign page can be pushed to target users through the task delivery agent. This can be done through various channels. The specific channel pages and placements can be configured by the operations team. Within the same country / region, there may be multiple different trending events, leading to the creation of multiple different marketing campaigns. When delivering pages to target users, the Anget delivery agent can also match marketing campaigns with users. Based on users' in-site behavior paths and interest tags, it matches the most suitable marketing campaign theme and strategy to users in real time and delivers the corresponding page. This means that different users in the same country / region may see different marketing campaign information when viewing the same channel page, allowing them to browse different marketing campaign pages, thus achieving a personalized experience for each user.
[0030] By leveraging the multiple agents described above, a strategy generated from automated market insights can be implemented, connecting all stages from product selection, creative content generation, page construction, and page deployment. Furthermore, in practical implementation, an online experiment can be automatically initiated before actual online deployment. During this experiment, the deployment agent performs semantic understanding of the deployment strategy description, and based on the semantic understanding results, delivers the marketing campaign page to users. Real-time online performance data can then be collected. Generative models can also be used to statistically analyze the performance data of multiple different marketing campaigns. This allows for decisions on whether to take a particular campaign offline or allocate different traffic to different campaigns, ensuring that the actual marketing campaigns are better suited to the localized needs of specific countries / regions.
[0031] In other words, such as Figure 1 As shown in this embodiment, a "strategy center" can be provided firstly for collecting data from both inside and outside the platform and for data analysis using AI capabilities. The data from inside the platform mainly includes search data, traffic data, product structure, and user demographics; the data from outside the platform mainly includes marketing events, weather conditions, and competitor trends. The AI-powered data analysis component can perform statistical analysis and attribution diagnosis on the data from inside the platform. For the data from outside the platform, it can combine external trends with internal user and product data to optimize strategies by country, demographic, and scenario, thereby generating marketing campaign strategies at the country / region level. Multiple marketing campaign strategies can be generated for the same country / region.
[0032] The "Selection and Distribution" stage involves audience selection, product selection, and the generation of promotional materials, culminating in the setup of the event page. Audience selection primarily uses trending events or specific product-specific rules to identify target audiences. Product selection automatically generates product selection rules based on trending events or the audience's ability to choose products, creating a product pool. Material generation includes generating materials based on the campaign scenario and guidelines, such as multilingual text and images.
[0033] In the "Online Experiment" phase, after the event venue page is built, the experiment can be automatically launched, displaying the specific event venue page to users. AI is then integrated to analyze the performance data obtained from all online experiments, providing data and strategic conclusions. Subsequently, based on the experiment's conclusions, traffic rules between different marketing campaigns can be set, or it can be determined whether to launch or take a particular marketing campaign offline, and so on.
[0034] The specific implementation schemes provided in the embodiments of this application will be described in detail below.
[0035] Example 1 First, this first embodiment provides a method for page construction and information delivery processing, see [link to documentation]. Figure 2 The method may include: S201: Collect localized data relevant to the target country / region.
[0036] This step primarily involves localized data collection on a country / region basis. Localized data collection refers to gathering relevant data within a specific country / region to identify key events and trends. Specifically, this can include tracking search fluctuations within a product information service website or analyzing external trend data. External data can originate from public digital platforms, such as social media, news and forums, search engine systems, and publicly available government and institutional data. Data acquisition methods can include obtaining data through interfaces provided by external platforms or acquiring relevant offline data from local partners in the specific country / region.
[0037] For on-site data, the main focus can be on the location of on-site search data. Specifically, data related to fluctuations in on-site search activity can be collected. For example, the search volume for a certain term last week and this week can be used to identify changes in search volume for that term. These changes in search volume can reflect the product trends and interests of users in that country / region.
[0038] S202: Analyze the localized data using a generative model to determine hot topic information related to the operation of the product information service system, and generate selectable marketing activity suggestions based on the hot topic information and the localized knowledge information corresponding to the target country / region. The marketing activity suggestions include marketing activity themes and marketing strategy descriptions. The marketing strategy descriptions include scenario descriptions, product selection strategy descriptions, and placement strategy descriptions.
[0039] For the collected localized data from specific countries / regions, analysis can be performed using a perception task agent for market insights. This perception task agent can include a generative model that analyzes the localized data to extract trending events relevant to the operation of the product information service system. For example, whether extreme weather is occurring, whether certain sporting events are underway, or whether there is a surge in searches for a particular keyword. Subsequently, based on the trending event information and localized knowledge of the target country / region, marketing campaign themes and marketing strategy descriptions can be generated. The specific marketing strategy descriptions can include scenario descriptions, product selection strategy descriptions (including theme tone, price range, etc.), and campaign placement strategy descriptions (including target countries / regions, campaign start and end times, target audience, etc.).
[0040] Specifically, localized knowledge information related to the target country / region can include commonalities among users in that country / region regarding product preferences. For example, young people in a certain country generally enjoy esports culture and are correspondingly interested in related products and accessories. This localized knowledge information can be provided to the generative model through prompts or external knowledge bases.
[0041] Regarding trending event information, the specific implementation can include various types of trending events, which can be determined by analyzing the aforementioned collected localized data. For example, specific trending event types can include "site-wide trending searches," "national holidays / celebrations," "sports events," "film and television IPs," and "competitor dynamics." The specific data sources and analysis methods used can differ for different types of trending events. For instance, for site-wide trending events, the primary data source for analysis is site-wide search data. The analysis method can be to periodically statistically analyze keyword search volume and compare it with historical search volumes to analyze trends. A relatively short period, such as one week, can be used to promptly identify short-term changes in search term volume. For trending events such as "sports events" or "film and television IPs," the primary data source for analysis can be data collected from external public digital platforms, and so on. Therefore, in practice, different prompts can be constructed to express the specific types of hot events that need to be obtained, as well as the corresponding data sources, analysis methods, and other information. The generative model can then be called using these different prompts to obtain information on various types of hot events.
[0042] After extracting the aforementioned trending event information, the generative model can make decisions based on this information and the localized knowledge, determining whether a marketing campaign is suitable. If so, it can generate suggested campaign themes and strategy descriptions. As mentioned earlier, the specific strategy descriptions can include scenario descriptions, product selection strategy descriptions, and campaign placement strategy descriptions. In other words, it can not only determine the campaign theme but also describe specific scenarios, product selection methods, and campaign placement strategies to guide subsequent work.
[0043] By collecting localized data from the target country / region and performing analysis and content generation based on generative models, the system can output recommended marketing campaign suggestions. These suggestions can include the campaign's theme and strategy description. This content can be presented to operations staff for further evaluation and decision-making. To further assist operations personnel in decision-making and selection from multiple AI-generated marketing campaigns, optionally, after generating the campaign suggestions, the generative model can also generate evaluations across multiple dimensions. These evaluations include text and scores to aid in determining whether to adopt the suggestions. Specific evaluation dimensions may include product purchase intent, industry demand intensity, and short-term growth potential, among others.
[0044] For example, such as Figure 3 The diagram illustrates the interface displayed to operations and other staff after a marketing campaign has been generated by a generative model, including the campaign theme, strategy description, and campaign evaluation. The diagram shows the specific strategy type (in this example, trending search terms within the site), strategy name (including the campaign theme: "Ultimate Gear Set for Gamers," and the scenario description: "High-performance equipment and personalized accessories designed for home gaming spaces"), strategy details (including its performance in localized data and the rationale based on localized knowledge), product selection strategy, placement strategy, AI evaluation, and more. Operations and other staff can use this information to evaluate the campaign and decide whether to adopt the suggested activities.
[0045] Through this AI data analysis and the generation of event themes and strategies, we can move away from passively receiving information and actively capture strategies. Based on hot topics, we can generate strategy suggestions, thereby eliminating the perceived time difference caused by geographical location and making the event themes more suitable for local user needs.
[0046] S203: After receiving a request to adopt the marketing campaign suggestion information and create a marketing campaign, the marketing strategy description is semantically understood through a generative model, and a set of products that match the product selection strategy description is selected. Material content corresponding to the scenario description is generated, and the corresponding marketing campaign page is built by filling the product set and the material content into the corresponding area of the preset page frame structure template.
[0047] After operations staff select and adopt a marketing campaign suggestion, they can simultaneously initiate a request to create the corresponding marketing campaign. At this point, the generative model in the product selection task agent can perform semantic understanding on the AI-generated product selection strategy description and conduct a highly relevant search within the site's product pool to select a set of products that match the product selection strategy description, thereby achieving automated product selection.
[0048] Additionally, the generative model within the content generation agent can perform semantic understanding on the aforementioned AI-generated scene descriptions to generate content that matches the scene description information. Specific content can include text and images. These text and images can be displayed on the generated event page; for example, images can be displayed in the header area to highlight the theme and create atmosphere. Furthermore, as mentioned earlier, these images can also be displayed in the resource slots of the channel pages used as entry points to ensure continuity and consistency in browsing (the image displayed at the entry point on the channel page is the same as the image displayed in the header area of the page after clicking the link). Of course, since the event page header area and the aforementioned channel page resource slots have different requirements for the size of specific image materials, after the generative model generates the image materials, specific business rules can be used to process the image materials to meet the requirements of the specific display slots. The same marketing campaign page may be distributed through multiple different channels, and the requirements for image materials may also be different for different channels. Therefore, it is possible to process the images separately for each channel and save the corresponding processing results for display in the resource positions of different channels.
[0049] After completing product selection and material generation, the AI-generated product selection results, AI-generated copy and material content can be dynamically assembled according to the preset page framework template. Interface adjustments can be achieved without manual intervention, and a complete event venue page can be generated in seconds, supporting rapid online deployment.
[0050] It's important to note that in practical implementation, in addition to product selection, audience selection can also be performed, primarily to identify the target audience for a specific marketing campaign. Specifically, a generative model can automate audience selection based on the aforementioned AI-generated campaign theme and scenario descriptions, as well as audience profiles from the product information service system. This automated audience selection result can also be output as part of the specific campaign strategy.
[0051] S204: After the page is built, a generative model is used to perform semantic understanding on the content describing the delivery strategy, and page delivery about the marketing campaign page is performed to users based on the semantic understanding results.
[0052] After the marketing campaign venue page is built, the generative model in the delivery task agent can perform semantic understanding on the previously AI-generated delivery strategy content. Then, based on the semantic understanding results, the page is delivered to consumer users; that is, the specifically built marketing campaign venue page is pushed to specific end users. In practice, this can be achieved through pre-defined delivery channels. For example, assuming the delivery channel page is the homepage of the current product information service system's client application, a specific resource slot on that homepage will be used to provide an access point to this marketing campaign venue page. When a user visits the homepage, they can see this access point in that resource slot. If the user clicks or performs other actions within that resource slot, they will be redirected to the marketing campaign venue page. Of course, in practice, the same access point can also be provided through other delivery channel pages.
[0053] It's important to note that in practice, multiple trending events may be identified for the same target country / region, resulting in multiple corresponding marketing campaigns (the number of campaigns depends on the actual trending events perceived by the AI). In this case, when the delivery agent needs to display marketing campaign page information to the target user, it can use a generative model based on semantic understanding results and the target user's behavioral path and / or interest tags to determine a suitable target marketing campaign for that user, thus displaying the corresponding marketing campaign page. This achieves precise matching of "people-product-place," enhancing traffic value. Specifically, during the delivery process, the access address information of the marketing campaign page corresponding to the target marketing campaign can be provided to the delivery channel page. This allows the delivery channel page to include an access link to the marketing campaign page in its resource slots. Once this access link is triggered, the marketing campaign page corresponding to the target marketing campaign can be displayed.
[0054] Furthermore, in practical applications, an automated experimental process can be initiated. In this process, a generative model can be used to perform semantic understanding of the description of the campaign strategy. Based on the semantic understanding results, the marketing campaign page is then delivered to users. The generative model then statistically analyzes the performance data of multiple different marketing campaigns to determine whether a particular campaign should be taken offline, or whether different traffic allocations should be made for different campaigns. The performance data can include page click-through rate, conversion rate, etc. Campaigns with poor performance can be taken offline, meaning they are no longer officially promoted to users. Alternatively, more traffic can be allocated to better-performing campaigns, ensuring they are seen by users in specific target countries / regions; otherwise, less traffic can be allocated to improve the overall traffic value.
[0055] In summary, through the embodiments of this application, localized data related to a target country / region can be collected. Then, the localized data is analyzed using a generative model to determine hot topic information related to the operation of the product information service system. Based on the hot topic information and the localized knowledge information corresponding to the target country / region, marketing activity suggestion information is generated, which may include a marketing activity theme and a marketing strategy description. The marketing strategy description includes scenario description, product selection strategy description, and placement strategy description. After receiving a request to adopt the marketing activity suggestion information and create a marketing activity, the generative model can perform semantic understanding on the marketing strategy description and select a set of products that match the product selection strategy description to generate material content corresponding to the scenario description information. Then, the corresponding marketing activity page can be built by filling the product set and the material content into the corresponding area of a preset page frame structure template. After the page is built, the generative model can also perform semantic understanding on the placement strategy description and, based on the semantic understanding results, deliver the marketing activity page to the user. This approach eliminates the reliance on passively receiving information manually, instead proactively capturing strategies and generating strategy suggestions based on trending topics. This eliminates perceived time differences caused by geographical location, ensuring that the event theme better matches the needs of local users. Furthermore, AI-powered product selection results, AI-generated copywriting, and content can be dynamically assembled, enabling interface adjustments without manual intervention. Complete event page generation is completed in seconds, supporting rapid deployment and automated page placement. Clearly, this solution connects all stages—product selection, content generation, page construction, and page placement—around strategies generated from automated market insights, achieving an automated "insight-based deployment" system. This facilitates a shift from a "passive follow-up" to a "proactive prediction" operational model, improving responsiveness to market changes and increasing deployment efficiency.
[0056] Example 2 This second embodiment corresponds to the first embodiment and provides a page building and information delivery processing system. See [link / reference] Figure 4 The system may include: Data collection module 401 is used to collect localized data related to the target country / region; The perception task intelligent engine 402 is used to analyze the localized data through a generative model, determine hot event information related to the operation of the commodity information service system, and generate selectable marketing activity suggestion information based on the hot event information and the localized knowledge information corresponding to the target country / region. The marketing activity suggestion information includes the marketing activity theme and the marketing strategy description content. The marketing strategy description content includes scenario description information, product selection strategy description content and placement strategy description content. The task intelligence engine 403 is used to perform semantic understanding of the marketing strategy description content through a generative model after receiving a request to adopt the marketing activity suggestion information and create a marketing activity. Based on the semantic understanding results, it selects a set of products that match the product selection strategy description content, generates material content corresponding to the scenario description information, and builds the corresponding marketing activity page by filling the product set and the material content into the corresponding area of the preset page frame structure template. The intelligent engine 404 for delivery tasks is used to perform semantic understanding of the delivery strategy description content through a generative model after the page is built, and to deliver the page about the marketing campaign to users based on the semantic understanding results.
[0057] In practical implementation, the perception task intelligence engine can also be used for: After generating marketing campaign suggestion information, a generative model is used to generate evaluation content on the marketing campaign theme and marketing strategy description across multiple evaluation dimensions. The evaluation content includes evaluation text and evaluation scores to help determine whether to adopt the marketing campaign theme and marketing strategy description.
[0058] Specifically, it can generate the theme text content to be displayed on the marketing campaign page, as well as the theme image materials related to the marketing campaign page.
[0059] In addition, the task intelligence engine can also be used for: Based on the image material requirements of the header area of the marketing campaign page, the theme image material is processed so that it can be displayed in the header area of the marketing campaign page; Based on the image material requirements of the resource slots on the distribution channel page, the theme image material is processed so that when providing the access link to the marketing campaign page in the resource slots on the distribution channel page, the same theme image material as the header area of the marketing campaign page is used for processing and display.
[0060] Among them, the hot topics identified for the target countries / regions include multiple events, and multiple marketing campaigns are created accordingly; Specifically, the intelligent engine for delivery tasks can be used to: when it is necessary to display marketing campaign page information to target users, determine the target marketing campaign suitable for the target users based on the semantic understanding results and the target users' behavior paths and / or interest tag information through the generative model, so as to display the marketing campaign page corresponding to the target marketing campaign to the target users.
[0061] In addition, the task delivery intelligent engine can also be used for: During the display of the advertising channel page, the access address information of the marketing activity page corresponding to the target marketing activity is provided to the advertising channel page so that an access entry link to the marketing activity page is provided in the resource slot of the advertising channel page. After the access entry link is triggered, the marketing activity page corresponding to the target marketing activity is displayed.
[0062] Furthermore, the device may also include: The experiment engine is used to initiate the experiment process. In the experiment process, the generative model performs semantic understanding on the content of the delivery strategy description, and delivers the page about the marketing campaign to the user based on the semantic understanding results. Then, the generative model statistically analyzes the delivery effect data corresponding to multiple different marketing campaigns, so as to determine whether to take a certain marketing campaign offline or to allocate different traffic to different marketing campaigns based on the delivery effect data.
[0063] For the parts of this embodiment that are not described in detail, please refer to the description in embodiment one and other parts of this specification, which will not be repeated here.
[0064] It should be noted that the embodiments of this application may involve the use of user data. In practical applications, user-specific personal data may be used in the scheme described herein within the scope permitted by applicable laws and regulations, provided that it complies with the applicable laws and regulations of the country (e.g., with the user's explicit consent, with the user being properly notified, etc.).
[0065] In addition, embodiments of this application also provide a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method described in any of the foregoing method embodiments.
[0066] And an electronic device, comprising: One or more processors; and A memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method described in any of the foregoing method embodiments.
[0067] A computer program product includes a computer program / computer executable instructions that, when executed by a processor in an electronic device, implement the steps of the method described in the foregoing method embodiments.
[0068] in, Figure 5 An exemplary architecture of an electronic device is shown, which may include a processor 510, a video display adapter 511, a disk drive 512, an input / output interface 513, a network interface 514, and a memory 520. The processor 510, video display adapter 511, disk drive 512, input / output interface 513, network interface 514, and memory 520 can communicate with each other via a communication bus 530.
[0069] The processor 510 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solution provided in this application.
[0070] The memory 520 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 520 can store the operating system 521 for controlling the operation of the electronic device 500, and the basic input / output system (BIOS) for controlling the low-level operations of the electronic device 500. Additionally, it can store a web browser 523, a data storage management system 524, and a page building and deployment processing system 525, etc. The aforementioned page building and deployment processing system 525 can be the application program that specifically implements the aforementioned steps in this embodiment. In summary, when implementing the technical solution provided in this application through software or firmware, the relevant program code is stored in the memory 520 and executed by the processor 510.
[0071] Input / output interface 513 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, speakers, vibrators, indicator lights, etc.
[0072] Network interface 514 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).
[0073] Bus 530 includes a pathway for transmitting information between various components of the device, such as processor 510, video display adapter 511, disk drive 512, input / output interface 513, network interface 514, and memory 520.
[0074] It should be noted that although the above-described device only shows the processor 510, video display adapter 511, disk drive 512, input / output interface 513, network interface 514, memory 520, bus 530, etc., in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the solution of this application, and does not necessarily include all the components shown in the figures.
[0075] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this application.
[0076] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for system or system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and relevant parts can be referred to the descriptions in the method embodiments. The systems and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.
[0077] The foregoing has provided a detailed description of the page construction and information delivery processing method, system, and electronic device provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.
Claims
1. A method for page construction and information delivery processing, characterized in that, include: Collect localized data relevant to the target country / region; By analyzing the localized data through a generative model, hot topic information related to the operation of the product information service system is identified. Based on the hot topic information and the localized knowledge information corresponding to the target country / region, selectable marketing activity suggestions are generated. The marketing activity suggestions include marketing activity themes and marketing strategy descriptions. The marketing strategy descriptions include scenario descriptions, product selection strategy descriptions, and placement strategy descriptions. After receiving a request to adopt the marketing campaign suggestion information and create a marketing campaign, the marketing strategy description is semantically understood through a generative model, and a set of products that match the product selection strategy description is selected. Material content corresponding to the scenario description is generated, and the corresponding marketing campaign page is built by filling the product set and the material content into the corresponding area in the preset page frame structure template. After the page is built, a generative model is used to perform semantic understanding on the content describing the delivery strategy, and page delivery about the marketing campaign page is delivered to users based on the semantic understanding results.
2. The method according to claim 1, characterized in that, Also includes: After generating marketing campaign suggestion information, a generative model is used to generate evaluation content on the marketing campaign theme and marketing strategy description across multiple evaluation dimensions. The evaluation content includes evaluation text and evaluation scores to help determine whether to adopt the marketing campaign theme and marketing strategy description.
3. The method according to claim 1, characterized in that, The generation of material content corresponding to the scene description information includes: Generate the theme text content to be displayed on the marketing campaign page, as well as the theme image materials related to the marketing campaign page.
4. The method according to claim 3, characterized in that, Also includes: Based on the image material requirements of the header area of the marketing campaign page, the theme image material is processed so that it can be displayed in the header area of the marketing campaign page; Based on the image material requirements of the resource slots on the distribution channel page, the theme image material is processed so that, during the process of providing the access link to the marketing campaign page in the resource slots on the distribution channel page, the same theme image material as the header area of the marketing campaign page is used for processing and display.
5. The method according to claim 1, characterized in that, The identified hot topics for the target countries / regions include multiple events, and multiple marketing campaigns were created accordingly. The step of delivering the marketing campaign page to the user based on the semantic understanding results includes: When it is necessary to display marketing campaign page information to target users, the generative model determines the target marketing campaign suitable for the target users based on semantic understanding results and the target users' behavior paths and / or interest tag information, so as to display the marketing campaign page corresponding to the target marketing campaign to the target users.
6. The method according to claim 5, characterized in that, The step of delivering the marketing campaign page to the user based on the semantic understanding results also includes: During the display of the advertising channel page, the access address information of the marketing activity page corresponding to the target marketing activity is provided to the advertising channel page so that an access entry link to the marketing activity page is provided in the resource slot of the advertising channel page. After the access entry link is triggered, the marketing activity page corresponding to the target marketing activity is displayed.
7. The method according to claim 5, characterized in that, Also includes: The experiment process is initiated, in which a generative model is used to perform semantic understanding on the description of the delivery strategy. Based on the semantic understanding results, the page about the marketing campaign is delivered to the user. Then, the generative model is used to statistically analyze the delivery performance data corresponding to multiple different marketing campaigns, so as to determine whether to take a certain marketing campaign offline or to allocate different traffic to different marketing campaigns based on the delivery performance data.
8. A page creation and information delivery processing system, characterized in that, include: The data collection module is used to collect localized data related to the target country / region; The perception task intelligent engine is used to analyze the localized data through a generative model, determine the hot event information related to the operation of the commodity information service system, and generate selectable marketing activity suggestions based on the hot event information and the localized knowledge information corresponding to the target country / region. The marketing activity suggestions include the marketing activity theme and the marketing strategy description. The marketing strategy description includes scenario description, product selection strategy description and placement strategy description. A task intelligence engine is built to perform semantic understanding of the marketing strategy description content through a generative model after receiving a request to adopt the marketing activity suggestion information and create a marketing activity. Based on the semantic understanding results, a set of products that match the product selection strategy description content is selected, material content corresponding to the scenario description information is generated, and the corresponding marketing activity page is built by filling the product set and the material content into the corresponding area of the preset page frame structure template. The intelligent engine for campaign delivery is used to perform semantic understanding of the content describing the delivery strategy through a generative model after the page is built, and to deliver the page about the marketing campaign to users based on the semantic understanding results.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method described in any one of claims 1 to 8.
10. An electronic device, characterized in that, include: One or more processors; as well as A memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method according to any one of claims 1 to 8.
11. A computer program product comprising a computer program / computer-executable instructions, characterized in that, When the computer program / computer executable instructions are executed by a processor in an electronic device, they implement the steps of the method according to any one of claims 1 to 8.