Intelligent interaction method and electronic device

Abstract

Disclosed in embodiments of the present disclosure are an intelligent interaction method and an electronic device. The method comprises: when a current page is displayed, at least on the basis of a function provided by the current page and / or behavioral data generated by a user in the current page, sensing shopping needs that the user may currently have; when a target shopping need that the user may currently have is sensed, on the basis of the target shopping need, generating a prompt text for dialogue with an artificial intelligence (AI) content generation model; and on the basis of the prompt text, calling a pre-trained AI content generation model, so as to generate target content corresponding to the target shopping need for display. By means of the embodiments of the present disclosure, on the basis of the sensing results of users' shopping needs, solutions corresponding to the shopping needs can be intelligently and dynamically provided, improving the information acquisition efficiency of users.

Description

Intelligent interaction method and electronic device

[0001] This disclosure claims priority to Chinese patent application number 202410021667.3, filed with the China Patent Office on January 5, 2024, and entitled “Intelligent Interaction Method and Electronic Device,” the entire contents of which are incorporated herein by reference. Technical Field

[0002] The present disclosure relates to the field of information processing technology, and in particular to an intelligent interaction method and electronic equipment. Background Art

[0003] As a centralized traffic scenario, product information service systems (commonly referred to as e-commerce platforms) need to provide comprehensive shopping guide services that address the shopping needs of various users. For example, user shopping needs include clear shopping goals, a specific shopping scope, and seeking product consultation. However, on current e-commerce platforms, users must search or browse through recommended lists on the homepage to find products that meet their needs, then browse the product details page to view the details, and then decide whether to purchase. This entire process requires multiple steps to obtain more information, requiring users to spend a lot of time searching for products and browsing shopping guides, which is very inefficient. Summary of the Invention

[0004] The present disclosure provides an intelligent interaction method and electronic device that can perceive the results of user shopping needs, intelligently and dynamically provide solutions corresponding to shopping needs, and improve the user's information acquisition efficiency.

[0005] The present disclosure provides the following solutions:

[0006] An intelligent interaction method, comprising:

[0007] During the display of the current page, the user's current shopping needs are perceived based on at least the functions provided by the current page and / or the user's behavior data generated on the current page;

[0008] After sensing the user's current potential shopping needs, a prompt text is generated based on the target shopping needs for dialogue with the artificial intelligence (AI) content generation model;

[0009] Based on the prompt text, a pre-trained AI content generation model is called to generate target content corresponding to the target shopping needs and display it.

[0010] The method is applied to an artificial intelligence (AI) interaction module, and the AI ​​interaction module is used to be implanted into any page.

[0011] The AI ​​interaction module exists in the form of a software development kit (SDK), so that the AI ​​interaction module can be implanted into any page through the SDK.

[0012] The multiple pages include: pages corresponding to multiple nodes on the shopping guidance link in the commodity information service system.

[0013] Among them, the AI ​​interaction module corresponds to a page element for display in the target page, wherein when the shopping demand is not perceived, the page element is in a folded or hidden state, and after the shopping demand is perceived, the page element is in an expanded or displayed state, so as to interact with the AI ​​interaction module through the page element.

[0014] Among them, also include:

[0015] After sensing the target shopping demand, an operation option for confirming the target shopping demand is provided in the target page element, so that after receiving the user's confirmation message, a prompt text for communicating with the artificial intelligence AI content generation model is generated according to the target shopping demand.

[0016] Wherein, the generated prompt text is multiple, and the method further includes:

[0017] An operation option for selecting multiple prompt texts is provided in the target page element, so that after receiving the prompt text selected by the user, a call is initiated to the AI ​​content generation model.

[0018] Among them, the information used to perceive the user's current possible needs also includes: the asset information of the user in the product information service system, the marketing activity information published in the product information service system, the marketing activity information associated with the product, and the user's shopping preference information.

[0019] Among them, the functions provided by the current page include: providing product recommendation information;

[0020] The sensing of the user's current needs based at least on the functions provided by the current page and / or the user's behavior data generated on the current page includes:

[0021] Based on the behavioral data generated by the user on the page containing the product recommendation information flow, it is judged whether the user has a clear shopping target product or a clear shopping target range, and the user's possible needs are perceived based on the judgment result.

[0022] Among them, the functions provided by the current page include: providing product search information;

[0023] The sensing of the user's current needs based at least on the functions provided by the current page and / or the user's behavior data generated on the current page includes:

[0024] Based on the product search condition information input by the user on the current page, it is judged whether the user has a clear shopping target product or a clear shopping target range, and the user's possible needs are perceived based on the judgment result.

[0025] The step of perceiving the user's possible needs based on the judgment result includes:

[0026] If it is determined that the user has a clear shopping target product, it is perceived that the user's target shopping needs are information provision and / or decision-making assistance content based on the product dimension, so that by generating corresponding prompt text, the AI ​​content generation model generates information provision and / or decision-making assistance content based on the product dimension.

[0027] The information provided and / or decision-making assistance content based on product dimensions includes: comparisons of the same products provided by different merchants in multiple dimensions and / or product recommendation information.

[0028] The step of perceiving the user's possible needs based on the judgment result includes:

[0029] If it is determined that the user has a clear shopping target range, and the target range is related to a certain target scenario, then it is perceived that the user's target shopping needs are shopping guidance based on the target scenario, so that by generating corresponding prompt text, the AI ​​content generation model generates product recommendation information in the form of a shopping list aggregated by the target scenario.

[0030] In particular, when providing the product recommendation information in the form of the shopping list, the AI ​​content generation model is also used to generate an atmosphere background image corresponding to the target scene.

[0031] The step of perceiving the user's possible needs based on the judgment result includes:

[0032] If it is determined that the user has a clear shopping target range, and the target range includes multiple different products under a certain target category, then it is perceived that the user's target shopping demand is to purchase products in the target category, or to compare the multiple different products, so that by generating corresponding prompt text, the AI ​​content generation model generates product recommendation information about the target category, or compares and / or recommends information on the multiple different products in multiple dimensions.

[0033] The step of perceiving the user's possible needs based on the judgment result includes:

[0034] If it is determined that the user has no clear shopping goal, it is determined whether the user has clear shopping drivers based on the user's historical behavior data. If so, it is perceived that the user's target shopping needs are information related to the shopping drivers, so that the AI ​​content generation model can generate content aggregated by the shopping drivers by generating corresponding prompt text.

[0035] The functions provided by the current page include: providing detailed information about the target product;

[0036] The sensing of the user's current needs based at least on the functions provided by the current page and / or the user's behavior data generated on the current page includes:

[0037] If the user stays on the current page for longer than a threshold or swipes multiple screens but does not add the product to the set to be settled or perform a purchase operation, it is perceived that the user's target shopping demand is to obtain more comprehensive information about the target product, or to obtain information about the effect of the target product in the desired usage scenario, so that the AI ​​content generation model can summarize the detailed information of the target product by generating corresponding prompt text and generate a summarized text, or after the user provides relevant images of the usage scenario, generate an image of the usage effect when the target product is simulated in the usage scenario.

[0038] Among them, the functions provided by the current page include: providing a list of products that the user has added to the set of products to be settled;

[0039] The sensing of the user's current needs based at least on the functions provided by the current page and / or the user's behavior data generated on the current page includes:

[0040] Based on the commodities in the set of commodities to be settled, it is determined whether the user has a need to obtain price changes of commodities in the set of commodities to be settled, add more commodities to the set of commodities to be settled to meet the conditions for using a certain user's rights, or recommend related commodities to the commodities in the set of commodities to be settled, so that by generating corresponding prompt text, the AI ​​content generation model can bring the commodities with price changes to the front of the current page for display, or provide recommendation information for commodities that meet the conditions for using or belong to related commodities.

[0041] An intelligent interactive device, comprising:

[0042] a demand sensing unit configured to sense a user's current shopping needs during display of a current page, based at least on the functions provided by the current page and / or the user's behavioral data generated on the current page;

[0043] A prompt text generation unit is used to generate prompt text for dialogue with the artificial intelligence (AI) content generation model based on the target shopping needs that the user may currently have after sensing the target shopping needs;

[0044] The content generation unit is used to call a pre-trained AI content generation model based on the prompt text to generate target content corresponding to the target shopping needs and display it.

[0045] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of any of the aforementioned methods.

[0046] An electronic device, comprising:

[0047] one or more processors; and

[0048] A memory associated with the one or more processors, the memory being used to store program instructions, wherein the program instructions, when read and executed by the one or more processors, execute the steps of any of the aforementioned methods.

[0049] A computer program product comprises a computer program, wherein when the computer program is executed by a processor, the steps of any one of the aforementioned methods are implemented.

[0050] According to the specific embodiments provided by the present disclosure, the present disclosure discloses the following technical effects:

[0051] Through the embodiments of the present disclosure, in the process of displaying the current page, the user's current possible needs can be perceived based on the functions provided by the current page and / or the behavioral data generated by the user in the current page. After perceiving the user's current possible target shopping needs, a prompt text for communicating with the AI ​​(Artificial Intelligence) content generation model can be generated based on the target shopping needs. After that, the pre-trained AI content generation model can be called based on the prompt text to generate target content corresponding to the target shopping needs and display it. In this way, since the user's possible target shopping needs can be perceived, and prompt text can be automatically generated based on the perceived target shopping needs, and content for solving the corresponding problems can be generated, it is a demand-responsive active AI interaction method. In this way, the user does not need to initiate interaction with the AI, but the AI ​​interaction module actively initiates it, and can also automatically construct prompt text for communicating with the AI ​​model. Therefore, the difficulty and threshold of user interaction with AI are reduced, which is conducive to improving user experience. It can also perceive the results based on different user shopping needs and provide intelligent and dynamic solutions to corresponding shopping needs, so as to flexibly adapt to diverse user needs and scenarios and improve user information acquisition efficiency.

[0052] In an optional implementation method, an AI interaction module can be provided in the form of SDK (Software Development Kit), etc., and the AI ​​interaction module can be implanted in a variety of different pages. For example, it can be pages corresponding to multiple nodes on the shopping guidance link in the product information service system, etc., so that the AI ​​interaction module is untied from the specific page, so as to provide users with a unified interactive experience across different pages.

[0053] Furthermore, this approach is more user-friendly and convenient for new users who are unfamiliar with e-commerce systems. Furthermore, by providing support for interactive methods such as voice, gestures, and video, it can also provide a more user-friendly experience for user groups such as the visually impaired.

[0054] Of course, any product implementing the present disclosure does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0056] FIG1 is a schematic diagram of a system architecture provided by an embodiment of the present disclosure;

[0057] FIG2 is a flow chart of a method provided by an embodiment of the present disclosure;

[0058] 3 to 6 are schematic diagrams of user interfaces provided by embodiments of the present disclosure;

[0059] FIG7 is a schematic diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0060] The following will be combined with the accompanying drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the embodiments described are only part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field are within the scope of protection of the present disclosure.

[0061] In the embodiments of the present disclosure, AIGC (Artificial Intelligence Generated Content) technology can be used to help users of the commodity information service system improve the efficiency of obtaining information. Among them, the so-called AIGC is to use the AI ​​content production model to produce some content, including text, images, videos, etc. Driven by the wave of AI big models, AI has rapidly improved in natural language processing, complex scene understanding, etc., combined with massive data and algorithm model optimization, and has a relatively solid underlying foundation in actual landing scenarios. It is in this context that the embodiments of the present disclosure provide an application method of AI in e-commerce scenarios.

[0062] It's important to note that some application systems have already proposed implementing large AI models in e-commerce scenarios. However, there are still some implementation challenges. For example, existing applications typically provide a permanent, fixed interactive portal on a specific page. While browsing the page, users can access the AI ​​interactive interface through this portal at any time. They can then engage in conversations with the AI ​​by asking questions, and the AI ​​can then generate content tailored to the user's questions.

[0063] The problems with the above implementation are at least:

[0064] Users are required to actively initiate interaction with the AI ​​and ask questions. However, in practice, it can be difficult for users to understand whether they can get help from the AI, or to determine when and how to interact with the AI. Many users struggle with how to ask questions, especially when it comes to asking questions to the AI. In other words, users struggle to clearly describe their needs. Furthermore, users may not fully understand some potential needs. For example, some users may be accustomed to purchasing lower-priced items without realizing it. Consequently, it's difficult for users to construct high-quality prompts for communicating with large AI models, making it difficult to obtain the content they truly desire through AI interaction.

[0065] In addition, specific AI interaction functions are provided by specific pages, that is, users can only interact with AI on such pages. For example, assuming that an AI interaction entrance is provided on the client's homepage, users can only interact with AI while browsing the homepage. However, e-commerce platforms usually provide multiple shopping guide links, each of which may include multiple nodes and involve multiple different pages. If only some pages implement AI interaction functions, it means that users will not be able to interact with AI when visiting other pages. If AI interaction functions are required to be implemented separately on each page, on the one hand, the workload will be very large, and on the other hand, since different pages may correspond to different development teams, it is difficult to unify the AI ​​interaction functions implemented separately. Users may need to familiarize themselves with the interaction methods on different pages, which will bring new troubles to the user's browsing process.

[0066] In response to the above problems, in the embodiments of the present disclosure, the application method of the AI ​​big model in the e-commerce platform is improved. Specifically, a demand-responsive AI interactive "robot" module can be implanted in the e-commerce platform. This module can be implanted in the pages of the full-link scenario of user shopping (for example, home page, channel page, venue page, details page, shopping cart page, order page, etc.) in the form of an independent plug-in. In this way, users can get AI help in multiple different pages, and the interface, interaction logic, etc. can be kept unified, without the need for the developers of each page to carry out specific development work. In addition, the "AI" interactive robot can perceive the shopping needs or demands that the user may have based on the type or function of the specific page, the actual behavioral data generated by the user on the page, etc., and then intelligently generate prompt text for dialogue with AI. If the user approves the prompt text, the AI ​​content generation model can generate specific consumption suggestions or solutions corresponding to specific shopping needs / demands, thereby flexibly adapting to diverse user needs and business scenarios. In other words, the so-called "demand-responsive" AI interaction means that it does not directly provide a permanent interactive entrance on a certain page and allow users to initiate questions. Instead, it can perceive the user's shopping needs on any page based on the page's own type, function and other information as well as the user's current behavioral data on the page. If it can be perceived that the user may have a certain shopping need, the AI ​​can initiate interaction with the user and provide dynamically changing dialogue prompts and generate content based on the specific perception results. Of course, in another implementation method, if there is no need to consider the consistency of the AI ​​interaction experience between different pages, the above-mentioned "demand-responsive" AI interaction module can also be directly implemented on different pages.

[0067] Among them, the specific AI interaction module may also correspond to a page element for display in the target page, wherein, when the user's shopping needs are not perceived, the page element may be in a folded or hidden state, and after the shopping needs are perceived, the page element may be switched to an expanded or displayed state, so as to interact with the AI ​​interaction module through the page element.

[0068] Specifically, after sensing a user's shopping needs, the AI ​​can proactively generate prompt text based on the user's shopping needs. Optionally, the user can be asked to confirm the perceived shopping needs. For example, prompt information can be displayed through the aforementioned page elements, including "Do you need to prepare a shopping list for Christmas?" and so on. If the user approves the perceived result, corresponding prompt text can be generated, and the AI ​​content generation model can be invoked based on the prompt text to generate content. In this way, since the specific prompt text is generated based on the user's shopping needs, the specific content generated is also corresponding to the shopping needs. Furthermore, the user does not need to manually construct specific prompt text, thus reducing the difficulty or threshold of user-AI interaction and further improving the intelligence of the AI ​​interaction process. The content generated by the AI ​​content generation model can still be displayed through the aforementioned page elements. Of course, the size of the page element can be dynamically adjusted based on the specific content required to be displayed, and can even be enlarged to full screen. In addition, it can jump to another follow-up page to display the content generated by the specific AI content generation model, etc.

[0069] Among them, from the perspective of technical implementation, as mentioned above, each page can develop support for the above functions independently, or, in order to reduce the development costs of page developers, a unified SDK (Software Development Kit, that is, a toolkit provided by the service provider to implement a certain function of the product software) can be provided. In this way, the above functional modules can be implanted into specific pages through this SDK. In this way, the AI ​​interaction function is not strongly coupled with the specific page, and the specific page developer does not need to pay attention to the specific implementation of the function, thereby reducing the development cost of the page developer. At the same time, the specific intelligent AI interaction interface, interaction form, etc. can be unified across multiple different pages to provide users with a unified experience across scenarios and pages.

[0070] From the perspective of system architecture, referring to FIG1 , in a specific implementation method, the embodiment of the present disclosure can provide the above-mentioned AI interaction module in the form of SDK or the like (it can also be in the form of SaaS (Software as a Service) or the like), so that the specific AI interaction model can be implanted into different pages corresponding to multiple nodes on a specific shopping guide link through SDK or the like, so that these pages can have demand-responsive AI interaction functions. In the above-mentioned demand-responsive AI interaction process, the specific AI interaction module can obtain relevant information of the current page, and can also obtain information such as the behavior data generated by the user in the current page through the user behavior data collection module. In addition, it can also obtain the asset information of the user in the current commodity information service system through the user asset data center on the server side, obtain the marketing activity information published in the commodity information service system from the marketing activity center, obtain the marketing activity information associated with the commodity through the commodity data center, obtain the current user's shopping preference information from the user personalized preference data center, and so on. The user's shopping needs can be perceived in combination with this information. After detecting a target shopping need, prompt text can be generated to communicate with the artificial intelligence (AI) content generation model. Based on the prompt text, the pre-trained AI content generation model can be invoked to generate and display target content corresponding to the target shopping need. The specific perceived shopping need can also be confirmed by the user. Upon user approval, the corresponding prompt text and content generation will be performed, and the generated content will be displayed.

[0071] From the perspective of application scenarios, the main application scenarios of the demand-responsive intelligent AI interaction function provided by the embodiments of the present disclosure can be divided into pre-sales shopping guide scenarios and after-sales service scenarios. Among them, in terms of pre-sales shopping guides, compared with the traditional online shopping process, the demand-responsive intelligent AI interaction function has built corresponding capabilities for real-time response to user needs, complex scene understanding, user intent recognition, and deep personalized shopping guides in the shopping process. Especially when the wireless and mobile screens are small and the displayed information is limited, more accurate information and product access can be provided through simple human-computer interaction. In the after-sales scenario, the demand-responsive AI assistant can also provide users with logistics inquiries, refunds and returns, dispute protection and other services in the form of dialogue.

[0072] The specific implementation scheme provided by the embodiment of the present disclosure is introduced in detail below.

[0073] First, the embodiment of the present disclosure provides an intelligent interaction method. In one implementation, the method can be applied to an AI interaction module. The AI ​​interaction module can be specifically used to be implanted in any page, so that multiple different pages can have demand-responsive AI interaction functions and provide users with a unified experience across scenarios and pages. For example, any specific page can include any page among multiple pages of the same application, and can be implanted in multiple pages. For example, the above-mentioned AI interaction module can be implanted in multiple pages corresponding to multiple nodes on the shopping guide link in the commodity information service system, and so on. In an optional manner, the AI ​​interaction module can exist in the form of an SDK so that the specific AI interaction module can be implanted into multiple different pages through the SDK. Alternatively, the AI ​​interaction module can also exist in the form of SaaS, etc. This SDK plug-in or SaaS method can not only quickly implant the AI ​​interaction module into multiple pages of various e-commerce platforms, but also into pages of systems such as AI large model plug-ins and social software, with strong scalability and reusability. It should be noted here that, regardless of whether it exists in the form of SDK or SaaS, the AI ​​interaction module in the embodiment of the present disclosure and the specific page can be independent of each other, that is, there is no strong coupling relationship between the AI ​​interaction function and the specific page. Different pages can be implanted with the same AI interaction module, and the AI ​​interaction module can interact with users in different pages according to the same data processing logic and front-end display strategy.

[0074] Specifically, referring to FIG2 , the method may include:

[0075] S201: In the process of displaying the current page, the user's current needs may be perceived based on at least the functions provided by the current page and / or the behavior data generated by the user in the current page.

[0076] Among them, the current page is the page that the user is currently browsing. Specifically, it can be the page corresponding to any node on any shopping guide link in the product information service system. For example, it can include the client homepage, product recommendation page, search page, channel page, event venue page, store page, product details page, and can also include the "shopping cart" page, "complement order" page, after-sales service-related pages, etc.

[0077] In the embodiment of the present disclosure, the AI ​​interaction entrance is not provided directly in a specific page. Instead, after analyzing the user's behavioral data, if it is perceived that the user may have a certain shopping demand, the user can be actively interacted with to respond to the shopping demand. Therefore, it is called demand-responsive AI interaction.

[0078] Specifically, when perceiving the user's shopping needs, at least the functions provided by the current page and / or the behavioral data generated by the user in the current page can be used. That is, in the embodiment of the present disclosure, since AI interactive functions can be provided in the pages corresponding to multiple nodes of the full link, when the user browses different pages, the functions provided by the specific pages are different, and therefore, the needs they have may also be different. In addition, even in the same page, if different behavioral data are generated, the user needs that can be expressed will also be different. Therefore, in a specific implementation, it is possible to combine the functions of the page itself, the behavioral data generated by the user in the current page, etc., to perceive whether the user has a certain shopping need during this browsing process. Of course, in a specific implementation, the information based on which the user's current needs may be perceived may also include: the asset information of the user in the commodity information service system (for example, whether there is a "red envelope" that has not been written off, etc.), the marketing activity information published in the commodity information service system, the marketing activity information associated with the commodity, and the user's shopping preference information (for example, whether the user is a user who is sensitive to price attributes, whether he prefers certain brands, etc.). By integrating multiple aspects of information, a more accurate perception of the user's shopping needs can be achieved.

[0079] Specifically, there are multiple ways to perceive a user's shopping needs. For example, one approach involves pre-establishing rules such as: if a user generates certain behavioral data on a certain page, it can be assumed that the user may have a certain shopping need. For example, if a user scrolls through recommended products on a product recommendation page and first clicks to view the details of a certain mobile phone, then returns to the recommended product page and clicks to view the details of another mobile phone of the same model, then based on this behavioral data, it can be determined that the user may have a need to purchase this mobile phone. Alternatively, if a user scrolls through recommended products on a product recommendation page and clicks to view the details of multiple products, all of which are related to a specific scenario, it can be inferred that the user may have a shopping need related to that scenario. Alternatively, if a user enters information such as the brand and model of a specific mobile phone while searching for a product, this indicates that the user has a clear need to purchase this mobile phone. Alternatively, if a user simply enters keywords such as "Christmas," this indicates that the user may have a need to purchase products related to Christmas, without specifically specifying a specific product or type. In this way, after collecting behavioral data, it is possible to detect whether the user has a certain target shopping demand by matching it with this rule.

[0080] In another way, the perception of user shopping needs can also be completed by the AI ​​big model. Specifically, the type and function of the page, the behavioral data information generated by the user on the page, etc. can be input into the AI ​​big model, and the AI ​​big model can understand the specific data to perceive whether the user has a certain shopping need. Of course, in this way, the asset information of the user in the product information service system, the marketing activity information published in the product information service system, the marketing activity information associated with the product, the user's shopping preference information, etc. can also be input into the AI ​​big model to be used for the perception of the user's current shopping needs. In addition, in the specific implementation, the AI ​​big model can be trained in advance with some training samples, and the parameters of the AI ​​big model can be tuned to enable it to have the ability to understand the above information and perceive the specific user shopping needs.

[0081] Alternatively, pre-defined rules can be combined with the perception of the AI ​​model. In practice, this can be done by categorizing and summarizing potential user needs across various pages to facilitate rule development or AI model training. The following describes the categorization and perception of user needs, using the client homepage (typically used for product recommendations), search page, shopping cart page, and product details page as examples of pre-sales guide scenarios.

[0082] 1. Client homepage: As a public domain centralized traffic scenario, the homepage is distributed to all types of users. Therefore, the homepage usually needs to take into account the shopping needs of all types of users. For example, it can include:

[0083] 1. Targeted shopping needs: This refers to a specific brand and product that users are interested in, requiring them to quickly find the product, obtain pricing, discounts, comparisons, and other information, and make an order. In this case, users can be provided with product-specific information and shopping guides. However, existing homepages typically don't directly address this user need. Users must access search, detail pages, and other clickable content to obtain more information about the product.

[0084] 2. Only a clear shopping scope: This means that users have a clear shopping scenario and desired categories / brands, but not yet a specific product they're looking for. They need to first find products that meet their needs. In this case, a scenario-based shopping guide can be provided to users, understanding their needs and making more precise product recommendations. However, existing homepages generally lack this capability, requiring users to spend time searching for products and browsing the shopping guide.

[0085] 3. No shopping goals at all: In this case, specific needs may vary depending on the user's driving factors. For example, users driven by discounts or promotions may want to learn about platform promotions, etc. To address this situation, users can be provided with clearer promotional information that is aligned with their preferences. However, existing technologies, such as homepages, lack aggregated information on promotions. Instead, it is often scattered across various business modules on the platform, requiring users to independently search for it.

[0086] Alternatively, users who browse for "grass-seeking" products may want to know if there are any good product recommendations. In this case, we can provide users with more good product recommendations and content exposure. Existing technologies can meet certain good product recommendation scenarios through recommendation cards on the homepage, but lack more interesting and rich good product recommendations, which cannot meet user needs.

[0087] Furthermore, some users may need advice on daily life issues, such as "My white shoes tend to get dark easily. What can I do?" In this case, the platform can answer specific questions based on specific scenarios. Furthermore, since there is a potential for purchase conversion, relevant product recommendations can be provided. However, existing homepages do not meet this need. Users often seek advice from the content platform, "seeking" something, and then proceed to the shopping platform to place an order.

[0088] Based on the above summary and analysis of the possible needs that users may have when browsing the homepage, specifically in the process of users browsing this type of homepage page, it is possible to judge whether the user has a clear shopping target product or a clear shopping target range based on the behavioral data generated by the user on the page, and then perceive the possible needs of the user based on the judgment result.

[0089] For example, suppose a user slides through specific recommended products on the home page that contains a product recommendation information flow, first clicks to view the details page of a certain mobile phone, then returns to the product recommendation page, and clicks to view the details page of another mobile phone of the same model. Based on the above behavioral data, it can be determined that the user has a clear demand for shopping target products, specifically, the user may want to buy this mobile phone. Furthermore, it can be perceived that the user's target shopping demand may be based on the provision of information and / or decision-making assistance content in the product dimension, for example, a detailed introduction to this mobile phone, or a summary of user reviews of this mobile phone, etc., to facilitate shopping decisions.

[0090] Alternatively, suppose a user scrolls through a specific recommended product on a product recommendation page and clicks to view the detailed information of multiple products. These products are all related to a certain scenario. It can be inferred that the user may have shopping needs related to that scenario, that is, a user with only a clear shopping scope. In this case, the specific perceived need can be to obtain shopping guidance information based on the target scenario. For example, if the multiple products clicked by the user are all related to "Christmas", it can be perceived that the user may have a need for shopping guidance based on the "Christmas" scenario, and so on.

[0091] Alternatively, suppose that a user slides through the recommended product cards on the product recommendation page, but never performs operations such as clicking to view details. In this case, it can be inferred that the user is currently "browsing" on the page without a clear goal (neither a clear product nor a clear scene, brand, etc.). In this case, it can be inferred that the user is in a situation without a clear goal. When performing demand perception, the user's asset information in the product information service system, the marketing activity information published in the product information service system, the marketing activity information associated with the product, and the user's shopping preference information (which can specifically include the main driving factors, for example, price-sensitive users may belong to price / promotion-driven users, or "planting grass"-driven users, etc.) can be combined to comprehensively perceive the user's possible needs. For example, it can be determined based on the user's historical behavior data whether the user has a clear shopping driving factor. If so, it can be perceived that the user's target shopping demand is information related to the shopping driving factor, for example, including the aforementioned information related to price reduction promotions, or "planting grass"-related content, etc.

[0092] 2. Search page: For search scenarios, users generally have clear shopping demands, including clear scenarios, categories, or clear products of interest. For example, these may include:

[0093] 1. Scenario-based search needs: This typically involves fuzzy searches based on the input keywords, such as "Christmas outfit," "mobile phone," "pants," and so on. This helps users identify the specific context and create a list of recommended products that align with their shopping habits. However, existing technologies lack continuous interaction capabilities, relying solely on users to continuously adjust keywords, resulting in a long decision cycle.

[0094] 2. Product-based search needs: This typically involves entering keywords for precise searches, such as "a certain brand, a certain model of mobile phone." In this case, users may need more information about the product, including product features and regulations, pricing, store and specific product services, reviews, shipping locations, fulfillment information, and more. Existing technologies support product-based filtering, but require users to manually filter and compare features, and offer no shopping guide recommendations for the SKU (Stock Keeping Unit) dimension.

[0095] For this type of search page, in the embodiment of the present disclosure, it is possible to judge whether the user has a clear shopping target product or a clear shopping target range based on the product search condition information entered by the user on the current page, and perceive the user's possible needs based on the judgment result.

[0096] For example, assuming the user enters the search keyword "Christmas", it can be perceived that the user's need may be to obtain a list of items related to the "Christmas" scene, or assuming the user enters the search keyword "a certain brand and model of mobile phone", it can be perceived that the user's need may be to obtain a comparison of similar products related to the product, etc., to help the user make a selection decision.

[0097] 3. Product Details Page: The product details page primarily provides information about a specific product, including a brief introduction to its features, color / color guides, merchant services, product reviews, and delivery fulfillment. While browsing this page, if a user hasn't yet taken actions like adding an item to a shopping cart or placing an order, they may want to obtain more information about the specific product, including product features / color guides, pricing, the store, and the specific product's services, reviews, shipping locations, and fulfillment. For products in certain categories, they may also want to know how the item will perform in their intended use scenario. For example, before placing an order for clothing or furniture, users may want to know how a specific item will look on them, or how it will coordinate with their home decor or other furniture. In existing technology, users can only obtain more comprehensive information by reading the entire page. Alternatively, for size recommendations, shipping time, and other information, they can usually contact the store's customer service staff, which can be affected by the merchant's customer service resources, online hours, time zone differences, and other factors.

[0098] In an embodiment of the present disclosure, if the user's stay time on the current page exceeds a threshold or swipes multiple screens, but does not perform an operation of adding the item to the set of items to be settled (i.e., the "shopping cart") or a purchase operation, then it is perceived that the user's target shopping needs may result in obtaining more comprehensive information about the target item, or obtaining information on the effect of the target item in the desired usage scenario.

[0099] Fourth, the pending checkout page, also known as the shopping cart page, is where users typically focus on promotions, discounts, and price reductions for items already added to their cart, as well as discounts offered during promotional events, before making their final purchase decision. In this scenario, users' target shopping needs typically include: price changes for added items, price reduction alerts when added items reach their expected price, recommendations for items to add to their cart, and related recommendations for items already added to their cart.

[0100] Therefore, in the embodiment of the present disclosure, based on the product information in the set of products to be settled, including the comparison of the current price of the product and the price when it was added to the shopping cart, whether the product participates in a certain promotional activity, etc., it can be determined whether the user has the need to obtain price changes of the products in the set of products to be settled, add more products to the set of products to be settled to meet the conditions for using a certain user benefit, or recommend related products to the products in the set of products to be settled.

[0101] For pages related to after-sales scenarios, such as logistics inquiry pages, user needs can be directly determined as obtaining information such as logistics status and logistics details. Or, if some special status occurs, such as being detained at customs, corresponding solutions may also be needed, etc. Alternatively, for reverse transaction pages such as returns and refunds, user needs can be perceived based on the status of the transaction. For example, if a refund or return request has not yet been submitted, the user may need to obtain descriptive information related to the refund instructions, or if a refund or return request has already been submitted, the user may want to be informed of the refund progress, etc.

[0102] S202: After sensing the target shopping needs that the user may currently have, generate prompt text for communicating with the artificial intelligence AI content generation model based on the target shopping needs.

[0103] After detecting that the user may have a specific target shopping need based on the page information and user behavior data, a prompt text can be generated based on the target shopping need to communicate with the artificial intelligence (AI) content generation model. Of course, in specific implementations, the page elements corresponding to the aforementioned AI interaction module can also provide an operation option for confirming the perceived target shopping need, and the prompt text will be generated after the user completes the confirmation operation.

[0104] For example, in the homepage scenario, if the user is detected to have a clear shopping desire for a specific product, a confirmation question can be generated for the user, such as "Are you looking for a certain product?" This can be displayed through the aforementioned page elements. If the user confirms by selecting "Yes" or directly clicking on the question, a specific prompt text can be generated for input into the AI ​​model for content generation. Of course, in a specific implementation, the prompt text can also be generated first, and the AI ​​model can perform preliminary content generation before confirming the user. For example, assuming that the user has a clear shopping desire for a specific product, the corresponding prompt text can be generated first, and the AI ​​model can generate content related to this product, determine whether multiple merchants are selling the product, whether any merchants are running promotions for the product, etc. Afterwards, text is generated to confirm the user, such as "Are you looking for a certain product? There are 10 products currently on sale," etc. In this way, since it is not only possible to perceive that the user may be looking for a certain product, but also to provide the user with information with obvious benefits such as "10 products are currently on sale", it can be more conducive to increasing the user's interest in the detailed content, etc.

[0105] After the user confirms their perceived target shopping needs, prompt text can be generated for that specific target shopping need. For example, in the example above, the prompt text could be: "Check which similar products are on sale for a certain product, which ones are participating in promotions, compare them based on price, performance, logistics, after-sales service, and other dimensions, and provide recommended purchase suggestions." This prompt text can be used as input into the AI ​​content generation model to generate specific content.

[0106] In other words, when using AI content generation models to generate content, the construction of prompt text is crucial. To obtain high-quality content, high-quality prompt text must first be constructed. Traditionally, users construct prompt text based on their own needs and then input it into the AI ​​master model for content generation. However, in the present embodiment, considering that constructing high-quality prompt text can be quite challenging for average users, such as the difficulty of clearly describing their needs in the prompt text and the time it takes to construct the prompt text, a solution is provided to help users construct prompt text. In other words, in the present embodiment, not only can the AI ​​master model intelligently generate content for users, but even the prompt text can be automatically generated. From another perspective, the input information for the AI ​​master model is no longer the traditional user-entered prompt text, but can include page type, function, user behavior data within the page, and even information about the user's assets in the current system, marketing activities within the system, product-related marketing activities, and the user's personalized preferences. Of course, in a specific implementation, this information can also be pre-processed to generate specific prompt text, which is then input into the AI ​​master model for content generation.

[0107] S203: Based on the prompt text, a pre-trained AI content generation model is called to generate target content corresponding to the target shopping needs and display it.

[0108] After generating specific prompt text, the pre-trained AI content generation model can be called based on the prompt text to generate and display target content corresponding to the target shopping needs. In various cases, such as different pages and different user behavior data, the corresponding prompt text and generated content will be different due to different user needs.

[0109] For example, if the current page being browsed is the home page or the search page, and it is determined that the user has a clear shopping target product, it can be perceived that the user's target shopping needs are information provision and / or decision-making assistance content based on the product dimension. At this time, a prompt text corresponding to the need can be generated. Accordingly, the AI ​​content generation model can generate information provision and / or decision-making assistance content based on the product dimension.

[0110] Specifically generated product-based information provision and / or decision-making assistance content may include: a summary of detailed information, evaluation information, and other content about the product, or a comparison of the same product offered by different merchants across multiple dimensions and / or product recommendations. Regarding the latter, when a user clearly needs to purchase a certain product, since the current system may include multiple merchants selling the same product, it is important for the user to choose from these merchants, and when making a choice, it is usually necessary to compare the same products offered by different merchants across multiple dimensions. In the prior art, users can only click to view the details page of each product of the same type, obtain relevant information, and then compare them. However, in the disclosed embodiment, if the user is found to have the above needs, corresponding prompt text can be automatically generated, and the AI ​​model can generate relevant content such as comparisons of multiple products of the same type. Users can directly view the comparison results of multiple products of the same type on the same page and can compare them separately from multiple dimensions, or they can also provide purchase recommendations from the overall perspective. In this way, users can make choices based on the dimensions they are more concerned about. For example, after comparing identical products across multiple dimensions like price, logistics, after-sales service, and reviews, the recommended products for each dimension can be presented. This way, if a user is more focused on a particular dimension when shopping, they can choose the recommended product for that dimension. For example, if a user is more concerned with price, they can choose a product with an advantage in that dimension. If a user prefers to receive their goods as quickly as possible, they can choose a product with an advantage in logistics, and so on. This approach helps users more quickly compare multiple identical products and make purchasing decisions.

[0111] For example, as shown in Figure 3(A), suppose a user enters "Brand A X1 mobile phone" on the search page and initiates a search. This is a precise search, meaning that the user is aware of a specific shopping target product, specifically "Brand A X1 mobile phone." At this point, an action option for confirming this perceived result can be provided at the location shown in 31, where a message might be displayed: "Do you want to know a detailed description of Brand A X1 mobile phone?" If the user clicks this option, they will be directed to the AI ​​interaction page, as shown in Figure 3(B). Based on the keywords entered by the user on the search page and their behavioral data on the search page, several optional question examples are provided, such as "How is the camera on Brand A X1 mobile phone?", "What is the battery life on Brand A X1 mobile phone?", "Find highly rated Brand A X1 mobile phones," and so on. Users can select specific questions from these questions based on their actual needs and ask the AI ​​assistant. These questions can then serve as prompts for interacting with the AI ​​content generation model. For example, if a user selects the question "How is the camera on the X1 model of brand A?", the AI ​​assistant can generate content as shown in Figure 3(C). It can provide an overall evaluation of the phone's camera performance, such as "above average," and can also elaborate on several key points. This content can be generated by the AI ​​content generation model based on the phone's detailed description information, attribute parameter information, user evaluation information, etc.

[0112] If it is determined based on the user's browsing behavior on pages such as the homepage, or the keywords entered on the search page that the user has a clear shopping target range, it can be further determined whether the target range is related to a certain target scenario. If so, it can be perceived that the user's target shopping needs are shopping guidance based on the target scenario. At this time, after generating the corresponding prompt text, the AI ​​content generation model can generate product recommendation information in the form of a shopping list aggregated by the target scenario.

[0113] For example, suppose a user browses multiple recommended products on the homepage, many of which are related to "home renovation." It can be determined that the user's target scope is likely related to the "home renovation" scenario. Therefore, it can be perceived that the user's target shopping needs may be based on shopping guides for this scenario. In this case, as shown in Figure 4(A), interactive content such as "Looking for home renovation ideas?" can be provided in the page element corresponding to the AI ​​interaction module. If the user clicks on this page element, it indicates that the user may have accepted the perceived result. Furthermore, corresponding prompt text can be generated, and content related to the "home renovation" scenario can be generated through the AI ​​content generation model. On the front-end page, the page element corresponding to the AI ​​interaction module can be enlarged, displaying content related to the "home renovation" scenario. For example, as shown in Figure 4(B), the generated content can include product recommendations in the form of a shopping list aggregated from the "home renovation" scenario. Specifically, recommendations can be categorized by living room, bedroom, kitchen, and so on. In addition, if it is perceived based on previous user behavior data that the user may have other needs, this can also be displayed through the page element. For example, as shown in Figure 4(B), "You can also ask me the following questions: Are there any other discounts? Are you looking for children's clothing?" and so on.

[0114] It should be noted that, under existing methods, users can use the search function to search for product information related to a specific scenario. However, the search logic typically uses keyword matching to provide search results, and multiple complex search result sorting mechanisms are involved. As a result, while users can view search results related to specific scenario keywords on the search results page, it is difficult for them to quickly purchase the desired products. For example, after entering the keyword "Christmas," the search results may include multiple "Christmas"-related products, where multiple consecutive products may be Christmas trees, or multiple consecutive products may be Christmas stockings, and so on. However, in the disclosed embodiments, if a user is found to have shopping needs related to a specific scenario, the AI ​​content generation model can generate product recommendation information in the form of a shopping list aggregated by the target scenario. For example, this may include Christmas trees, Christmas stockings, bells, wreaths, etc., with one or more recommended products corresponding to each category, allowing users to directly complete the purchase of Christmas-related products according to the list on the page, etc. In addition, in the preferred real-time method, when providing the product recommendation information in the form of a shopping list, the AI ​​content generation model can also generate an atmospheric background image corresponding to the target scenario. For example, when generating a shopping list related to a "Christmas" scene, a background image with a "Christmas" atmosphere can also be generated at the same time, so that the displayed content generation results have more scene atmosphere and enhance the user experience.

[0115] In addition, when it is determined that the user has a clear target range, the target range may also be related to a certain category. For example, the user may browse multiple different products under the same category on the current page. In this case, it can be perceived that the user's target shopping demand may be to purchase products in the target category, or to compare multiple different products in the target category. At this time, the AI ​​content generation model can also generate corresponding prompt text to generate product recommendation information about the target category, or compare and / or recommend information for the multiple different products in multiple dimensions.

[0116] For example, suppose a user browses the recommended products on the homepage and first browses the details page of a pair of wireless headphones as shown in Figure 5(A), and then browses the details page of another pair of wireless headphones as shown in Figure 5(B). It can be determined that the user has a relatively clear target range, that is, related to the category of wireless headphones. It can then be determined that the user may need to compare these two wireless headphones. At this time, the message "Compare them?" can be provided on the page as shown in Figure 5(B), and thumbnails of the two headphones can also be displayed for the user's reference. After the user clicks on the page element, the page element can be enlarged as shown in Figure 5(C), displaying AI-generated content. Specifically, it can include content comparing two products based on dimensions such as price, model, service, and rating, etc.

[0117] It should be noted that, if it is perceived that the user has a clear target range, after initiating AI interaction with the user, it is also possible to conduct a dialogue with the user to further understand the user's needs and then make more precise product recommendations. For example, as shown in Figure 6, 6(A) shows the client homepage. While the user is browsing this page, assuming that based on the user's behavioral data, etc., it is perceived that the user has a need to "buy a high-performance mobile phone", an operation option can be provided at the position shown in 61 to initiate a question to the user, such as "Are you looking for a high-performance mobile phone?" If the user clicks this operation option, they will enter the AI ​​interaction interface shown in Figure 6(B). In this interface, multiple candidate questions can be provided based on previously obtained user behavior data on the current page. The user can choose one to ask the AI ​​assistant, or they can enter a specific question in the input box below. Assuming that the question actually entered by the user is "Can you recommend me a high-performance mobile phone?" as shown in Figure 6(C), the AI ​​can further clarify the user's needs through dialogue with the user. For example, the AI ​​assistant can ask the user, "What do you mainly want to use this phone for: office / study, entertainment / socializing, or taking photos / playing games?" If the user answers, "I mainly use it for gaming, so I hope it has good performance," the assistant can then continue to ask the user, "Can you tell me roughly what your budget is?" The user can answer, "No more than $350," and so on. This way, the user can understand that they want to buy a phone for gaming, and their primary focus is on performance, with a price tag of no more than $350. Based on this information, more accurate recommendations can be made to the user. For example, the specific content generated by the AI ​​assistant can be shown in Figure 6(D), which includes three recommended phones and provides reasons for each recommendation. The user can then make further choices based on the recommended results.

[0118] In addition, if it is determined that the user does not have a clear shopping goal, it is possible to determine whether the user has a clear shopping driver based on the user's historical behavior data. If so, it can be perceived that the user's target shopping demand is product information related to the shopping driver. After that, the AI ​​content generation model can generate product content aggregated by the shopping driver by generating corresponding prompt text. In other words, a judgment can be made based on the user's own situation. If it is analyzed that a certain user is marketing-sensitive and has purchased heavily discounted "flash sale" products in the past, a list of discounted products can be provided; if the user is brand-conscious, product recommendation information related to these brands can be provided, and so on.

[0119] For the product details page, if it is perceived that the user's target shopping needs are to obtain more comprehensive information about the target product, or to obtain information about the effect of the target product in the expected usage scenario, the AI ​​content generation model can summarize the details of the target product and generate a summarized text, or, after the user provides relevant images of the usage scenario, generate an image of the usage effect when the target product is simulated in the usage scenario. For example, the detailed description information of the product can be summarized, or the user evaluation content of the product can be summarized, and so on. In this way, online consulting services can be provided to users 24 / 7, supplementing the customer service resources of the platform and merchants, providing more patient and real product shopping guides, helping to summarize buyer feedback and evaluations, and so on.

[0120] Alternatively, if the product is clothing, the user can be asked if they want to see how the item would look on them. If so, the user can upload a photo of themselves, and the AI ​​content generation model can then use its AI mapping capabilities to generate a visual representation of how the item would look on them for reference. Similarly, if the product is furniture, the user can upload a photo of their home, and the AI ​​content generation model can generate a visual representation of how the specific piece would look in their room.

[0121] If the current page is a shopping cart page, based on the items in the set of items to be settled, it is determined that the user has the need to obtain price changes of the items in the set of items to be settled, add more items to the set of items to be settled to meet the usage conditions of a certain user's rights, or recommend related items to the items in the set of items to be settled. The AI ​​content generation model can display the items with price changes at the front of the current page, or provide product recommendation information for items that meet the usage conditions or are related items.

[0122] In summary, through the embodiments of the present disclosure, in the process of displaying the current page, the AI ​​interaction module can perceive the user's current possible needs based on the functions provided by the current page and / or the behavioral data generated by the user in the current page. After perceiving the user's current possible target shopping needs, it can generate prompt text for dialogue with the artificial intelligence AI content generation model based on the target shopping needs. Afterwards, the pre-trained AI content generation model can be called based on the prompt text to generate target content corresponding to the target shopping needs and display it. In this way, since the user's possible target shopping needs can be perceived, prompt text can be automatically generated based on the perceived target shopping needs, and content for solving the corresponding problems can be generated, it is a demand-responsive active AI interaction method. In this way, the user does not need to initiate interaction with the AI, but the AI ​​interaction module actively initiates it, and can also automatically construct prompt text for dialogue with the AI ​​model. Therefore, the difficulty and threshold of user interaction with AI are reduced, which is conducive to improving user experience. It can also perceive the results based on different user shopping needs and provide intelligent and dynamic solutions to corresponding shopping needs, so as to flexibly adapt to diverse user needs and scenarios and improve user information acquisition efficiency.

[0123] In an optional implementation method, an AI interaction module can be provided in the form of SDK, etc., and the AI ​​interaction module can be implanted in a variety of different pages. For example, it can be pages corresponding to multiple nodes on the shopping guidance link in the product information service system, etc., so that the AI ​​interaction module is untied from the specific page, so as to provide users with a unified interactive experience across different pages.

[0124] Furthermore, this approach is more user-friendly and convenient for new users who are unfamiliar with e-commerce systems. Furthermore, by providing support for interactive methods such as voice, gestures, and video, it can also provide a more user-friendly experience for user groups such as the visually impaired.

[0125] It should be noted that the embodiments of the present disclosure may involve the use of user data. In actual applications, user-specific personal data may be used in the scenarios described herein within the scope permitted by applicable laws and regulations, subject to compliance with applicable laws and regulations of the country where the user is located (for example, with the user's explicit consent, effective notification to the user, etc.).

[0126] Corresponding to the aforementioned method embodiment, the present disclosure further provides an intelligent interaction device, which may include:

[0127] a demand sensing unit configured to sense a user's current shopping needs during display of a current page, based at least on the functions provided by the current page and / or the user's behavioral data generated on the current page;

[0128] A prompt text generation unit is used to generate prompt text for dialogue with the artificial intelligence (AI) content generation model based on the target shopping needs that the user may currently have after sensing the target shopping needs;

[0129] The content generation unit is used to call a pre-trained AI content generation model based on the prompt text to generate target content corresponding to the target shopping needs and display it.

[0130] The method is applied to an artificial intelligence (AI) interaction module, and the AI ​​interaction module is used to be implanted into any page.

[0131] Specifically, the AI ​​interaction module exists in the form of a software development kit SDK, so that the AI ​​interaction module can be implanted into any page through the SDK.

[0132] The multiple pages include: pages corresponding to multiple nodes on the shopping guidance link in the commodity information service system.

[0133] In a specific implementation, the AI ​​interaction module corresponds to a page element for displaying in the target page, wherein when the shopping demand is not perceived, the page element is in a folded or hidden state; after the shopping demand is perceived, the page element is in an expanded or displayed state, so as to interact with the AI ​​interaction module through the page element.

[0134] In addition, the device may further include:

[0135] The demand confirmation unit is used to provide an operation option for confirming the target shopping demand in the target page element after sensing the target shopping demand, so as to generate a prompt text for communicating with the artificial intelligence AI content generation model according to the target shopping demand after receiving the user's confirmation message.

[0136] Wherein, the generated prompt text is multiple, and the device further includes:

[0137] A prompt text selection unit is used to provide an operation option for selecting multiple prompt texts in the target page element, so as to initiate a call to the AI ​​content generation model after receiving the prompt text selected by the user.

[0138] Among them, the information used to perceive the user's current possible needs also includes: the asset information of the user in the product information service system, the marketing activity information published in the product information service system, the marketing activity information associated with the product, and the user's shopping preference information.

[0139] Specifically, the functions provided by the current page include: providing product recommendation information; in this case, the demand perception unit can be specifically used to:

[0140] Based on the behavioral data generated by the user on the page containing the product recommendation information flow, it is judged whether the user has a clear shopping target product or a clear shopping target range, and the user's possible needs are perceived based on the judgment result.

[0141] Alternatively, the function provided by the current page includes providing product search information; in this case, the demand sensing unit may be specifically used to:

[0142] Based on the product search condition information input by the user on the current page, it is judged whether the user has a clear shopping target product or a clear shopping target range, and the user's possible needs are perceived based on the judgment result.

[0143] Specifically, the demand sensing unit can be used to:

[0144] If it is determined that the user has a clear shopping target product, it is perceived that the user's target shopping needs are information provision and / or decision-making assistance content based on the product dimension, so that by generating corresponding prompt text, the AI ​​content generation model generates information provision and / or decision-making assistance content based on the product dimension.

[0145] The information provided and / or decision-making assistance content based on product dimensions includes: comparisons of the same products provided by different merchants in multiple dimensions and / or product recommendation information.

[0146] Alternatively, the demand sensing unit may be specifically used to:

[0147] If it is determined that the user has a clear shopping target range, and the target range is related to a certain target scenario, then it is perceived that the user's target shopping needs are shopping guidance based on the target scenario, so that by generating corresponding prompt text, the AI ​​content generation model generates product recommendation information in the form of a shopping list aggregated by the target scenario.

[0148] Among them, when providing the product recommendation information in the form of the shopping list, the AI ​​content generation model can also be used to generate an atmosphere background image corresponding to the target scene.

[0149] In addition, the demand sensing unit can be specifically used for:

[0150] If it is determined that the user has a clear shopping target range, and the target range includes multiple different products under a certain target category, then it is perceived that the user's target shopping demand is to purchase products in the target category, or to compare the multiple different products, so that by generating corresponding prompt text, the AI ​​content generation model generates product recommendation information about the target category, or compares and / or recommends information on the multiple different products in multiple dimensions.

[0151] In addition, the demand sensing unit can be specifically used to:

[0152] If it is determined that the user has no clear shopping goal, it is determined whether the user has clear shopping drivers based on the user's historical behavior data. If so, it is perceived that the user's target shopping needs are information related to the shopping drivers, so that the AI ​​content generation model can generate content aggregated by the shopping drivers by generating corresponding prompt text.

[0153] Furthermore, the functions provided by the current page include: providing detailed information about the target product;

[0154] At this time, the demand sensing unit can be specifically used to:

[0155] If the user stays on the current page for longer than a threshold or swipes multiple screens but does not add the product to the set to be settled or perform a purchase operation, it is perceived that the user's target shopping demand is to obtain more comprehensive information about the target product, or to obtain information about the effect of the target product in the desired usage scenario, so that the AI ​​content generation model can summarize the detailed information of the target product by generating corresponding prompt text and generate a summarized text, or after the user provides relevant images of the usage scenario, generate an image of the usage effect when the target product is simulated in the usage scenario.

[0156] Alternatively, the functions provided by the current page include: providing information on a list of commodities that the user has added to the set of commodities to be settled;

[0157] The demand sensing unit can be specifically used for:

[0158] Based on the commodities in the set of commodities to be settled, it is determined whether the user has a need to obtain price changes of commodities in the set of commodities to be settled, add more commodities to the set of commodities to be settled to meet the conditions for using a certain user's rights, or recommend related commodities to the commodities in the set of commodities to be settled, so that by generating corresponding prompt text, the AI ​​content generation model can bring the commodities with price changes to the front of the current page for display, or provide recommendation information for commodities that meet the conditions for using or belong to related commodities.

[0159] In addition, an embodiment of the present disclosure further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps of any one of the methods in the aforementioned method embodiments are implemented.

[0160] And an electronic device comprising:

[0161] one or more processors; and

[0162] A memory associated with the one or more processors, the memory being used to store program instructions, wherein the program instructions, when read and executed by the one or more processors, execute the steps of the method described in any one of the aforementioned method embodiments.

[0163] 7 exemplarily shows the architecture of an electronic device. For example, device 700 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, an aircraft, etc.

[0164] 7 , device 700 may include one or more of the following components: a processing component 702 , a memory 704 , a power component 706 , a multimedia component 708 , an audio component 710 , an input / output (I / O) interface 712 , a sensor component 714 , and a communication component 716 .

[0165] The processing component 702 generally controls the overall operation of the device 700, such as operations associated with display, phone calls, data communications, camera operation, and recording operations. The processing component 702 may include one or more processors 720 to execute instructions to complete all or part of the steps of the method provided by the technical solution of the present disclosure. In addition, the processing component 702 may include one or more modules to facilitate interaction between the processing component 702 and other components. For example, the processing component 702 may include a multimedia module to facilitate interaction between the multimedia component 708 and the processing component 702.

[0166] The memory 704 is configured to store various types of data to support operations on the device 700. Examples of such data include instructions for any application or method operating on the device 700, contact data, phone book data, messages, pictures, videos, etc. The memory 704 can be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0167] The power supply component 706 provides power to the various components of the device 700. The power supply component 706 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the device 700.

[0168] The multimedia component 708 includes a screen that provides an output interface between the device 700 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, slides, and gestures on the touch panel. The touch sensor can not only sense the boundaries of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 708 includes a front camera and / or a rear camera. When the device 700 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera can receive external multimedia data. Each front camera and rear camera can be a fixed optical lens system or have focal length and optical zoom capabilities.

[0169] The audio component 710 is configured to output and / or input audio signals. For example, the audio component 710 includes a microphone (MIC), which is configured to receive external audio signals when the device 700 is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal can be further stored in the memory 704 or transmitted via the communication component 716. In some embodiments, the audio component 710 also includes a speaker for outputting audio signals.

[0170] I / O interface 712 provides an interface between processing component 702 and peripheral interface modules, such as a keyboard, click wheel, buttons, etc. These buttons may include but are not limited to: a home button, volume buttons, a start button, and a lock button.

[0171] Sensor assembly 714 includes one or more sensors for providing various aspects of device 700 status assessment. For example, sensor assembly 714 can detect the open / closed state of device 700, the relative positioning of components, such as the display and keypad of device 700. Sensor assembly 714 can also detect changes in the position of device 700 or a component of device 700, the presence or absence of user contact with device 700, the orientation or acceleration / deceleration of device 700, and changes in the temperature of device 700. Sensor assembly 714 can include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 714 can also include an optical sensor, such as a CMOS (Complementary Metal-Oxide-Semiconductor) or CCD (Charge-Coupled Device) image sensor, for use in imaging applications. In some embodiments, sensor assembly 714 can also include an accelerometer, a gyroscope, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0172] The communication component 716 is configured to facilitate wired or wireless communication between the device 700 and other devices. The device 700 can access a wireless network based on a communication standard, such as WiFi (Wireless Fidelity), or a mobile communication network such as 2G (Second Generation), 3G (Third Generation), 4G (Fourth Generation) / LTE (Long Term Evolution), or 5G (Fifth Generation). In an exemplary embodiment, the communication component 716 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 716 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0173] In an exemplary embodiment, the device 700 can be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above method.

[0174] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 704 including instructions. The instructions can be executed by the processor 720 of the device 700 to perform the method provided by the technical solution of the present disclosure. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM (Compact Disc Read-Only Memory), a magnetic tape, a floppy disk, an optical data storage device, etc.

[0175] An embodiment of the present disclosure further provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps of any one of the methods in the aforementioned method embodiments are implemented.

[0176] Through the description of the above embodiments, it can be seen that those skilled in the art can clearly understand that the present disclosure can be implemented by means of software plus a necessary general hardware platform. Based on this understanding, the technical solution of the present disclosure, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of the present disclosure.

[0177] Each embodiment in this specification is described in a progressive manner. The same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on 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. For relevant parts, refer to the partial description of the method embodiment. The system and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without expending creative work.

[0178] The above describes in detail the intelligent interaction method and electronic device provided by the present disclosure. Specific examples are used herein to illustrate the principles and implementation methods of the present disclosure. The description of the above embodiments is only intended to help understand the method and core concept of the present disclosure. At the same time, for those skilled in the art, based on the concept of the present disclosure, there may be changes in the specific implementation methods and application scope. In summary, the contents of this specification should not be understood as limiting the present disclosure.

Claims

1. An intelligent interaction method, wherein, The method includes: During the process of presenting the current page, at least sense the possible shopping needs of the user currently based on the functions provided by the current page and / or the behavioral data generated by the user on the current page; After sensing the possible target shopping needs of the user currently, generate a prompt text for conversing with the artificial intelligence (AI) content generation model according to the target shopping needs; Based on the prompt text, call the pre-trained AI content generation model to generate target content corresponding to the target shopping needs and present it.

2. The method according to claim 1, wherein The method is applied to an artificial intelligence (AI) interaction module, and the AI interaction module is used to be implanted into any page.

3. The method according to claim 2, wherein The AI interaction module exists in the form of a software development kit (SDK) so that the AI interaction module can be implanted into any page through the SDK.

4. The method according to claim 2, wherein The page includes: pages corresponding to multiple nodes on the shopping guidance link in the commodity information service system.

5. The method according to claim 2, wherein The AI interaction module corresponds to a page element for display on the target page. When the shopping needs are not sensed, the page element is in a collapsed or hidden state. After the shopping needs are sensed, the page element is in an expanded or displayed state for interacting with the AI interaction module through the page element.

6. The method according to claim 5, wherein, It further includes: After sensing the target shopping needs, provide an operation option for confirming the target shopping needs in the page element of the target page, so that after receiving the confirmation message from the user, generate a prompt text for conversing with the artificial intelligence (AI) content generation model according to the target shopping needs.

7. The method according to claim 6, wherein Multiple generated prompt texts are provided, and the method further includes: Provide an operation option for selecting multiple prompt texts in the page element of the target page, so that after receiving the selected prompt text from the user, initiate a call to the AI content generation model.

8. The method according to claim 1, wherein The information relied on when sensing the possible needs of the user currently further includes: the asset information of the user in the commodity information service system, the marketing activity information published in the commodity information service system, the marketing activity information associated with the commodity, and the shopping preference information of the user.

9. The method according to claim 1, wherein The functions provided by the current page include: providing commodity recommendation information; The at least sensing the possible needs of the user currently based on the functions provided by the current page and / or the behavioral data generated by the user on the current page includes: Based on the behavioral data generated by the user on the page containing the commodity recommendation information flow, judge whether the user has a clear target commodity for shopping or a clear shopping target range, and sense the possible needs of the user according to the judgment result.

10. The method according to claim 1, wherein the functions provided by the current page include: providing product search information; perceiving the possible current needs of the user based at least on the functions provided by the current page and / or the behavior data generated by the user on the current page, including: judging whether the user has a clear target product for shopping or a clear shopping target range according to the product search condition information input by the user on the current page, and perceiving the possible needs of the user according to the judgment result.

11. The method according to claim 9 or 10, wherein perceiving the possible needs of the user according to the judgment result includes: if it is judged that the user has a clear target product for shopping, then it is perceived that the target shopping need of the user is the provision of information based on the product dimension and / or decision-making assistance content, so as to generate corresponding prompt text, and the AI content generation model generates information provision based on the product dimension and / or decision-making assistance content.

12. The method according to claim 11, wherein the information provision based on the product dimension and / or decision-making assistance content includes: comparison between the same products provided by different merchants in multiple dimensions and / or product recommendation information.

13. The method according to claim 9 or 10, wherein perceiving the possible needs of the user according to the judgment result includes: if it is judged that the user has a clear shopping target range, and the target range is related to a certain target scenario, then it is perceived that the target shopping need of the user is shopping guidance based on the target scenario, so as to generate corresponding prompt text, and the AI content generation model generates product recommendation information in the form of a shopping list aggregated by the target scenario.

14. The method according to claim 13, wherein when providing the product recommendation information in the form of a shopping list, the AI content generation model is further used to generate an atmosphere background image corresponding to the target scenario.

15. The method according to claim 9 or 10, wherein perceiving the possible needs of the user according to the judgment result includes: if it is judged that the user has a clear shopping target range, and the target range includes multiple different products under a certain target category, then it is perceived that the target shopping need of the user is to purchase products in the target category or compare the multiple different products, so as to generate corresponding prompt text, and the AI content generation model generates product recommendation information about the target category, or compares and / or recommends the multiple different products in multiple dimensions.

16. The method according to claim 9, wherein perceiving the possible needs of the user according to the judgment result includes: If it is determined that the user has no clear shopping goal, then based on the user's historical behavior data, it is determined whether the user has a clear shopping driving factor. If so, it is perceived that the user's target shopping need is information related to the shopping driving factor, so as to generate corresponding prompt text, and the AI content generation model generates content aggregated by the shopping driving factor.

17. The method according to any one of claims 1-16, wherein the functions provided by the current page include: providing detailed information about the target commodity; perceiving the possible current needs of the user at least based on the functions provided by the current page and / or the behavior data generated by the user on the current page, including: If the user's stay time on the current page exceeds the threshold or swipes multiple screens, but does not perform the operation of adding to the set of goods to be settled or purchasing, it is perceived that the user's target shopping need is to obtain more comprehensive information about the target commodity, or to obtain the effect information of the target commodity in the desired usage scenario, so as to generate corresponding prompt text, and the AI content generation model summarizes the detailed information of the target commodity to generate a summarized text, or, after the user provides relevant images of the usage scenario, generate the usage effect image when the target commodity is simulated into the usage scenario.

18. The method according to any one of claims 1-17, wherein the functions provided by the current page include: providing a list information of the goods that the user has added to the set of goods to be settled; perceiving the possible current needs of the user at least based on the functions provided by the current page and / or the behavior data generated by the user on the current page, including: Based on the goods in the set of goods to be settled, it is determined whether the user has the need to obtain the price change situation of the goods in the set of goods to be settled, add more goods to the set of goods to be settled to meet the usage conditions of a certain user right, or recommend associated goods for the goods in the set of goods to be settled, so as to generate corresponding prompt text, and the AI content generation model displays the goods with price changes at a prominent position on the current page, or provides goods recommendation information for meeting the usage conditions or belonging to associated goods.

19. A computer-readable storage medium having a computer program stored thereon, wherein, When the program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 18.

20. An electronic device, wherein, including: one or more processors; and a memory associated with the one or more processors, the memory is used to store program instructions, and when the program instructions are read and executed by the one or more processors, they execute the steps of the method according to any one of claims 1 to 18.

21. A computer program product, wherein, including a computer program, and when the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 18.

Images