CN117974260A

CN117974260A - Intelligent interaction method and electronic equipment

Info

Publication number: CN117974260A
Application number: CN202410021667.3A
Authority: CN
Inventors: 董晖; 程斯祈
Original assignee: Hangzhou Alibaba Overseas Internet Industry Co ltd
Current assignee: Hangzhou Alibaba Overseas Internet Industry Co ltd
Priority date: 2024-01-05
Filing date: 2024-01-05
Publication date: 2024-05-03

Abstract

The embodiment of the application discloses an intelligent interaction method and electronic equipment, wherein the method comprises the following steps: in the process of displaying a current page, sensing shopping requirements which a user may currently have at least according to functions provided by the current page and/or behavior data generated by the user in the current page; after sensing a target shopping demand which a user may currently have, generating a prompt text for generating a model dialogue with artificial intelligent AI content according to the target shopping demand; and calling the AI content generation model which is trained in advance based on the prompt text so as to generate and display target content corresponding to the target shopping requirement. According to the embodiment of the application, the solution corresponding to the shopping demand can be intelligently and dynamically provided according to the shopping demand sensing result of the user, and the information acquisition efficiency of the user is improved.

Description

Intelligent interaction method and electronic equipment

Technical Field

The present application relates to the field of information processing technologies, and in particular, to an intelligent interaction method and an electronic device.

Background

The commodity information service system (generally called as an e-commerce platform) is used as a centralized flow scene, and needs to provide comprehensive shopping guide service capability and consider shopping demands of various users. For example, the shopping requirements of the user include that the shopping target is clear, only the shopping range is clear, the problem of goods is sought, and the like, but in the current electronic commerce platform, the user needs to search for goods meeting the requirements of the user by searching or browsing from a recommendation list of a home page, enter a detail page of the goods for detail viewing, then determine whether to purchase, and the like. The whole process needs to obtain more information through multistage operation, so that a user needs to spend a lot of time for commodity searching, shopping guiding and browsing and the like, and the efficiency is low.

Disclosure of Invention

The application provides an intelligent interaction method and electronic equipment, which can intelligently and dynamically provide a solution corresponding to shopping requirements according to the shopping requirement sensing result of a user, and improve the information acquisition efficiency of the user.

The application provides the following scheme:

an intelligent interaction method, the method comprising:

in the process of displaying a current page, sensing shopping requirements which a user may currently have at least according to functions provided by the current page and/or behavior data generated by the user in the current page;

After sensing a target shopping demand which a user may currently have, generating a prompt text for generating a model dialogue with artificial intelligent AI content according to the target shopping demand;

and calling the AI content generation model which is trained in advance based on the prompt text so as to generate and display target content corresponding to the target shopping requirement.

The method is applied to an artificial intelligence AI interaction module, and the AI interaction module is used for being implanted into any page.

The AI interaction module exists in the form of a Software Development Kit (SDK) so as to be implanted into any page through the SDK.

Wherein the plurality of pages comprises: pages corresponding to a plurality of nodes on a shopping guide link in the commodity information service system.

The AI interaction module corresponds to a page element for displaying in the target page, wherein the page element is in a packed or hidden state when the shopping demand is not perceived, and is in an unfolded or display state after the shopping demand is perceived, so as to be used for interacting with the AI interaction module through the page element.

Wherein, still include:

and after the target shopping requirement is perceived, providing operation options for confirming the target shopping requirement in the target page element, so that prompt text for dialogue with an artificial intelligence AI content generation model is generated according to the target shopping requirement after a confirmation message of a user is received.

Wherein the generated prompt text is a plurality of pieces, the method further comprises:

And providing operation options for selecting a plurality of prompt texts in the target page element so as to initiate a call to the AI content generation model after receiving the prompt texts selected by the user.

The information according to which the user can feel the current requirements further comprises: the user has asset information in the commodity information service system, marketing activity information issued in the commodity information service system, marketing activity information related to commodities and shopping preference information of the user.

The functions provided by the current page include: providing commodity recommendation information;

The sensing the possible demands of the user according to at least the functions provided by the current page and/or the behavior data generated by the user in the current page comprises the following steps:

Judging whether the user has an explicit shopping target commodity or an explicit shopping target range according to behavior data generated by the user in the page containing the commodity recommendation information flow, and sensing possible demands of the user according to a judgment result.

The functions provided by the current page include: providing commodity searching information;

Judging whether the user has an explicit shopping target commodity or an explicit shopping target range according to commodity search condition information input by the user on the current page, and sensing possible demands of the user according to a judgment result.

The sensing the possible needs of the user according to the judgment result includes:

if the user is judged to have an explicit shopping target commodity, the target shopping requirement of the user is perceived to be information providing and/or decision-making auxiliary content based on commodity dimension, so that the information providing and/or decision-making auxiliary content based on commodity dimension is generated by the AI content generation model through generating corresponding prompt text.

Wherein the commodity dimension-based information provides and/or decides auxiliary content, comprising: contrast and/or merchandise recommendation information in multiple dimensions between the same merchandise offered by different merchants.

If the user is judged to have an explicit shopping target range, wherein the target range is related to a certain target scene, the target shopping requirement of the user is perceived to be shopping guide based on the target scene, so that commodity recommendation information in the form of a shopping list aggregated by the target scene is generated by the AI content generation model through generating a corresponding prompt text.

And when commodity recommendation information in the shopping list form is provided, the AI content generation model is also used for generating an atmosphere background image corresponding to the target scene.

If the user is judged to have an explicit shopping target range, and the target range comprises a plurality of different commodities under a certain target category, the target shopping requirement of the user is perceived to be that the commodity of the target category is purchased, or the plurality of different commodities are compared, so that commodity recommendation information about the target category is generated by the AI content generation model through generating corresponding prompt text, or the plurality of different commodities are compared and/or recommended in a plurality of dimensions.

if the user has no explicit shopping target, judging whether the user has an explicit shopping driving factor according to the historical behavior data of the user, and if so, sensing that the target shopping requirement of the user is information related to the shopping driving factor so as to generate the content aggregated by the shopping driving factor by the AI content generation model through generating a corresponding prompt text.

The functions provided by the current page include: providing detailed information about the target commodity;

if the residence time of the user in the current page exceeds a threshold value or a plurality of screens are slid, but no operation of joining a to-be-settled commodity set or purchasing is performed, sensing that the target shopping requirement of the user is to acquire more comprehensive information about the target commodity or to acquire effect information of the target commodity under a desired use scene, summarizing the detail information of the target commodity by generating a corresponding prompt text through the AI content generation model, generating a summarized text or generating a use effect image when the target commodity is simulated into the use scene after the user provides a related image of the use scene.

The functions provided by the current page include: providing commodity list information that the user has added to the commodity set to be settled;

Judging whether a user has the condition of acquiring price variation of the commodities in the commodity set to be settled or not according to the commodities in the commodity set to be settled, adding more commodities into the commodity set to be settled to meet the use condition of the rights of a user or the requirement of recommending the related commodities with the commodities in the commodity set to be settled, so that the commodities with the price variation are presented by the AI content generation model at the position before the current page by generating the corresponding prompt text, or providing commodity recommendation information for meeting the use condition or belonging to the related commodities.

An intelligent interaction device, comprising:

the demand sensing unit is used for sensing shopping demands which a user can currently have according to at least functions provided by the current page and/or behavior data generated by the user in the current page in the process of displaying the current page;

A prompt text generation unit, which is used for generating a prompt text for dialogue with an artificial intelligence AI content generation model according to target shopping requirements after sensing the target shopping requirements which the user may currently have;

and the content generation unit is used for calling the AI content generation model which is trained in advance based on the prompt text so as to generate and display target content corresponding to the target shopping demand.

A computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the method of any of the preceding claims.

An electronic device, comprising:

one or more processors; and

A memory associated with the one or more processors, the memory for storing program instructions that, when read for execution by the one or more processors, perform the steps of the method of any of the preceding claims.

According to the specific embodiment provided by the application, the application discloses the following technical effects:

According to the embodiment of the application, in the process of displaying the current page, the current possible demand of the user can be perceived according to the function provided by the current page and/or the behavior data generated in the current page by the user, after the current possible target shopping demand of the user is perceived, a prompt text for dialogue with an AI (ARTIFICIAL INTELLIGENCE ) content generation model can be generated according to the target shopping demand, and then a pre-trained AI content generation model can be called based on the prompt text to generate target content corresponding to the target shopping demand and display the target content. In this way, the user may have a target shopping demand perceived, and based on the perceived target shopping demand, an automatic prompt text is generated, and content for solving the corresponding problem is generated, so the method belongs to a demand-responsive active touch AI interaction method. In the mode, the user is not required to initiate interaction with the AI, but the AI interaction module is used for actively initiating the interaction, and the prompt text for carrying out the dialogue with the AI model can be automatically constructed, so that the difficulty and threshold for the user to interact with the AI are reduced, and the user experience is improved. And the solution corresponding to the shopping demand can be intelligently and dynamically provided according to the sensing results of different shopping demands of the users, so that the method is flexibly adapted to various user demands and scenes, and the information acquisition efficiency of the users is improved.

In an alternative implementation manner, the AI interaction module may be provided in the form of an SDK (Software Development Kit ) or the like, and the AI interaction module may be implanted into a plurality of different pages, for example, may be pages corresponding to a plurality of nodes on a shopping guide link in a merchandise information service system, so that the AI interaction module is unbinding with a specific page, and is convenient for providing a unified interaction experience across scene pages for a user.

In addition, the method is more friendly and convenient for user groups such as new users unfamiliar with the operation of the e-commerce system. Furthermore, by providing support for interactive forms of voice, gestures, video, etc., a more friendly experience may also be provided for user groups such as visually impaired people.

Of course, it is not necessary for any one product to practice the application to achieve all of the advantages set forth above at the same time.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic diagram of a system architecture provided by an embodiment of the present application;

FIG. 2 is a flow chart of a method provided by an embodiment of the present application;

FIGS. 3 through 6 are schematic diagrams of user interfaces provided by embodiments of the present application;

fig. 7 is a schematic diagram of an electronic device according to an embodiment of the present application.

Detailed Description

The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which are derived by a person skilled in the art based on the embodiments of the application, fall within the scope of protection of the application.

In the embodiment of the application, AIGC (ARTIFICIAL INTELLIGENCE GENERATED Content) technology can be utilized to help users of the commodity information service system to improve the efficiency of acquiring information. Among them, so-called AIGC is to produce some content including text, images, video, etc. using the AI content production model. Under the promotion of AI big model wave tide, AI promotes fast in aspects such as natural language processing, complex scene understanding, combines mass data and algorithm model optimization, has had the relatively firm bottom basis in the scene of actually falling to the ground. The embodiment of the application provides an application mode of AI in the electronic market scene.

It should be noted that, some application systems have proposed implementation of applying the AI large model to the e-commerce scenario, but there are some problems in specific implementation. For example, in the existing application manner, a resident fixed interaction entrance is generally provided in a specific page, and a user can enter an AI interaction interface through the entrance at any time in the process of browsing the page, and then can converse with the AI by asking the AI, and accordingly, the AI can generate corresponding content for the problem posed by the user.

The problems with the above implementations are at least:

the user is required to actively initiate interaction with the AI and to raise questions to the AI by himself, but in practical applications, it may be difficult for many users to realize whether assistance can be obtained by the AI or when to initiate interaction with the AI and how to raise questions to the AI. In particular, in how to ask the AI, many users are not aware of the question, i.e., the users have difficulty in clearly describing their own needs. In addition, for some potential needs, the user may not be able to actually recognize themselves, e.g., some users may be accustomed to purchasing some relatively low priced goods, but the user may not perceive themselves, etc. This means, therefore, that it is difficult for the user to construct high quality Prompt text (Prompt) for a dialogue with the AI large model, which results in difficulty in obtaining the truly desired content by way of AI interaction.

In addition, specific AI interaction functionality is provided by specific pages, i.e., users can interact with AI only in such pages. For example, assuming that an AI interaction portal is provided in the client home page, the user can interact with the AI only during browsing of the home page. However, there are usually multiple shopping guide links provided in the e-commerce platform, each shopping guide link may include multiple nodes, and multiple different pages are respectively involved, and if only part of the pages implement the AI interaction function, this means that a user cannot interact with the AI when accessing other pages. If the AI interaction functions are required to be respectively implemented in each page, on one hand, the workload is high, and on the other hand, because different pages may correspond to different development teams, the AI interaction functions respectively implemented are difficult to be unified, and users may need to be respectively familiar with the interaction modes in the different pages, so that new troubles are brought to the browsing process of the users.

Aiming at the problems, in the embodiment of the application, the application mode of the AI large model in the e-commerce platform is improved, specifically, a demand-responsive AI interaction 'robot' module can be implanted in the e-commerce platform, the module can be implanted in a page (such as a front page, a channel page, a meeting place page, a detail page, a shopping cart page, a compact single page and the like) of a full-link scene of user shopping in the form of an independent plug-in, so that the user can obtain the assistance of AI in a plurality of different pages, and the aspects of interfaces, interaction logic and the like can be kept uniform without specific development work of developers of each page. In addition, the 'AI' interactive robot can sense shopping requirements or demands and the like possibly possessed by a user based on the type or function of a specific page, behavior data actually generated in the page by the user and the like, and then can intelligently generate prompt texts for dialogue with the AI, and can generate specific consumption suggestions or solutions corresponding to specific shopping requirements/demands through an AI content generation model under the condition that the user approves the prompt texts, so that the interactive robot flexibly adapts to various user requirements and business scenes. That is, the "demand-responsive" AI interaction means that, instead of providing a resident interaction portal directly in a certain page and initiating a question by a user, the shopping demand of the user can be perceived in any page according to information such as the type and function of the page and behavior data generated by the user in the page at present, and if the user can be perceived as possibly having a certain shopping demand, the AI can initiate an interaction with the user and provide a dynamically changing dialogue prompt and generate content according to a specific perception result. Of course, in another implementation, if consistency of AI interaction experience between different pages is not required to be considered, the AI interaction modules of the "demand response" may also be implemented directly on different pages respectively.

The specific AI interaction module may further correspond to a page element for displaying in the target page, where the page element may be in a collapsed or hidden state when a shopping demand of a user is not sensed, and may be switched to an expanded or display state after the shopping demand is sensed, so as to be used for interaction with the AI interaction module through the page element.

Specifically, after the shopping requirement of the user is sensed, the AI can actively generate a corresponding prompt text according to the shopping requirement of the user, and of course, in an optional manner, the user can be requested to confirm the sensed shopping requirement. For example, the prompt may be presented via the aforementioned page elements, including "whether you need to prepare a list of babies for christmas," and so on. If the user approves the perception result, corresponding prompt text can be generated, and an AI content generation model is called for content generation based on the prompt text. In this way, the specific prompt text is generated based on the shopping requirement of the user, so that the specific generated content also corresponds to the shopping requirement, and the user is not required to manually construct the specific prompt text, so that the difficulty or threshold of the user to interact with the AI can be reduced, and the intelligence in the AI interaction process is further improved. The content specifically generated by the AI content generation model may still be displayed through the foregoing page element, and of course, the size of the page element may be dynamically scaled and adjusted according to the amount of content specifically required to be displayed, or may even be enlarged to a full-screen state, or may jump to another receiving page to display the content generated by the specific AI content generation model, and so on.

From the technical implementation point of view, as described above, the support of the above functions may be developed by each page, or in order to reduce the development cost of the page developer, a unified SDK (Software Development Kit, a software development kit, i.e., a kit provided by a service side for implementing a function of product software) may be further provided, so that the above functional modules may be implanted into a specific page through such an SDK. Through the mode, the AI interaction function is not strongly coupled with a specific page, a specific page developer does not need to pay attention to specific implementation of the function, so that development cost of the page developer is reduced, and meanwhile, a specific intelligent AI interaction interface, interaction form and the like can be realized to be uniformly expressed among a plurality of different pages, so that scene-crossing and page-crossing uniform experience is provided for a user.

From the system architecture perspective, referring to fig. 1, in a specific implementation manner, the embodiment of the present application may provide the AI interaction module in the form of an SDK (or in the form of SaaS (Software as a service) AS A SERVICE) or the like, so that a specific AI interaction model may be implanted into different pages corresponding to a plurality of nodes on a specific shopping link by the SDK or the like, so that the pages may have a demand-responsive AI interaction function. In the demand response type AI interaction process, the specific AI interaction module can acquire related information of a current page, can acquire information such as behavior data generated in the current page by a user behavior data collection module, can acquire asset information of the user in a current commodity information service system by a user asset data center of a server side, can acquire marketing activity information issued in the commodity information service system from a marketing activity center, can acquire marketing activity information related to commodities by the commodity data center, can acquire shopping preference information of the current user and the like from a user personalized preference data center, and can sense shopping demands of the user by combining the information. After the target shopping demand is perceived, prompt text for dialogue with an artificial intelligent AI content generation model can be generated so as to call a pre-trained AI content generation model based on the prompt text to generate target content corresponding to the target shopping demand and display the target content. The specifically perceived shopping demand can be confirmed by the user, and corresponding prompt text generation and content generation are performed and the generated content is displayed under the condition that the user approval is obtained.

From the application scene, the main application scenes of the demand-responsive intelligent AI interactive function provided by the embodiment of the application can be divided into a pre-sale shopping scene and an after-sale service scene. In the aspect of shopping before sale, compared with the traditional online shopping flow, the demand-response intelligent AI interactive function builds corresponding capability for real-time response, complex scene understanding, user intention recognition and deep personalized shopping in the shopping flow, and particularly under the conditions of smaller wireless end and mobile end screens and limited display information, more accurate information and commodity touch can be provided through simple man-machine interaction. In an after-sales scenario, the demand-responsive AI assistant may also provide the user with services such as logistic inquiry, refund returns, dispute maintenance, etc. in the form of a dialogue.

The following describes in detail the specific implementation scheme provided by the embodiment of the present application.

Firstly, the embodiment of the application provides an intelligent interaction method, which can be applied to an AI interaction module in one implementation mode, wherein the AI interaction module can be specifically used for being implanted into any page, so that a plurality of different pages can have a demand-responsive AI interaction function, and unified experience of crossing scenes and pages is provided for a user. For example, a particular arbitrary page may include any of a plurality of pages of the same application, and may be populated in a plurality of pages. For example, the AI interaction module may be implanted into a plurality of pages corresponding to a plurality of nodes on a shopping guide link in the merchandise information service system, and so on. In an alternative manner, the AI interaction module may exist in the form of an SDK, such that a particular AI interaction module is implanted into a plurality of different pages via the SDK. Or the AI interaction module can also exist in the form of SaaS and the like, and the mode of the SDK plug-in or the SaaS can not only be quickly implanted into a plurality of pages of each e-commerce platform, but also be implanted into pages of systems such as an AI large model plug-in, social software and the like, so that the system has strong expansibility and reusability. It should be noted that, whether in the form of SDK or SaaS, the AI interaction module and the specific page in the embodiment of the present application may be independent from each other, that is, there is no strong coupling relationship between the AI interaction function and the specific page, different pages may be implanted with the same AI interaction module, and the AI interaction module may interact with the user in different pages according to the same data processing logic and front-end display policy.

Specifically, referring to fig. 2, the method may specifically include:

S201: and in the process of displaying the current page, sensing the possible current requirements of the user at least according to the functions provided by the current page and/or the behavior data generated by the user in the current page.

The current page, that is, the page currently browsed by the user, may specifically be a page corresponding to any node on any shopping guide link in the merchandise information service system, for example, may include a client front page, a merchandise recommendation page, a search page, a channel page, an event venue page, a store page, a merchandise detail page, and may further include a "shopping cart" page, a "play list" page, a page related to after-sales service, and so on.

In the embodiment of the application, instead of providing the AI interaction portal directly in a specific page, the user may actively interact with the user to respond to a shopping demand if the user may have a shopping demand after analyzing behavior data of the user, etc., and thus is referred to as demand-responsive AI interaction.

In particular, at least the functionality provided to the current page and/or the behavior data generated by the user in the current page may be used when perceiving the shopping needs of the user. That is, in the embodiment of the present application, since the AI interaction function may be provided in the pages corresponding to the plurality of nodes in the full link, and when the user browses in different pages, the specific page may provide different functions, so that the requirements may also be different. In addition, even in the same page, if different behavior data are generated, the expressed user requirements are different, so that, in specific implementation, the user can sense whether the browsing process of the user has a certain shopping requirement or not by combining the functions of the page itself, the behavior data generated by the user in the current page, and the like. Of course, in specific implementations, the information upon which the user may currently have a need is based upon may also include: asset information that the user has in the merchandise information service system (e.g., whether there is a "red pack" that has not yet been approved, etc.), marketing campaign information that is published in the merchandise information service system, marketing campaign information associated with merchandise, shopping preference information for the user (e.g., whether it is a user that is relatively sensitive to price attributes, whether certain brands are preferred, etc.). By integrating the information in multiple aspects, the shopping demands of the user can be perceived more accurately.

In particular, when sensing the shopping demand of the user, there may be multiple implementations, for example, in one way, some rules may be preset, which are as follows: if the user generates certain behavioral data in a certain page, it may be considered that the user may have certain shopping needs. For example, if the user clicks and views the detail page of a certain mobile phone before returning to the commodity recommendation page and clicks and views the detail page of another mobile phone of the same type in the process of sliding and browsing the specific recommended commodity in the commodity recommendation page, it may be determined that the user may have a need to purchase the mobile phone of the same type based on the behavior data. Or in the process of sliding and browsing specific recommended commodities in the commodity recommendation page, clicking and viewing detailed information of a plurality of commodities, wherein the commodities are related to a certain scene, and then it can be presumed that the user possibly has shopping requirements related to the scene. Or if the user inputs the brand, model and other information of a specific mobile phone in the commodity searching process, the user has a clear requirement of purchasing the mobile phone; or if only a keyword of "christmas" or the like is entered, it means that the user may have a need to purchase goods related to christmas scenes without being specific to a certain or a certain type of goods, or the like. In this way, it is possible to perceive whether a user has some target shopping need by matching with such rules after the behavior data is collected.

Alternatively, the sensing of the shopping demand of the user may be performed by an AI large model, specifically, information such as the type and function of the page, information such as behavior data generated in the page by the user, etc. may be input into the AI large model, and the AI large model senses whether the user has a certain shopping demand by understanding specific data. Of course, in this way, asset information of the user in the merchandise information service system, marketing campaign information issued in the merchandise information service system, marketing campaign information associated with merchandise, shopping preference information of the user, and the like may also be input into the AI large model for perception of shopping needs that the user may have at present. In addition, in the concrete implementation, the AI large model can be trained in advance through a plurality of training samples, and parameters of the AI large model are adjusted so that the AI large model has the capability of understanding the information and perceiving shopping demands of concrete users.

Or may also combine pre-established rules with the perception of AI large models. In specific implementation, classification summary can be performed in advance according to possible demand conditions of users in various pages so as to complete rule formulation or training of an AI large model and the like. The classification summary and the perception of the user demands are described below by taking a client-side front page (which can be used for recommending commodities in general), a search page, a shopping cart page, a commodity detail page and the like as pages in a shopping scene before sale as examples.

1. Client first page: the home page is used as a centralized flow scene in the public domain, and various users can distribute via the home page, so that the home page generally needs to consider shopping requirements of various users, for example, the home page specifically can comprise:

1. explicit demand for shopping target merchandise: that is, there is clear appeal to brands and commodities, the commodities need to be found out quickly, commodity price, preferential, contrast and other information are obtained, and a decision is made. In this case, information provision based on commodity dimensions and shopping guide decision assistance can be performed to the user. However, the home page in the prior art is not generally realized by the direct demands of users, and the users need to enter a search, a detail page and the like, and obtain more information about commodities through clicking operation.

2. Only a clear shopping range: that is, there are clear shopping scenes, required categories/brands, but there is no specific required commodity, and it is necessary to find a commodity meeting the demand first. In this case, scene-based shopping guide guidance can be provided for the user, and the user needs to know the commodity recommendation more accurately later. However, the home page in the prior art generally does not have this capability, and requires the user to spend time in commodity searching and shopping guide browsing.

3. No shopping goals at all: at this time, specific requirements may also be different according to the user driving factors. For example, for a discount/promotion driven user, it may be desirable to learn platform promotional information, and so forth. For this case, the user may be provided with clearer promotional information and can be associated with user preferences. However, in the prior art, the information about promotion angle aggregation is less in pages such as the first page, and is usually scattered in each service module of the platform, so that users are required to find the information autonomously.

Or for a "grass" browse-driven user, may want to know if there is a good recommendation. At this time, more good recommendations and content access may be provided to the user. In the prior art, a certain good recommending scene can be met through the recommending card of the home page, but the better interesting and rich good recommending is lacked, and the requirements of users cannot be met.

In addition, some users may need to conduct living consultations to learn about living problems, such as "white shoes are easily blackened, and what is done). At this time, life answering can be performed from a specific scene, and meanwhile, due to the possibility of shopping conversion, some related commodity recommendation and the like can be performed. However, the home page in the prior art does not meet similar requirements, users obtain suggestions in a content field, and go to a shopping platform for ordering after 'grass planting'.

On the basis of summarizing and analyzing the possible demands of the user in the browsing process of the home page, particularly in the browsing process of the user on the page of the home page, whether the user has an explicit shopping target commodity or an explicit shopping target range can be judged according to behavior data generated in the page by the user, and then the possible demands of the user are perceived according to a judging result.

For example, suppose that in the process of sliding and browsing a specific recommended commodity in a first page including a commodity recommendation information stream, a user clicks and views a detail page of a certain mobile phone, then returns to the commodity recommendation page, and clicks and views a detail page of another mobile phone of the same type. Based on the behavior data, it can be determined that the user has explicit needs for the shopping target commodity, specifically, that the user may want to purchase the mobile phone. Further, it may be perceived that the target shopping requirement of the user may be information providing and/or decision-making assistance content based on commodity dimensions, for example, a detailed description of the mobile phone, or a summary of user evaluation situations of the mobile phone, etc., so as to make a shopping decision.

Or in the process of sliding and browsing specific recommended commodities in the commodity recommendation page, clicking and viewing detailed information of a plurality of commodities, wherein the commodities are related to a certain scene, it can be presumed that the user possibly has a shopping demand user related to the scene, namely, the user belongs to the condition of only having an explicit shopping range. At this time, the specifically perceived demand may be to acquire shopping guide information based on the target scene. For example, if a plurality of items clicked and viewed by the user are all related to "christmas", the user may be perceived to have a need to conduct shopping guidance based on the "christmas" scenario, and so on.

Further, or assuming that the user is sliding through the recommended product card in the product recommendation page, but is not always performing operations such as clicking to view details, at this time, it can be inferred that the user is currently "strolling" in the page and has no explicit target (neither explicit product nor explicit scene, brand, etc. range). In this case, it may be presumed that the user belongs to a situation without an explicit target, and when the demand sensing is specifically performed, the user may comprehensively sense the demands that the user may have in combination with the asset information that the user has in the merchandise information service system, the marketing campaign information issued in the merchandise information service system, the marketing campaign information associated with the merchandise, and the shopping preference information of the user (may specifically include main driving factors, for example, include a user that the price-sensitive user may belong to a price/promotion-driven type, or may further include a user of a "grass" driven type, etc.). For example, it may be determined from historical behavioral data of a user whether the user has an explicit shopping driver, and if so, it may be perceived that the user's target shopping needs are information related to the shopping driver, including, for example, the aforementioned information related to reduced-price promotional campaigns, or "grass" related content, and so forth.

2. Search page: for search scenarios, users typically have explicit shopping appeal, including explicit scenarios, category ranges, or explicit concerns about merchandise, etc. For example, specific examples may include:

1. scene-based search requirements: at this time, generally, a fuzzy search of an input keyword, such as: christmas dressing, cell phones, pants, etc. In this case, the user can be helped to define a specific scene atmosphere, and a commodity recommendation list and the like conforming to shopping habits of the user can be formed. However, the prior art does not have the capability of continuous interaction, and only the user can continuously adjust the keywords, so that the decision period is long.

2. Search requirements based on merchandise: an accurate search is typically performed for an input keyword, such as "a certain brand of a certain model of mobile phone," and so on. At this time, the user may need to acquire more information about the merchandise, merchandise function/rule introduction, price, service of the store and specific merchandise, evaluation, delivery place, performance information, and the like. In the prior art, screening of commodity dimensions can be supported, but the screening and functional comparison are required to be performed manually by a user, and no shopping guide proposal is provided in the SKU (Stock Keeping Unit, minimum stock unit) dimension.

For such search type pages, in the embodiment of the present application, whether the user has an explicit shopping target commodity or an explicit shopping target range can be determined according to commodity search condition information input by the user on the current page, and the possible demands of the user are perceived according to the determination result.

For example, assuming that the user enters a search keyword of "christmas", it may be perceived that the user's need may be to obtain a list of items related to the "christmas" scenario, or assuming that the user enters a search keyword of "a certain brand of mobile phone", it may be perceived that the user's need may be to obtain a comparison of the same type of merchandise related to the merchandise, etc., to aid the user in making a selection decision.

3. Item detail page: the commodity detail page is mainly used for providing information such as functional introduction, color/rule shopping guide information, business service condition, commodity evaluation, distribution performance and the like of a specific commodity. In browsing such pages, a user may typically have the option of acquiring more information about the specific merchandise, merchandise function/rule introduction, price, store and specific merchandise services, evaluation, shipping location, performance, etc., or may want to learn about the use of the specific merchandise in the desired scenario for some specific categories of merchandise if no further operations such as joining a "shopping cart" or ordering a purchase are performed late. For example, apparel, furniture, etc., a user may want to know the upper body effect of a particular apparel on himself or herself, or to match the style of room finishing in his home, other furniture, etc., prior to ordering. In the prior art, a user can acquire more comprehensive information only after finishing looking up the whole content in a page, or can only contact customer service personnel of a store in the aspects of code number recommendation, delivery time information and the like, but can be influenced by customer service resources of a merchant, on-line time length, time difference and the like.

In the embodiment of the application, if the residence time of the user in the current page exceeds a threshold value or a multi-screen is slid, but no to-be-settled commodity set (i.e. shopping cart) or purchasing operation is performed, the target shopping requirement of the user is perceived to possibly acquire more comprehensive information about the target commodity or acquire the effect information of the target commodity under the expected use scene.

4. The to-be-settled commodity collection page, namely the shopping cart page, is usually more concerned about sales promotion, discount, price reduction and other conditions of the purchased commodities and the conditions of full reduction of the sales slip during the sales promotion when a user accesses the page, so as to make a final purchase decision. In this case, the user's target shopping needs may generally include: price change of the purchased goods, price reduction reminding when the purchased goods reach the expected price, recommendation of the ordered goods, associated recommendation of the purchased goods, and the like.

Therefore, in the embodiment of the application, according to the commodity information in the commodity set to be settled, including the condition that whether the commodity participates in a sales promotion activity or not when the commodity is compared with the current price of the commodity when the commodity is added to the shopping cart, whether the user has the price change condition of acquiring the commodity in the commodity set to be settled, adding more commodities to the commodity set to be settled so as to meet the use condition of the rights of a user, or carrying out related commodity recommendation with the commodity in the commodity set to be settled or the like can be judged.

Regarding relevant pages of the after-sales scenario, for example, a logistics inquiry page, the user's requirement may be directly determined to obtain information of logistics status, logistics details, etc., or if some special status occurs, for example, is detained at customs, etc., a corresponding solution may also need to be obtained, etc. Or for reverse transaction pages such as refund, the user demand can be specifically perceived according to the state of the transaction, for example, if a refund request is not submitted, the user may need to obtain description information related to refund description, or if a refund request is submitted, the user may want to know refund progress, and so on.

S202: after sensing a target shopping demand which a user may currently have, generating prompt text for dialogue with an artificial intelligence AI content generation model according to the target shopping demand.

After perceiving that the user may have some target shopping need at the present time based on the information of the page, the behavior data of the user, etc., a prompt text for a model dialog with the artificial intelligence AI content generation model may be generated based on the target shopping need. Of course, in a specific implementation, an operation option for confirming the perceived target shopping requirement may also be provided through the page element corresponding to the AI interaction module, and the user may generate the prompt text after completing the confirmation operation.

For example, in the home scenario, if the user is perceived to have an explicit shopping need for a particular item, a question for confirmation to the user may be generated, such as "whether you are looking for a particular item" and may be presented through the aforementioned page element, and if the user confirms by selecting "yes" or clicking directly on the question, etc., a specific prompt text may be generated for input into the AI big model for content generation. Of course, in a specific implementation, the prompt text may be generated first, and after the AI big model performs the preliminary content generation, the user may be checked again, for example, if it is also assumed that the user has an explicit shopping requirement for a specific commodity, the corresponding prompt text may be generated first, and the AI big model generates the content related to the commodity, to determine whether the commodity is sold by multiple merchants, whether some merchants are developing a sales promotion for the commodity, and so on. Thereafter, text for confirmation to the user is regenerated, for example, "whether you are looking for a certain commodity, 10 money is now being reduced in price", and so on. Thus, the user can be perceived to be likely to find a certain commodity, and meanwhile, the information of obvious benefit points such as 10 money in price reduction can be provided for the user, so that the interest of the user in detail content can be promoted more favorably, and the like.

After the user confirms the perceived target shopping requirement, a corresponding prompt text may be generated for the specific target shopping requirement, for example, for the foregoing example, the generated prompt text may be: "check which of the same items are on sale, which of them are on the part of the promotional program, compare and give recommended purchase advice from the dimensions of price, performance, logistics, after-sales, etc., respectively", etc. Such hint text may be used for input into the AI content generation model for generating specific content.

That is, in the process of content generation using the AI content generation model, the construction of the presentation text is critical, and in order to obtain high-quality content, first, the high-quality presentation text is constructed. In the traditional mode, a user constructs a prompt text according to the own requirements, and then inputs the prompt text into an AI large model to generate content. However, in the embodiment of the present application, considering that there may be a high difficulty in making a normal user construct a high-quality prompt text, it may be difficult for the user to clearly describe his own needs in the prompt text, and in addition, the process of constructing the prompt text may occupy the time of the user, so that a scheme for helping the user construct the prompt text is provided. In other words, in the embodiment of the application, not only can some content be intelligently generated for the user through the AI large model, but also even the prompt text can be automatically generated. From another perspective, the input information for the AI large model is no longer the traditional prompt text entered by the user, but may include the type of page, the function, the behavior data of the user in the page, even the asset information of the user in the current system, the marketing campaign situation in the system, the marketing campaign situation associated with the merchandise, the personalized preference information of the user, and so on. Of course, in specific implementation, the above information may be preprocessed first to generate specific prompt text, and then input into the AI big model to generate content.

S203: and calling the AI content generation model which is trained in advance based on the prompt text so as to generate and display target content corresponding to the target shopping requirement.

After a specific prompt text is generated, a 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 requirement. In the case of various different pages, different user behavior data and the like, the generated prompt text is different due to different user requirements, and the generated content is also different.

For example, if the user currently browses a home page or a search page, and determines that the user has an explicit shopping target commodity, the user's target shopping requirement can be perceived to provide and/or make a decision to assist content based on commodity dimension information, at this time, a prompt text corresponding to the requirement can be generated, and accordingly, the AI content generation model can generate commodity dimension-based information to provide and/or make a decision to assist content.

Wherein, the specifically generated information providing and/or decision-making auxiliary content based on commodity dimension can comprise: summary of details information, evaluation information and other contents of the commodity, or comparison and/or commodity recommendation information among the same commodity provided by different merchants in multiple dimensions. With respect to the latter, when a user explicitly needs to purchase a certain item, since it is likely that multiple merchants are selling the same item of the item in the current system, it is important for the user to choose from among the merchants, and it is often necessary to compare the same items offered by different merchants in multiple dimensions when making the choice. In the prior art, a user can only click and view detail pages of all the same commodities respectively, and acquire related information for comparison. The user can directly view the comparison results of the plurality of same-style commodities from the same page, and can respectively compare the same-style commodities from a plurality of dimensions, or can also give purchase suggestions as a whole, and the like. In this way, the user can make a selection according to a dimension or the like in which he or she is more concerned. For example, after comparing the same type of commodity from a plurality of different dimensions such as price, logistics, after-sales, and evaluation, the commodity recommended to be purchased in each dimension can be respectively given, so that if the user pays more attention to a certain dimension during shopping, the commodity recommended in the dimension can be selected. For example, the user may be compared to the price dimension, may select an item that is advantageous in the price dimension, may be more advantageous in the logistics dimension if the user is more willing to receive the item as soon as possible, and so on. In this way, the user can be helped to more quickly complete the comparison of multiple items of the same money and make shopping decisions.

For example, as shown in fig. 3 (a), it is assumed that the user inputs "a brand X1 model cell phone" through the search page and initiates the search, which belongs to an accurate search, that is, it is possible to perceive that the user has an explicit shopping target commodity demand, that is, the "a brand X1 model cell phone". At this point, an operational option for confirming the perceived result may be provided at the location shown at 31, where: "do you want to know the detailed description of the phone model a brand X1? ". If the user clicks on the option, the user may enter an AI interaction page, as shown in fig. 3 (B), where several alternative examples of questions may be provided, for example, "what is the camera of the type a brand X1 mobile phone? "how long the battery life of the mobile phone is, the type A brand X1 model," the type A brand X1 model mobile phone with high evaluation, and so on. The user can select specific questions from the questions according to actual requirements to initiate questions to the AI assistant, and the questions can be used as prompt texts for interacting with the AI content generation model. For example, what is the user selected "what is the camera of the mobile phone model a brand X1? The content generated by the AI assistant can be shown in fig. 3 (C), and the overall evaluation of the camera performance of the mobile phone can be performed, for example, the mobile phone belongs to the "middle upper level", and in addition, the mobile phone can be respectively illustrated through several key points, and the like. The content can be generated by the AI content generation model after understanding according to the detailed description information, attribute parameter information, user evaluation information and the like of the mobile phone.

If the user has an explicit shopping target range according to the browsing behavior of the user in pages such as a home page or keywords input in a search page, whether the target range is related to a certain target scene can be further judged, if so, the target shopping requirement of the user can be perceived as shopping guidance based on the target scene, at the moment, after a corresponding prompt text is generated, commodity recommendation information in the form of a shopping list aggregated by the target scene can be generated by the AI content generation model.

For example, assuming that the user browses a plurality of recommended products in the home page, wherein a plurality of products are related to "home improvement", it may be determined that the target range of the user may be related to the scene of "home improvement", and thus it may be perceived that the target shopping demand of the user may be shopping guide based on the scene of "home improvement". At this time, as shown in fig. 4 (a), it is possible to provide information about "is home improvement creative sought" in the page element corresponding to the AI interaction module? And (3) the interactive contents, if the user clicks the page element, proving that the user possibly approves the perception result, further, generating a corresponding prompt text, and generating contents related to shopping guide of a home decoration scene through an AI content generation model. In the front page, the page element corresponding to the AI interaction module may be enlarged, and content related to shopping guide of the "home decoration" scene may be displayed therein. For example, as shown in fig. 4 (B), the specifically generated content may include merchandise recommendation information in the form of shopping lists aggregated by "home improvement" scenes. Specifically, the classification recommendation may be performed according to living rooms, bedrooms, kitchens, etc., and so on. In addition, if the user may have other needs as perceived based on previous user behavior data or the like, the presentation may be performed through the page element, for example, as shown in fig. 4 (B), "you may ask me the following questions: is there other offers? Is you looking for children's garments? "and the like.

It should be noted that, in the existing manner, the user may search the commodity information related to a certain scene through the search function, but under the search logic, the search result is usually given through the keyword matching manner, and in addition, multiple complicated search result sorting mechanisms are involved, so that although the user can view the search result related to the keyword of a specific scene in the search result page, it is difficult to quickly purchase the required commodity. For example, after entering the keyword "Christmas", the search result may include a plurality of items related to "Christmas", wherein the plurality of items may be all Christmas trees in succession, or the plurality of items may be all Christmas socks in succession, and so on. However, in the embodiment of the present application, if the user is found to have shopping requirements related to a certain scene, commodity recommendation information in the form of shopping list aggregated by the target scene may be generated by the AI content generation model. For example, it may be a Christmas tree, christmas socks, bell, wreath, etc., each category may correspond to one or more recommended merchandise, so that the user may complete the purchase of merchandise associated with Christmas directly according to the listing in the page, etc. In addition, in a preferred real-time manner, the specific AI content generation model may also generate an atmosphere background image corresponding to the target scene, specifically when providing the commodity recommendation information in the shopping list format. For example, when generating the shopping list related to the Christmas scene, a background image with the Christmas atmosphere can be generated at the same time, so that the displayed content generation result has scene atmosphere, and the user experience is improved.

In addition, in the case that the user has an explicit target range, the target range may be related to a certain category, for example, the user may browse a plurality of different commodities under a certain category in the current page, and then may perceive that the target shopping requirement of the user may be to purchase the commodity under the target category, or compare the plurality of different commodities under the target category. At this time, the AI content generation model may also generate product recommendation information about the target class, or compare and/or recommend the plurality of different products in a plurality of dimensions by generating a corresponding prompt text.

For example, assuming that the user browses the details page of one wireless earphone as shown in fig. 5 (a) and then browses the details page of another wireless earphone as shown in fig. 5 (B) during browsing the recommended merchandise in the home page, it may be determined that the user has a relatively clear target range, that is, it is related to the category of wireless earphone, and further it may be determined that the user may need to compare the two wireless earphones, at this time, a "do they compare? "information, and thumbnail images of the two types of headphones can be displayed at the same time, so that the user can refer to the information conveniently. After the user clicks on the page element, the page element may be enlarged to reveal AI-generated content, as shown in fig. 5 (C). Specifically, the method can comprise comparing two commodities from the dimensions of price, model, service, score and the like.

In this case, when the user is perceived to have a clear target range, the user may initiate the AI interaction with the user, and then further learn about the user's needs by performing a dialogue with the user, and then perform more accurate commodity recommendation. For example, as shown in fig. 6, where 6 (a) shows a client top page, and the user perceives from user behavior data or the like that the user has a need to "buy a high-performance mobile phone", an operation option for initiating a query to the user may be provided at a position shown therein 61, for example, "do you want to find a high-performance mobile phone? ". If the user clicks on the operation option, the AI interaction interface shown in fig. 6 (B) may be entered, where multiple candidate questions may be provided based on previously obtained behavior data of the user on the current page, or the like, and the user may select one of the questions to initiate a question to the AI assistant, or may input a specific question through an input box below. Assume that the question that the user actually inputs is "can help me recommend a high performance cell phone? By "AI may further define the user's needs by talking to the user. For example, the AI assistant may ask the user: what is you mainly doing with this handset, office/study, entertainment/social, or photo/game? ", if the user answers: "Main user plays games, so I want performance better"; thereafter, the user may also be asked: "do you ask you about what is probably? ", the user can answer: "no more than $350", etc. Thus, the user can know that the user wants to purchase a mobile phone for playing games, and the most important thing is that the performance of the mobile phone is as high as $350, so that more accurate recommendation can be made to the user based on the information. For example, the content generated by the specific AI assistant may be as shown in fig. 6 (D), where three recommended handsets are included, and respective recommendation reasons may be given respectively, and the user may make further selection based on the recommendation result.

In addition, if it is determined that the user does not have an explicit shopping target, it may be determined whether the user has an explicit shopping driver according to the historical behavior data of the user, if so, it may be perceived that the target shopping requirement of the user is merchandise information related to the shopping driver, and then merchandise content aggregated by the shopping driver may be generated by the AI content generation model by generating a corresponding prompt text. That is, the judgment can be made according to the situation of the user himself, and if a user is assumed to be marketing sensitive, and the past buying is a commodity with a large discount, namely, a commodity list with a reduced price can be provided; if branding is a concern, merchandise recommendation information may be provided relating to those brands, and so forth.

For the commodity detail page, if the target shopping requirement of the user is perceived to be obtaining more comprehensive information about the target commodity or obtaining effect information of the target commodity under a desired use scene, the detail information of the target commodity can be summarized by an AI content generation model to generate summarized texts, or after a related image of the use scene is provided by the user, a use effect image when the target commodity is simulated into the use scene is generated. For example, detailed description information of the commodity, or user evaluation content of the commodity may be summarized, and so on. In this way, online consultation service can be provided for users in 7 x24 h, customer service resources of platforms and merchants are supplemented, more durable and real commodity shopping guiding suggestions are provided, and buyer feedback and evaluation are helped to be summarized.

Or if the current commodity is a clothing commodity, the user can be inquired whether to look at the try-on effect of the clothing on the user, if so, the user can upload photos and the like of the user, and then the AI content generation model can generate an upper body effect diagram through AI drawing capability for reference by the user. Similarly, assuming that furniture-like merchandise is currently available, the user may be allowed to upload pictures of his/her home's room, etc., the AI content generation model may generate trial effects for a particular piece of furniture in his/her room, etc.

If the current page is a shopping cart page, judging that the user has the price change condition of acquiring the commodities in the commodity set to be settled according to the commodities in the commodity set to be settled, adding more commodities to the commodity set to be settled to meet the use condition of the rights of a user or carrying out associated commodity recommendation with the commodities in the commodity set to be settled, and enabling the commodity with the price change to be presented at the position in front of the current page by an AI content generation model or providing commodity recommendation information for meeting the use condition or belonging to the associated commodity.

In summary, through the embodiment of the application, in the process of displaying the current page, the AI interaction module can sense the current possible demand of the user according to the function provided by the current page and/or the behavior data generated by the user in the current page, after sensing the current possible target shopping demand of the user, the AI interaction module can generate a prompt text for dialogue with the artificial intelligent AI content generation model according to the target shopping demand, and then can call the pre-trained AI content generation model based on the prompt text to generate and display the target content corresponding to the target shopping demand. In this way, the user may have a target shopping demand perceived, and based on the perceived target shopping demand, an automatic prompt text is generated, and content for solving the corresponding problem is generated, so the method belongs to a demand-responsive active touch AI interaction method. In the mode, the user is not required to initiate interaction with the AI, but the AI interaction module is used for actively initiating the interaction, and the prompt text for carrying out the dialogue with the AI model can be automatically constructed, so that the difficulty and threshold for the user to interact with the AI are reduced, and the user experience is improved. And the solution corresponding to the shopping demand can be intelligently and dynamically provided according to the sensing results of different shopping demands of the users, so that the method is flexibly adapted to various user demands and scenes, and the information acquisition efficiency of the users is improved.

In an alternative implementation manner, the AI interaction module may be provided in the form of an SDK or the like, and the AI interaction module may be implanted into a plurality of different pages, for example, may be pages corresponding to a plurality of nodes on a shopping guide link in a merchandise information service system, so that the AI interaction module is unbinding with a specific page, and is convenient to provide a unified interaction experience across scene pages for a user.

It should be noted that, in the embodiment of the present application, the use of user data may be involved, and in practical application, the user specific personal data may be used in the solution described herein within the scope allowed by the applicable legal regulations in the country under the condition of meeting the applicable legal regulations in the country (for example, the user explicitly agrees to the user to notify practically, etc.).

Corresponding to the foregoing method embodiment, the embodiment of the present application further provides an intelligent interaction device, where the device may include:

Specifically, the AI interaction module exists in the form of a Software Development Kit (SDK) so as to be implanted into any page through the SDK.

When the shopping demand is sensed, the page element is in an unfolding or displaying state so as to be used for interacting with the AI interaction module through the page element.

In addition, the apparatus may further include:

And the demand confirming unit is used for providing operation options for confirming the target shopping demand in the target page element after the target shopping demand is perceived, so that prompt text for dialogue with an artificial intelligence AI content generating model is generated according to the target shopping demand after a confirmation message of a user is received.

Wherein the generated prompt text is a plurality of pieces, the device further comprises:

And the prompt text selection unit is used for providing operation options for selecting a plurality of 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.

Specifically, the functions provided by the current page include: providing commodity recommendation information; at this time, the demand sensing unit may specifically be configured to:

Or the functions provided by the current page include: providing commodity searching information; at this time, the demand sensing unit may specifically be configured to:

In particular, the demand sensing unit may specifically be configured to:

Or the demand sensing unit may be specifically configured to:

The AI content generation model can also be used for generating an atmosphere background image corresponding to the target scene when commodity recommendation information in the shopping list form is provided.

In addition, the demand sensing unit may specifically be configured to:

Further, the functions provided by the current page include: providing detailed information about the target commodity;

At this time, the demand sensing unit may specifically be configured to:

Or the functions provided by the current page include: providing commodity list information that the user has added to the commodity set to be settled;

The demand sensing unit may specifically be configured to:

In addition, the embodiment of the application also provides a computer readable storage medium, on which a computer program is stored, which when being executed by a processor, implements the steps of the method of any one of the previous method embodiments.

And an electronic device comprising:

one or more processors; and

A memory associated with the one or more processors for storing program instructions that, when read for execution by the one or more processors, perform the steps of the method of any of the preceding method embodiments.

In which fig. 7 illustrates an 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, an exercise device, a personal digital assistant, an aircraft, and so forth.

Referring to fig. 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.

The processing component 702 generally controls overall operation of the device 700, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing component 702 may include one or more processors 720 to execute instructions to perform all or part of the steps of the methods provided by the disclosed subject matter. Further, the processing component 702 can include one or more modules that 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.

Memory 704 is configured to store various types of data to support operations at device 700. Examples of such data include instructions for any application or method operating on device 700, contact data, phonebook data, messages, pictures, videos, and the like. The memory 704 may be implemented by any type or combination of volatile or nonvolatile memory devices 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 or optical disk.

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

The multimedia component 708 includes a screen between the device 700 and the user that provides an output interface. 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 a user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensor may sense not only the boundary of a touch or sliding action, but also the duration and pressure associated with the touch or sliding operation. In some embodiments, the multimedia component 708 includes a front-facing camera and/or a rear-facing camera. The front-facing camera and/or the rear-facing camera may receive external multimedia data when the device 700 is in an operational mode, such as a shooting mode or a video mode. Each front camera and rear camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

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

The I/O interface 712 provides an interface between the processing component 702 and peripheral interface modules, which may be a keyboard, click wheel, buttons, etc. These buttons may include, but are not limited to: homepage button, volume button, start button, and lock button.

The sensor assembly 714 includes one or more sensors for providing status assessment of various aspects of the device 700. For example, the sensor assembly 714 may detect an on/off state of the device 700, a relative positioning of the components, such as a display and keypad of the device 700, a change in position of the device 700 or a component of the device 700, the presence or absence of user contact with the device 700, an orientation or acceleration/deceleration of the device 700, and a change in temperature of the device 700. The sensor assembly 714 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 714 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 714 may also include an acceleration sensor, a gyroscopic sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

The communication component 716 is configured to facilitate communication between the device 700 and other devices, either wired or wireless. The device 700 may access a wireless network based on a communication standard, such as WiFi, or a mobile communication network of 2G, 3G, 4G/LTE, 5G, etc. In one 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 further includes a Near Field Communication (NFC) module to facilitate short range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, ultra Wideband (UWB) technology, bluetooth (BT) technology, and other technologies.

In an exemplary embodiment, the apparatus 700 may 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 elements for executing the methods described above.

In an exemplary embodiment, a non-transitory computer readable storage medium is also provided, such as memory 704 including instructions executable by processor 720 of device 700 to perform the methods provided by the disclosed subject matter. For example, the non-transitory computer readable storage medium may be ROM, random Access Memory (RAM), CD-ROM, magnetic tape, floppy disk, optical data storage device, etc.

From the above description of embodiments, it will be apparent to those skilled in the art that the present application may be implemented in software plus a necessary general hardware platform. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the embodiments or some parts of the embodiments of the present application.

In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for a system or system embodiment, since it is substantially similar to a method embodiment, the description is relatively simple, with reference to the description of the method embodiment being made in part. The systems and system embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.

The above describes the intelligent interaction method and the electronic device provided by the application in detail, and specific examples are applied to describe the principle and the implementation of the application, and the description of the above examples is only used for helping to understand the method and the core idea of the application; also, it is within the scope of the present application to be modified by those of ordinary skill in the art in light of the present teachings. In view of the foregoing, this description should not be construed as limiting the application.

Claims

1. An intelligent interaction method, characterized in that the method comprises the following steps:

2. The method of claim 1, wherein the step of determining the position of the substrate comprises,

3. The method of claim 2, wherein the step of determining the position of the substrate comprises,

4. The method of claim 2, wherein the step of determining the position of the substrate comprises,

The plurality of pages includes: pages corresponding to a plurality of nodes on a shopping guide link in the commodity information service system.

5. The method of claim 2, wherein the step of determining the position of the substrate comprises,

The AI interaction module corresponds to a page element for display in the target page, wherein the page element is in a packed or hidden state when the shopping demand is not perceived, and is in an unfolded or display state after the shopping demand is perceived, so as to be used for interacting with the AI interaction module through the page element.

6. The method as recited in claim 5, further comprising:

7. The method of claim 5, wherein the step of determining the position of the probe is performed,

The generated prompt text is a plurality of pieces, and the method further comprises:

8. The method of claim 1, wherein the step of determining the position of the substrate comprises,

The information on which the user may currently have needs is based when sensing it also includes: the user has asset information in the commodity information service system, marketing activity information issued in the commodity information service system, marketing activity information related to commodities and shopping preference information of the user.

9. The method of claim 1, wherein the step of determining the position of the substrate comprises,

10. The method of claim 1, wherein the step of determining the position of the substrate comprises,

11. The method according to claim 9 or 10, wherein,

The sensing the possible demand of the user according to the judging result comprises the following steps:

12. The method of claim 11, wherein the step of determining the position of the probe is performed,

The commodity dimension-based information providing and/or decision-making auxiliary content comprises: contrast and/or merchandise recommendation information in multiple dimensions between the same merchandise offered by different merchants.

13. The method according to claim 9 or 10, wherein,

14. The method of claim 13, wherein the step of determining the position of the probe is performed,

15. The method according to claim 9 or 10, wherein,

16. The method of claim 9, wherein the step of determining the position of the substrate comprises,

17. The method of claim 1, wherein the step of determining the position of the substrate comprises,

18. The method of claim 1, wherein the step of determining the position of the substrate comprises,

19. A computer readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements the steps of the method of any of claims 1 to 18.

20. An electronic device, comprising:

one or more processors; and

A memory associated with the one or more processors for storing program instructions that, when read for execution by the one or more processors, perform the steps of the method of any of claims 1 to 18.