Information interaction method and device, equipment, storage medium and product

Abstract

The embodiment of the invention discloses an information interaction method and device, equipment, a storage medium and a product, the method is applied to an e-commerce shopping scene, and the method comprises the following steps: receiving a user session message input by a user; obtaining a feedback session message corresponding to the user session message, and displaying the feedback session message in a current session interface; and in response to a received newly-added user session message input by the user according to the additional description prompt information, displaying a feedback session message corresponding to the newly-added user session message in the session interface. According to the technical scheme provided by the embodiment of the invention, gradual recommendation object optimization is realized, and the adaptation degree between the display object and the user is improved.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of information processing, and in particular, to an information interaction method, device, equipment, storage medium and product. BACKGROUND

[0002] With the development of e-commerce, online shopping through e-commerce platforms has become common. Currently, when shopping based on e-commerce platforms to obtain corresponding physical products or virtual services, the object name or keyword is mainly input in the search box, so that the server corresponding to the e-commerce platform determines the object adapted to the object name or keyword and displays it.

[0003] The inventor found the following problems when implementing the technical solution based on the above method:

[0004] When displaying objects based on the above method, the corresponding objects are mainly browsed and selected according to the preset display order, for example, the preset display order is mainly based on factors such as comprehensive ranking, price and sales. However, there are many search results corresponding to each keyword or object name, and users cannot obtain objects adapted to them from a large number of displayed objects, resulting in the problem of low adaptation between displayed objects and users. SUMMARY

[0005] Embodiments of the present application provide an information interaction method, device, equipment, storage medium and product to realize progressive recommendation object optimization and improve the adaptation between displayed objects and users.

[0006] In a first aspect, embodiments of the present application provide an information interaction method, which comprises:

[0007] receiving a user session message input by a user; wherein the user session message corresponds to a user demand description;

[0008] obtaining a feedback session message corresponding to the user session message, and displaying the feedback session message in a current session interface; wherein the feedback session message is used to represent a demand analysis result of the user demand description, and the demand analysis result at least includes demand analysis text information and additional description prompt information for indicating user additional questioning;

[0009] In response to receiving a new user session message input by a user according to the additional description prompt information, displaying a feedback session message corresponding to the new user session message in the session interface.

[0010] Further, the displaying the feedback session message in the current session interface comprises:

[0011] The requirement analysis result in the feedback session message is sequentially displayed in a streaming manner in the current session interface.

[0012] Further, the requirement analysis result in the feedback session message includes at least one first object available for the user to acquire, and the displaying of the feedback session message in the current session interface includes:

[0013] After the requirement analysis text information is displayed in the current session interface in a streaming manner, object information of at least one first object is sequentially displayed, and object recommendation reason information of the first object is displayed at an associated position of the object information.

[0014] The object recommendation reason information is determined based on one or more of object comment information, object attribute information, user session messages, and the user portrait of the first object.

[0015] Further, in the case where the feedback session message includes the first object, the feedback session message further includes a requirement description prompt identifier, the requirement description prompt identifier is used to represent the acquisition requirement of the user for the first object, and the displaying of the feedback session message in the current session interface includes:

[0016] In response to a triggering operation on the requirement description prompt identifier, the first object displayed in the feedback session message is adjusted.

[0017] Further, the feedback session message further includes an object set filtered based on the first session message, and the object set includes at least an object total number of a filtered second object and a viewing identifier for viewing the second object, and the method further includes:

[0018] In response to a triggering operation on the viewing identifier, an object display page including the second object is displayed.

[0019] The second object includes the first object.

[0020] Further, the method further includes: in response to a triggering operation on any first object, an object display panel is displayed in the current session interface to present object detail information and / or object acquisition information of the triggered first object based on the object display panel.

[0021] Further, the response to receiving the user input of the new user session message according to the additional description prompt information, displaying the feedback session message corresponding to the new user session message in the session interface, comprises: in response to the user triggering operation of the additional description prompt information and / or the event of inputting the session message in the edit box of the current session interface, generating the new user session message;

[0022] By processing the new user session message and the previous feedback session message, the feedback session message corresponding to the new user session message is determined; or,

[0023] By processing the new user session message and the historical session message in the current session interface, the feedback session message corresponding to the new user session message is determined.

[0024] Among them, the historical session message at least includes historical user session message and / or historical feedback session message.

[0025] Further, the method further comprises: in response to the triggering operation of replacing the first selectable item in the additional description prompt information, adjusting at least one first selectable item in the additional description prompt information.

[0026] Further, the feedback session message is determined based on the following way:

[0027] According to the user session message, the session intention of the user is determined;

[0028] According to the session intention and the user portrait of the user, the feedback session message is determined.

[0029] Further, the method further comprises:

[0030] Calling a target processing mode corresponding to the session intention;

[0031] Based on the target processing mode, the user session message and the user portrait are analyzed and processed to determine the feedback session message.

[0032] Further, the session intention is an object acquisition intention, and the object acquisition intention includes a commodity acquisition intention and / or a service acquisition intention. Based on the target processing mode, the user session message and the user portrait are analyzed and processed to determine the feedback session message, which comprises:

[0033] Rewriting understanding of the user session message to obtain first information;

[0034] Based on the first information and the user portrait, a demand analysis text information, a first preset number of first objects and an additional description prompt information are generated.

[0035] Further, the additional description prompt information is determined based on the following manner:

[0036] Based on the object attribute information of the second object, the comment information and the user session message, at least one selectable item is determined to generate the additional description prompt information based on the at least one selectable item.

[0037] Further, the method further comprises: in response to detecting an object acquisition event of acquiring the first object, recording an object acquisition progress of the acquired first object; and displaying the object acquisition progress.

[0038] In a second aspect, the embodiments of the present application further provide an information interaction device, which comprises:

[0039] a user session receiving module, configured to receive a user inputted user session message; wherein the user session message corresponds to a user demand description;

[0040] a feedback session display module, configured to acquire a feedback session message corresponding to the user session message, and display the feedback session message in a current session interface; wherein the feedback session message is used to represent a demand analysis result of the user demand description, and the demand analysis result at least comprises demand analysis text information and an additional description prompt information used to instruct a user to add a question;

[0041] an added session processing module, configured to, in response to receiving an added user session message inputted by a user according to the additional description prompt information, display a feedback session message corresponding to the added user session message in the session interface.

[0042] In a third aspect, the embodiments of the present application provide an electronic device, which comprises:

[0043] one or more processors;

[0044] a memory, configured to store one or more programs;

[0045] When the one or more programs are executed by the one or more processors, the one or more processors implement the information interaction method provided by any of the embodiments of the present application.

[0046] In a fourth aspect, the embodiments of the present application further provide a computer readable storage medium, which stores a computer program, and the program is executed by a processor to implement the information interaction method provided by any of the embodiments of the present application.

[0047] In a fifth aspect, an embodiment of the present application further provides a computer program product comprising a computer program, wherein the computer program, when executed by a processor, implements the information interaction method according to any one of the embodiments of the present application.

[0048] The technical solution provided by the embodiment of the present application can obtain the feedback session message corresponding to the user session message when receiving the user input user session message used to represent the user demand description, and display the feedback session message in the current session interface, wherein the feedback session message is used to represent the demand analysis result of the user demand description, and the demand analysis result at least includes the demand analysis text information and the additional description prompt information used to indicate the user additional question, so that the feedback session message corresponding to the new user session message input by the user according to the additional description prompt information is displayed in the session interface in response to receiving the new user session message, and the technical problem of low matching efficiency of massive goods and user demand in the e-commerce scenario is effectively solved. The user demand description is analyzed and processed to convert it into structured data of implicit demand, the intention understanding accuracy is improved, the recommendation logic is intuitively displayed through the demand analysis text information, and the recommendation direction is dynamically corrected in combination with the additional description prompt information, so that the progressive recommendation object optimization is realized, and the adaptation degree between the display object and the user is improved. BRIEF DESCRIPTION OF DRAWINGS

[0049] In order to more clearly illustrate the technical solutions of the example embodiments of the present application, the drawings needed in the description of the embodiments are briefly introduced below. Obviously, the drawings introduced are only a part of the drawings of the embodiments to be described by the present application, and not all the drawings. Those skilled in the art can obtain other drawings according to these drawings without creating creative labor.

[0050] Figure 1 The flowchart of the information interaction method provided by the embodiment of the present application;

[0051] Figure 2 The implementation process diagram of the information interaction method related to the embodiment of the present application;

[0052] Figure 3 The flowchart of another information interaction method provided by the embodiment of the present application;

[0053] Figure 4 The display effect diagram of the first object and the object selection reason information related to the embodiment of the present application;

[0054] Figure 5 The diagram of the demand description prompt identifier related to the embodiment of the present application;

[0055] Figure 6A flowchart of another information interaction method provided by an embodiment of the present application;

[0056] Figure 7 A flowchart of another information interaction method provided by an embodiment of the present application;

[0057] FIG. 8(a) is a schematic diagram of a feedback session message after a user inputs a user session message according to an embodiment of the present application;

[0058] FIG. 8(b) is a schematic diagram of an interface for displaying more goods according to an embodiment of the present application;

[0059] FIG. 8(c) is a schematic diagram of an information display page for triggering additional description prompt information according to an embodiment of the present application;

[0060] FIG. 8(d) is a schematic diagram of a display page for pushing content related to remaining service items of the user session content according to an embodiment of the present application;

[0061] FIG. 8(e) is a schematic diagram of a display page for pushing content related to medication guidance for purchasing a medicine according to an embodiment of the present application;

[0062] FIG. 8(f) is a schematic diagram of a display page for pushing content related to price reduction reminders and repeat purchase reminders for purchasing a medicine according to an embodiment of the present application;

[0063] FIG. 9(a) is a schematic diagram of a feedback session message after a user inputs a user session message according to an embodiment of the present application;

[0064] FIG. 9(b) is a schematic diagram of an information display page for triggering additional description prompt information according to an embodiment of the present application;

[0065] FIG. 9(c) is a schematic diagram of a service progress information display page according to an embodiment of the present application;

[0066] FIG. 9(d) is a schematic diagram of a display page for interpreting a detection report and providing subsequent suggestions according to an embodiment of the present application;

[0067] Figure 10 FIG. 10 is a schematic diagram of a background execution logic for implementing the information interaction method provided by an embodiment of the present application;

[0068] Figure 11 FIG. 11 is a schematic diagram of an information interaction device according to an embodiment of the present application;

[0069] Figure 12 FIG. 12 is a schematic diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0070] The application will be described in further detail below with reference to the drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the application and not to limit the application. In addition, it should be noted that, for the sake of description, only the parts related to the application are shown in the drawings and not all the structures.

[0071] It should be noted that the information (including but not limited to user equipment information, user personal information, etc.), data (including but not limited to data for analysis, stored data, displayed data, etc.) and signals involved in the application are authorized by the user or fully authorized by all parties, and the collection, use and processing of related data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0072] Before introducing the technical solutions provided by the embodiments of the application, the application scenario can be exemplarily described. The application scenario of the embodiments of the application can be any scenario that needs to accurately and efficiently screen out objects meeting the personalized needs of users. For example, the information interaction method provided in the embodiments can be integrated into any existing platform supporting online shopping, health examination application software or tool. A user who wants to shop can search for the commodity information he wants to buy in the platform, application software or tool, at this time, the technical solutions provided in the embodiments can be used to display the screened commodities to the client device of the user, and the screened commodities are presented on the client device. Optionally, the client device can include but is not limited to a mobile terminal, a wearable device, a vehicle-mounted device, a tablet computer, a network device or a desktop computer, etc. The scheme provided in the embodiments of the application can be presented in the form of a conversation, which realizes the effect that all the information needed can be previewed without page jumping.

[0073] For example, a user wants to buy a "smart bracelet suitable for sending to an elder" on an e-commerce platform, but is not sure about the specific needs, so the user can input the natural language description "want to buy a health monitoring bracelet for a 60-year-old father, hope to be simple to operate, can measure heart rate and blood pressure, and the budget is about 500 yuan" in the commodity search box. The technical solutions provided in the embodiments clarify the needs gradually through the conversation, and the effect finally realized is that the user does not need to manually screen hundreds of commodities, quickly locks the ideal choice through natural conversation, avoids the cumbersome process of comparing by himself after traditional keyword search "bracelet for the elderly", and quickly and efficiently displays the target commodity suitable for the user.

[0074] Figure 1This is a flowchart of an information interaction method provided by an embodiment of the present invention. This embodiment can be applied to scenarios where objects that meet the personalized needs of users are accurately and efficiently screened out. The information display device corresponding to the information display method can be integrated into the client. This method can be executed by the client or the server, or it can be implemented by the interaction between the client and the server. The client is installed in a terminal device, which can be a PC, a mobile terminal, or any electronic device that can support the client.

[0075] like Figure 1 As shown, the method specifically includes the following steps:

[0076] S110: Receive a user conversation message input by a user, where the user conversation message corresponds to a user demand description.

[0077] User conversation messages refer to initial requests or progressive descriptions of needs entered by users during an interaction and can be expressed in natural language. User conversation messages can be sentences, questions, or phrases expressed in natural language. User need descriptions refer to specific needs or questions proactively raised by users during an interaction, using natural language, keywords, or through dialogue. These descriptions can include core elements such as the functions, attributes, usage scenarios, and budget of the target product or service. User need descriptions can be ambiguous, fragmented, or implicit, requiring further analysis and interactive clarification in subsequent steps to transform them into executable search criteria for accurate matching and recommendations. For example, user conversation messages might be, "I want a thin and light laptop with more than 12 hours of battery life" or "Recommend an affordable sunscreen suitable for sensitive skin." User conversation messages serve as the starting point of a conversation, and their content is analyzed to generate corresponding responses, guiding subsequent interactions.

[0078] It should be noted that the user conversation message may be a conversation message input for the first time or a conversation message input again.

[0079] In a specific application, the information interaction method provided in the embodiment can be integrated into any existing platform, application software or tool supporting online shopping. In order to clearly introduce the technical solution, the shopping platform will be taken as an example for introduction. When a user needs to purchase a product, the user can enter the homepage of the shopping platform, and the homepage of the shopping platform is pre-configured with a demand dialogue box, in which the user can edit the user session message. The demand dialogue box is first presented in a similar presentation form as the search box. Thus, the background of the shopping platform can receive the user session message input by the user for the first time, and the user session message can be understood as the original demand expression of the user, which is a shopping intention or question directly stated in natural language form. In the embodiment, the core purpose of receiving the user session message input by the user is to capture the initial, possibly incomplete but core demand expression of the user, as the basis for further analysis and progressive demand clarification.

[0080] S120, obtaining a feedback session message corresponding to the user session message, and displaying the feedback session message in the current session interface.

[0081] The feedback session message is an interactive response generated based on the initial demand of the user (i.e., the user session message) through demand analysis. The current session interface is a real-time interactive dialogue window or page, which displays the user session message input by the user and the generated response (such as the feedback session message). More specifically, the feedback session message is used to represent the demand analysis result of the user demand description, and the demand analysis result at least includes demand analysis text information and additional description prompt information for indicating additional questions of the user.

[0082] The demand analysis result is a structured output generated after semantic understanding, intention recognition and demand disassembly of the user session message. The demand analysis text information is a demand understanding conclusion fed back in the form of natural language, such as “light and thin laptops meeting your expectations have been screened”.

[0083] The additional description prompt information is a guiding question designed for demand ambiguity points, such as “whether a touch screen function is needed?”. In the feedback session message, the guiding questions or options provided by the additional description prompt information are specially designed to stimulate the user to interact in the next round, and when the user responds to these prompts, the reply will be automatically converted into the input content of the next dialogue, i.e., the next session message. Thus, a coherent multi-round demand clarification process is formed to gradually narrow down the recommended range and gradually improve the matching accuracy.

[0084] Optionally, when the feedback conversation message is used to screen out commodity or service items that match the user's demand, one or more first objects can also be included in the feedback conversation message. In this case, the first objects respectively represent different entity commodities or different service items. The first objects can be presented in the form of a commodity display card, a commodity thumbnail, a commodity link, a service item display card, a service item link, etc. in the feedback conversation message. To clearly introduce the technical solution, the subsequent description will take the first object as an example of an entity commodity.

[0085] On the basis of the above embodiment, optionally, the demand analysis text information includes one or more of the preliminary analysis text of the user conversation message, the recommendation idea text, and the object push suggestion text.

[0086] In this embodiment, the preliminary analysis text of the user conversation message refers to the explanatory feedback text generated after semantic analysis of the user's original input user conversation message. The preliminary analysis text can contain a direct translation of the user's explicit demand, or it can be an implicit demand supplemented by an algorithm. The recommendation idea text refers to the recommendation logic explanation text transparently displayed to the user, which contains a direct response to the user's explicit conditions and also incorporates the explanations of collaborative filtering, commodity popularity, and other implicit recommendation strategies. The object push suggestion text refers to the specific explanation copy provided synchronously when displaying the recommended commodity (first object), and its function is to highlight the differentiated advantages of each recommended object and the matching point with the user's demand.

[0087] Next, taking the user conversation message "want a lightweight notebook with more than 12 hours of battery life" as an example, the information content that the demand analysis text information can include is described. When the user conversation message is "want a lightweight notebook with more than 12 hours of battery life", the demand analysis text information can include: the preliminary analysis text of the user conversation message, such as "locked to a model with a weight <1.5 kg and an official battery life ≥12 hours, a total of 8 models are selected"; the recommendation idea text, such as "preferably recommended: a model with actual test battery life of 14.2 hours, the lightest 1.2 kg in similar products, and a recent best-selling product with a 95% good rating"; and the object push suggestion text, such as the object push suggestion text corresponding to the first object, which can be represented as "actual test battery life of 14 hours / weight of 1.3 kg, 2 hours more battery life than required", and the object push suggestion text corresponding to the second object, which can be represented as "supports 65W fast charging, while similar products support 45W", etc. The three work together to achieve full-link guidance from demand understanding to decision support.

[0088] Further, when the user conversation message corresponds to a message for obtaining a specific commodity, the recommendation idea text includes brand selection text and / or specification selection text.

[0089] The message of obtaining a specific product refers to a conversation message in which the user directly inquires about a specific product, rather than a vague demand description. The brand selection text is an explanation and description for the brand dimension in the recommendation logic. The specification selection text is a description of the decision basis for the selection of specific parameters of the product.

[0090] For example, if the user conversation message is "camera", the corresponding recommendation idea text may include: brand selection text, such as "recommend brands A, B, and C, which have a market share of more than x% and a rich lens group"; specification selection text "prefer to show full-frame models, as your historical browsing records show a high level of interest in professional models; although entry-level prices are lower, full-frame sensor size is improved", supplemented by an industry trend description of certain specification products. The advantage of this setting is that by comparing brand market share data and specification technical indicators horizontally, it helps users establish a benchmark for purchasing in a general demand scenario.

[0091] In this embodiment, after receiving the user conversation message, the user conversation message can be analyzed by natural language processing technology to generate intelligent response demand analysis results including at least demand analysis text information, one or more first objects, and additional description prompt information. These demand analysis results are displayed in the current conversation interface. When the user triggers the additional description prompt information or supplementary answer, it is automatically converted into the next conversation message, forming a "demand-understanding-recommendation-clarification" closed-loop optimization process, and finally improving the product matching accuracy.

[0092] Next, a specific example is used to illustrate the above technical solutions. Figure 2 An implementation process diagram of an information interaction method is shown. The user can input the user conversation message "want a light and thin notebook with more than 12 hours of battery life" in the demand dialogue box on the home page of the shopping platform, and click the "confirm" control. At this time, the user conversation message can be analyzed by natural language processing technology to generate feedback conversation information. As shown in Figure 2 The feedback conversation information can include at least: demand analysis text information "has filtered light and thin notebooks that meet your expectations"; optionally, the additional description prompt information can be a label that the user can select, such as the "requirement 1" label, the "requirement 2" label, the "requirement 3" label, and the "requirement 4" label, as shown in Figure 2 The "need touch screen function" input by the user in the edit box, as shown in Figure 2 The feedback conversation message can also include a plurality of first object product cards that the user can obtain, namely, product A, product B, product C, and the like.

[0093] S130, in response to receiving the new user session message input by the user according to the additional description prompt information, displaying a feedback session message corresponding to the new user session message in the session interface.

[0094] The new user session message refers to a new message triggered by an interactive behavior of the user in the current dialogue process. The new user session message is generated by triggering the additional description prompt information and / or inputting information in the edit box of the current session interface. It can be understood that the generation method of the new user session message mainly includes three methods: the first method is to click the additional description prompt information provided in the current session interface, for example, Figure 2 the "requirement 1" label, the "requirement 2" label, the "requirement 3" label and the "requirement 4" label; the second method is to manually input new content in the edit box of the current session interface; and the third method is a combination of the first method and the second method.

[0095] In the embodiment, after the feedback session message is displayed in the current session interface, the user can continue to trigger new session messages in the current session interface. At this time, the existing information content in the current session interface and the new session information can be dynamically integrated as updated "user session messages", which can gradually refine the understanding of user demand, and each new interaction is based on the accumulation of previous dialogue. The accuracy of the demand analysis result can be continuously optimized, and the user can gradually supplement the demand details through the additional description prompt information, forming a spiral upward cognitive iteration, and the goods or service items suitable for the user can be gradually screened out. When the user triggers the additional description prompt information provided in the current session interface or manually inputs new content to trigger the new user session message event, the new message and the historical session message (the historical session message can be all message records in the current session thread that have been processed, including the initial user session message and subsequent multi-round interactive content) can be intelligently integrated. The complete context after integration is input into the recommendation engine as the updated user session message, so as to generate a more accurate feedback session message.

[0096] For example, after the user initially inputs the user session message "want a light and thin notebook with more than 12 hours of battery life", the recommended list that meets the condition can be displayed first, and the label information "budget 5000-6000" is prompted in the additional description prompt information. At this time, the user can trigger the label or input the new user session message "budget 5000-6000" through the edit box, so as to intelligently integrate the historical session message, i.e. the demand for battery life and lightness, with the new user session message, i.e. the budget of 5000-6000 yuan, to generate the updated user session message "light and thin notebook with more than 12 hours of battery life and budget within 5000 yuan". In the feedback session message generated based on the integrated demand, the first object recommended meets all the historical and new constraints, achieving accurate recommendation in the demand iteration process.

[0097] Through the above steps, after displaying the feedback session message in the current session interface, the user can continue to trigger new session messages in the current session interface. At this time, the existing information content in the current session interface and the new session information can be dynamically integrated as updated "user session messages", which can gradually refine the understanding of user demand, and each new interaction is based on the accumulation of previous dialogues, which can continuously optimize the accuracy of demand analysis results. At the same time, the user is guided to gradually supplement the demand details through the additional description prompt information, forming a spiral upward cognitive iteration, which can gradually filter out goods or service items that are suitable for the user, and through the continuous association of new and old session messages, the context of the dialogue is inherited and optimized, ensuring that the recommendation results are continuously refined as the dialogue deepens.

[0098] The technical solution provided by the embodiment of the application can obtain a feedback session message corresponding to the user session message when receiving a user input user session message representing a user demand description, and display the feedback session message in the current session interface, wherein the feedback session message represents a demand analysis result of the user demand description, and the demand analysis result at least includes demand analysis text information and additional description prompt information for indicating user additional questions. Therefore, the feedback session message corresponding to the new user session message input by the user according to the additional description prompt information can be displayed in the session interface in response to receiving the new user session message, effectively solving the technical problem of low matching efficiency of massive goods and user demand in the e-commerce scenario. By analyzing and processing the user demand description, it is converted into structured data of implicit demand, improving the intent understanding accuracy. The recommendation logic is intuitively displayed through the demand analysis text information, and the recommendation direction is dynamically corrected in combination with the additional description prompt information, realizing progressive recommendation object optimization, thereby improving the adaptation degree between the display object and the user.

[0099] Figure 3A flowchart of an information interaction method provided by an embodiment of the present application is shown in the foregoing embodiment. The display manner and content of the feedback conversation message can be further refined. The specific implementation can be referred to the detailed description of the embodiment. The solution provided by the embodiment of the present application can be integrated in an extended reality device to be implemented. The same or corresponding technical terms as the foregoing embodiments are not described herein again. As shown in Figure 3 The method specifically includes the following steps:

[0100] S210, receiving a user conversation message input by a user. The user conversation message corresponds to a user demand description.

[0101] S220, obtaining a feedback conversation message corresponding to the user conversation message, and sequentially displaying a demand analysis result in the feedback conversation message in a streaming manner in a current conversation interface.

[0102] The feedback conversation message is used to represent the demand analysis result of the user demand description. The demand analysis result at least includes demand analysis text information and additional description prompt information used to indicate that the user adds a question.

[0103] The streaming manner refers to that the feedback conversation message displays content in a dynamic and gradual manner in the interface.

[0104] In the embodiment, each component of the feedback conversation message, i.e., the demand analysis text and the additional description prompt information, can be dynamically displayed in batches according to a logical order. For example, the demand analysis text information is displayed first, and the additional description prompt information appears after 0.5 seconds. This design simulates the rhythm of human conversation and avoids visual oppression caused by information stacking.

[0105] Optionally, the demand analysis result in the feedback conversation message can be displayed according to a preset layout manner. The preset layout manner refers to a structured layout rule designed in advance according to content types. For example, the demand analysis text is displayed on the left, and the additional description prompt information is suspended at the bottom, so that different information levels are clear and visible, and the different terminal screen sizes are adapted. This standardized display strategy can help to improve the recognition efficiency of key information and reduce the cognitive load of the user.

[0106] On the basis of the foregoing embodiment, the demand analysis result of the feedback conversation message includes at least one first object available for the user to obtain. The feedback conversation message is displayed in the current conversation interface, including:

[0107] After the demand analysis text information is displayed in the current conversation interface in the streaming manner, the object information of the at least one first object is sequentially displayed, and the object recommendation reason information of the first object is displayed at the associated position of the object information.

[0108] The first object refers to an initial recommended commodity set matched based on the current understanding, and the first object can be an entity commodity or a virtual service item. The object information refers to detailed description data of the first object. For example, the object information can include core attributes of the first object, such as name, price, specification, picture, access portal, and possible behavior options, such as purchase, collection, preview, and the like. Optionally, the object information can be presented in the form of a commodity display card, a commodity thumbnail, a commodity link, and the like in the feedback session information.

[0109] The associated position of the first object refers to a specific display area directly adjacent to or bound to each first object. For example, the associated position can be in one of the following forms: 1) a folding explanation column below the commodity display card of the first object; 2) a position directly above adjacent to the commodity display card of the first object; 3) an information bubble suspended on the commodity picture; and 4) a corner mark prompt on the right side of the commodity parameter.

[0110] The object selection reason information is a decision basis description directly associated with each recommended commodity. The selection reason description information is determined based on one or more of the object review information, the object attribute information, the user session message, and the user portrait of the first object. The object review information refers to real evaluation data of the first object by other users; the object attribute information refers to the specification parameters and functional characteristics of the first object itself. The user portrait is a preference model constructed according to the historical behavior of the user.

[0111] In the embodiment, in the process of determining the feedback session message, a plurality of first objects available for the user to obtain can also be determined, and the object selection reason information corresponding to each first object can also be determined. Specifically, the object selection reason information can be generated according to at least one of the object review information, the object attribute information, the user session message, and the user portrait. The group consensus can be extracted through the object review information; the user demand can be directly matched through the object attribute information and the user session message; and the individualization element can be injected through the user portrait. The object selection reason information generated according to at least one of the four elements can simultaneously realize the multiple improvements of precision, reliability, and individualization, because it integrates multi-dimensional data such as commodity objective attributes, group evaluation consensus, and personal historical preferences.

[0112] Based on this, the way to display the feedback conversation message in the current conversation interface can be: the demand analysis text information can be completely displayed in the current conversation interface in a dynamic word-by-word or segmented loading streaming form; after the demand analysis text information is completely displayed, the object information of one or more first objects filtered and the object selection reason information corresponding to each first object can be sequentially presented one by one; finally, the additional description prompt information is presented, which simulates the rhythm of human conversation and avoids visual oppression caused by one-time information stacking.

[0113] It should be noted that the display order of the feedback conversation message in the streaming form is only exemplary, and the specific display order can be set according to actual needs.

[0114] Optionally, the number of the first objects is a first preset value. The first preset value refers to an initial recommended product quantity upper limit set in advance according to the interaction scene, for example, the default first preset value is 3-5. For example, according to the first preset value, for example, 5 are displayed by default, 5 first objects are gradually displayed in the feedback conversation message, and the corresponding display object selection reason information is synchronously displayed at the associated position of each first object.

[0115] For example, the display effect schematic diagram of the first object and the object selection reason information is shown in Figure 4 As shown in Figure 4 , the feedback conversation message displays 5 first object product display cards, which are product display card A, product display card B, product display card C, product display card D, and product display card E. The object selection reason information displayed above the product display card A is "92% of users think that the actual endurance meets the standard"; the object selection reason information displayed above the product display card B is "the brand you often buy"; the object selection reason information displayed above the product display card C is "compatible with your existing charger"; the object selection reason information displayed above the product display card D is "actual endurance of 14 hours (exceeds your requirement)"; and the object selection reason information displayed above the product display card E is "top rated in the same category".

[0116] In this embodiment, by directly displaying the selection reason information at the associated position of the first object, the transparency of the recommendation logic and the user decision-making efficiency can be significantly improved. The user can intuitively understand the matching basis of each recommended object and the user's demand without additional operation, which reduces the decision-making confusion caused by information asymmetry. This embedded explanation method maintains the simplicity of the interface, so that the user naturally completes the cognitive loop while obtaining the recommended result. In this way, the recommended result has both demand matching degree and user adaptability.

[0117] On the basis of the above-mentioned embodiments, optionally, in the case that the first object is included in the feedback session message, the feedback session message further includes a demand description prompt identifier, the demand description prompt identifier is used to represent the acquisition demand of the user on the first object, and the feedback session message is displayed in the current session interface, including: in response to the triggering operation on the demand description prompt identifier, the first object displayed in the feedback session message is adjusted.

[0118] The demand description prompt identifier is an interactive label or a filtering condition extracted from the user session message and displayed in the current session interface in a visual form. The acquisition demand refers to the core appeal characteristics of the target commodity. When the user clicks / modifies these identifiers, the recommendation result can be updated in real time to realize dynamic demand optimization.

[0119] In the embodiment, the user session message can be semantically analyzed to automatically generate structured demand description prompt identifiers, which are displayed in the session interface in the form of interactive labels, which not only explicitly represent the current filtering condition, that is, the acquisition demand, but also support user click modification. When the identifier is triggered to adjust, the commodity library is re-searched in real time to update the recommended first object list. In this way, the recommendation result dynamically adapts to the change of user demand, and the commodity filtering efficiency is greatly improved.

[0120] For example, the schematic diagram of the demand description prompt identifier is shown in Figure 5 The user session message is "urgently need a light and thin long-lasting notebook computer", and a plurality of structured demand description prompt identifiers are automatically generated by semantically analyzing the session message, including: demand description prompt identifier 1 "delivery time limit", demand description prompt identifier 2 "weight≤1.5kg", demand description prompt identifier 3 "battery life≥12 hours", and demand description prompt identifier 4 "free shipping" and the like. When at least one of the demand description prompt identifiers is triggered to adjust, the commodity library can be re-searched in real time to update the recommended first object. For example, the user clicks the "weight≤1.5kg" label and modifies it to "weight≤1.2kg", and at this time the recommended first object is all the models below 1.2kg.

[0121] On the basis of the above-mentioned embodiments, optionally, the feedback session message further includes an object set filtered based on the first session message, the object set at least includes the total number of objects of the filtered second object and a viewing identifier of the second object, and the method further includes: in response to the triggering operation on the viewing identifier, an object display page including the second object is displayed.

[0122] The object set refers to a set of all candidate goods initially screened according to the user session message. The second object represents all goods in the set that meet the conditions. The second object includes the first object, which can be understood as a part of the second object displayed on the front screen. These displayed second objects are also called first objects. The first object is actually a second object that is screened and has a higher matching degree with the user. The total number of objects is used to inform the user of the total size of the matching goods, for example, “a total of 128 notebooks meeting the requirements are found”. The viewing identifier is an identifier control that triggers a button or a link to jump to the object display page.

[0123] The object display page refers to a separate interface or pop-up module specially designed to centrally present the object set of the screening result when the user triggers the viewing identifier (such as the “view more” button or link) to jump or expand. The page will structure the list of all objects that meet the screening conditions, for example, in the form of cards, tables, pagination, or waterfall flow, and can include the simplified information of each object and the sorting and filtering controls.

[0124] In this embodiment, when generating the feedback session message, not only some recommended objects are directly displayed, but also more related objects are pre-screened based on the user session message to form a hidden object set. In the current session interface, the total number of objects of the object set and the interactive viewing identifier can be displayed. When the user clicks the viewing identifier, the page jumps or expands the floating window to enter the specially designed object display page, which will present all second objects and their detailed information in a more suitable layout for browsing. The purpose of this setting is to achieve seamless connection from instant dialogue recommendation to extended browsing, meet the user's viewing needs for a more comprehensive result set, and balance the recommendation accuracy and selection freedom through this hierarchical display strategy, so that the user can quickly obtain the preferred goods and expand the selection range at any time.

[0125] For example, continuing to refer to Figure 4 , Figure 4 The object set information in the upper dashed box in the figure is that “N items of goods have been shared for you”, which indicates that the total number of second objects in the object set information screened based on the user session message is N. “View more goods” represents the viewing identifier. When the user triggers the viewing identifier, the second objects can be displayed in the form of pagination or waterfall flow.

[0126] On the basis of the above embodiment, optionally, the information interaction method further includes: in response to a triggering operation on any first object, displaying an object display panel in the current session interface to present the object detail information and / or object acquisition information of the triggered first object based on the object display panel.

[0127] The object display panel refers to a pop-up window or an expanded area dynamically popped up in the current session interface, and is used for centralized display of detailed information of the first object triggered by the user. The object detail information can be represented by an object detail page, and the object detail page can include complete parameters, multi-angle text and video introduction, and use scene description of the goods, and the like. The object acquisition information refers to purchase related data, and the object acquisition information can be represented by an object settlement page. For example, the object settlement page can include the delivery address, the actual payment amount, the real-time inventory, the preferential activities, the delivery time limit and the like. In a specific application, under the condition that the page display content is limited, the object detail information and / or the object acquisition information can be displayed in the form of a scroll bar or a waterfall flow.

[0128] In the embodiment, when the user performs a triggering operation such as clicking or hovering on any first object in the recommendation list, an object display panel can be popped up immediately in the current session window. The object display panel can adopt an intelligent column layout to present at least one of the object detail information and the object acquisition information of the triggered first object. For example, the main area of the object display panel presents the object detail information, and when the user triggers the “confirm purchase control”, the object settlement page can be jumped to to display the object acquisition information, such as real-time data of “delivery address”, “actual payment amount”, “A area warehouse inventory”, “get a discount of 200 yuan” and the like. Through the hierarchical design of the floating panel display, the user can be ensured to continuously stay in the core decision-making scene without being lost, and an uninterrupted immersive product exploration process is realized.

[0129] S230, in response to receiving the new user session message input by the user according to the additional description prompt information, displaying a feedback session message corresponding to the new user session message in the session interface.

[0130] S240, in response to detecting an object acquisition event of the first object, recording an object acquisition progress of the acquired first object, and displaying the object acquisition progress.

[0131] The object acquisition event refers to a key interaction behavior of the user on the first object, which can be captured in real time through a burying point technology. The object acquisition progress is a visual index for quantifying the user decision-making process, which is usually displayed in the interface in the form of a progress bar, a score or a percentage.

[0132] In the embodiment, when the object acquisition event for a certain object is detected, the behavior is automatically recorded and the object acquisition progress is updated. The progress information is displayed in real time in the interface side bar through a dynamic visualization component such as a ring progress chart, a step-by-step guide bar and the like. For example,

[0133] The purpose of such design is to help users perceive the decision-making stage, which can reduce selection anxiety and provide behavioral data basis for optimizing subsequent recommendations.

[0134] The technical scheme provided by the embodiment of the application has the following beneficial effects: first, the feedback conversation message is dynamically presented in a streaming display manner, which not only conforms to the natural reading habit of the user, but also improves the smoothness of interaction through progressive information release; second, the object recommendation reason information that integrates the user portrait, comments and other multi-dimensional data is intelligently displayed beside the object information, which significantly enhances the transparency and persuasiveness of the recommended object; third, through the double-layer interactive design of the demand description prompt identifier and the object display panel, the user can quickly adjust the recommendation result and deeply view the object details, forming a three-dimensional decision support; finally, the first object of the instant recommendation is organically combined with the second object of the extended set, which ensures the accuracy of the first-screen information and retains the possibility of the user exploring more options, thereby balancing the recommendation efficiency and the degree of freedom of selection, and improving the efficiency of obtaining object information and the quality of decision-making in a complex demand scenario.

[0135] Figure 6 The flowchart of the information interaction method provided by the embodiment of the application can be further refined on the basis of the foregoing embodiment, and the specific implementation can be referred to the detailed description of the embodiment. The scheme provided by the embodiment of the application can be integrated in an extended reality device to realize. Among them, the same or corresponding technical terms as the above embodiments are not described in detail in this embodiment. As shown in the following Figure 6 The method specifically includes the following steps:

[0136] S310, receiving a user conversation message input by a user.

[0137] The user conversation message corresponds to a user demand description.

[0138] S320, obtaining a feedback conversation message corresponding to the user conversation message, and displaying the feedback conversation message in the current conversation interface.

[0139] The feedback conversation message is used to represent the demand analysis result of the user demand description, and the demand analysis result at least includes demand analysis text information and an additional description prompt information for indicating that the user adds a question.

[0140] S330, in response to a triggering operation of the user on the additional description prompt information and / or an event of inputting a conversation message in an edit box of the current conversation interface, generating a new user conversation message.

[0141] In this embodiment, when the user directly clicks the additional description prompt information provided by the feedback session message, the additional description prompt information may be, for example, a preset follow-up question button or a suggested question at this time, and the prompt content can be automatically converted into a new user session message at this time; or when the user actively enters new content in the input box of the chat interface and sends it, it will also be taken as a new user session message; or the user triggers the additional description prompt information and also enters new content in the input box of the chat interface, and these information contents can also be taken as a new user session message.

[0142] S340, by processing the new user session message and the previous feedback session message, determine the feedback session message corresponding to the new user session message; or, by processing the new user session message and the historical session message in the current session interface, determine the feedback session message corresponding to the new user session message.

[0143] Among them, the previous feedback session message specifically refers to the instant response content generated by the latest input of the user in the current interaction process, which represents the most adjacent context node in the dialogue thread. The historical session message is all past interaction records accumulated in the current session interface. The historical session message at least includes historical user session message and / or historical feedback session message. The historical session message contains the queries / instructions submitted by the user before, and also contains historical feedback session messages. These data together constitute a complete dialogue context, which is used to maintain semantic coherence and consistency of demand understanding in multiple rounds of interaction.

[0144] In this embodiment, the two context processing mechanisms for generating the feedback session message corresponding to the new user session message include the following two ways: the first is the short context mode, which can associate and analyze the new user session message with the last system reply, that is, the previous feedback session message, to generate the feedback session message corresponding to the new user session message. This type of way is suitable for simple and explicit follow-up questions; the second is the long context mode, which can comprehensively analyze the new message and all historical records accumulated in the current session (including user past questions and system historical replies) to generate the feedback session message corresponding to the new user session message. This type of way is suitable for complex requirements that need to be understood in combination with multiple rounds of dialogue. Both of these two processing methods can ensure that the system maintains dialogue coherence when generating new replies, and the first one focuses on recent dialogue focus, and the second one considers complete dialogue history.

[0145] The technical scheme provided by the embodiment of the application, when determining the feedback conversation message corresponding to the new user conversation message, first, the user operation threshold is significantly reduced and the interaction flexibility is improved by supporting multiple interaction modes such as a click-to-add description prompt information and / or manual input. Second, the short context or long context dual-mode processing mechanism is adopted, which not only guarantees the response efficiency in a simple scenario, but also accurately understands complex requirements through comprehensive analysis of complete dialogue history, so that the semantic understanding continuity and accuracy can be maintained in different scenarios. Finally, by dynamically integrating historical user messages and feedback messages, a progressive dialogue context is constructed, so that the user intent capture in multi-round interaction is more comprehensive, and progressive recommendation object optimization is further realized, thereby improving the adaptation degree between the display object and the user.

[0146] Figure 7 The information interaction method flowchart provided by the embodiment of the application is based on the foregoing embodiment, and the determination method of the feedback conversation message is further described in detail. The specific implementation can be referred to the detailed description of the embodiment. The same or corresponding technical terms as the foregoing embodiment are not described herein.

[0147] As shown in FIG. 7, the method comprises the following steps:

[0148] S410, receiving a user conversation message input by a user. The user conversation message corresponds to a user demand description.

[0149] S420, determining a conversation intent of the user according to the user conversation message.

[0150] The conversation intent of the user refers to a core demand target of the user identified from the user conversation message by a natural language processing technology, such as purchase decision, parameter comparison, or after-sales consultation information.

[0151] In the embodiment, the user conversation message can be analyzed and processed based on an intent analysis model to determine the conversation intent. The specific implementation can include: pre-processing, performing word segmentation, error correction, and noise removal processing on the original user conversation message input by the user; data loading, retrieving a historical behavior feature from a user portrait database and loading a domain knowledge graph; multi-model collaborative analysis, using an intent classification model to determine a basic intent category, extracting key parameters through an entity recognition model, and combining real-time context to calibrate the intent; strategy execution: according to a promotion mechanism, the weight of a high-frequency intent is improved, and finally a structured conversation intent with a confidence is output, which provides a directional guide for subsequent personalized recommendation.

[0152] S430, determining a feedback conversation message according to the conversation intent and a user portrait of the user.

[0153] The user portrait of the user refers to a preference feature model constructed according to a user historical behavior, and the user historical behavior may be, for example, browsing records, purchase records, evaluation records, and the like.

[0154] In the embodiment, the base response framework can be selected according to the session intent, and the user portrait can be combined for personalized optimization, and finally the generated feedback session message meets the functional requirements of the current intent and penetrates the long-term behavior characteristics of the user, so that the recommendation result has both scene adaptability and personal preference fitting degree, and the user satisfaction is improved.

[0155] Optionally, according to the session intent and the user portrait of the user, the specific implementation manner of the feedback session message can include: calling a target processing manner corresponding to the session intent; and analyzing and processing the user session message and the user portrait based on the target processing manner to determine the feedback session message.

[0156] The target processing manner refers to a special processing strategy and algorithm combination dynamically selected according to the session intent, and the essence is a differentiated service process predefined for different intent types.

[0157] In the embodiment, the target processing manner can be called from the strategy library through the session intent, and then the session intent and the user portrait are analyzed based on the manner, and finally the feedback session message is generated. For example, when the session intent is identified as a "price comparison intent", the target processing manner can be a price comparison engine combined with parameter comparison processing manner; if the session intent is an "after-sales consultation intent", the target processing manner can be to enable a work order system interface and a frequently asked question knowledge base matching. These processing manners are cooperatively configured through a rule engine and a machine learning model to ensure that the user session message and the portrait data of the user demand are analyzed in a targeted manner, and finally the feedback session message meeting the scene characteristics is generated, and accurate translation from intent recognition to service landing is realized.

[0158] On the basis of the above-mentioned embodiments, optionally, based on the session intent being an object acquisition intent, the object acquisition intent including a commodity acquisition intent and / or a service acquisition intent, the implementation manner of the feedback session message determined by analyzing and processing the user session message and the user portrait based on the target processing manner can include: rewriting understanding the user session message to obtain first information; generating demand analysis text information, a first preset number of first objects, and additional description prompt information based on the first information and the user portrait.

[0159] Among them, object acquisition intent refers to the core purpose of obtaining a certain type of entity object clearly expressed in the user session, which is divided into two subcategories: product acquisition intent and service acquisition intent. Among them, product acquisition intent refers to the purchase demand for specific products, for example, "want to buy a camera" or "recommended medicines." Service acquisition intent refers to the non-physical services provided by the platform, for example, "health consultation", "appointment for mobile phone repair", "consultation on return and exchange policies", etc. The first information refers to the standardized demand expression generated by semantic parsing and structured rewriting of the user session message originally input by the user, which converts fuzzy natural language into precise query conditions that can be processed by machines.

[0160] In this embodiment, when the conversation intent is determined to be an object acquisition intent, such as a product or service acquisition intent, semantic parsing is first used to convert the user's original request into structured primary information. For example, a user conversation message such as "I want a phone with good camera performance" would be rewritten as {Functional requirement: camera performance; Key parameters: [Main camera pixel ≥ 50 million, optical image stabilization support]}. Furthermore, a multi-stage process can be initiated based on the user profile. First, the recommendation logic can be explained in natural language to generate a demand analysis text message; second, the optimal first object can be selected from a candidate pool according to a first preset number; and third, guidance options are generated based on unclear parameters, generating additional descriptive prompts. Through this process of demand translation, profile integration, and gradual clarification, the final feedback conversation message meets both intent accuracy and personalization requirements.

[0161] It should be noted that the first object can be determined based on the user profile, the first information, the object information, and the preset display attributes. Specifically, the first object can be determined through a four-fold data fusion: for example, the user profile provides preference weights; the first information clarifies core needs; the object information screens qualified candidates; the preset display attributes control the presentation rules; and finally, a weighted algorithm is used to sort the products, ensuring that the first recommended display object meets both the user's explicit needs and implicit preferences, while also satisfying the platform's operating strategy.

[0162] Based on the above embodiment, optionally, the additional description prompt information is determined based on the following method: based on the object attribute information, comment information and user conversation message of the second object, at least one optional item is determined to generate the additional description prompt information based on the at least one optional item.

[0163] The object attribute information of the second object refers to inherent characteristic data of the candidate goods or services not directly displayed on the first screen. For example, hardware parameters such as CPU model and battery capacity of a notebook computer, or service indicators such as covered cities and hourly price of maintenance services; the comment information is the evaluation content generated by other users on these second objects. The optional item refers to an interactive filtering condition or follow-up item dynamically generated based on the user's potential decision dimension, which is derived from the analysis of the second object characteristics and the user's conversation message.

[0164] In this embodiment, the potential decision dimension can be automatically identified by the object attribute information and the comment information of all second objects, combined with the original demand of the user conversation message, to generate interactive optional items as additional description prompt information. For example, the user's core demand can be extracted as a root node based on the user conversation message, such as "light and thin notebook with a budget of 5k". The first layer of the tree diagram filters out the basic matching items through the attribute information of the second object, such as the CPU / weight / price of the notebook. The second layer of the tree diagram filters the reputation through the comment information, such as "poor battery life" and "good heat dissipation", and retains the attribute dimensions with high frequency and meeting the demand. The third layer of the tree diagram cross-analyzes the results of the previous two layers, eliminates attribute conflict items such as "light and thin but poor heat dissipation", and finally generates optional items sorted by matching degree, each item being attached with key attributes and comment summaries as the basis for generating additional description prompt information. In this way, the user can quickly narrow down the recommended range by selecting these one-key trigger operations of the optional items, accurately filtering the million-level goods library to dozens of most matching ones, while ensuring that the clarification questions are highly related to the user's initial demand, avoiding irrelevant questioning interference in the decision-making process.

[0165] Further, the method further comprises: in response to the trigger operation of replacing the first optional item in the additional description prompt information, adjusting at least one first optional item in the additional description prompt information.

[0166] The first optional item refers to an interactive preset option or filtering condition provided in the additional description prompt information, which is dynamically generated based on the intelligent analysis of the user's initial demand.

[0167] In this embodiment, a preset control for replacing the first optional item can be configured in the feedback conversation message, for example, Figure 2The "change a batch" control is shown. When the user triggers the "change a batch" control, one or more first selectable items in the additional description prompt information can be adjusted in real time. On the basis of the above example, when the user triggers the "change a batch" control, the selectable items can be queried level by level downward according to the hierarchical relationship of the tree diagram, as the adjusted first selectable items. On this basis, when the user triggers the at least one updated first selectable item, the structured options can be converted into machine-readable user session messages as input criteria for a new round of recommendations. Through this dynamically adjustable option design, the cost of user input is reduced, and the accuracy of demand expression is ensured.

[0168] S440, obtaining a feedback session message corresponding to the user session message, and displaying the feedback session message in the current session interface.

[0169] The feedback session message is used to represent the demand analysis result of the user demand description, and the demand analysis result at least includes demand analysis text information and additional description prompt information for indicating user additional questions.

[0170] S450, in response to receiving a new user session message input by the user according to the additional description prompt information, displaying a feedback session message corresponding to the new user session message in the session interface.

[0171] The technical scheme provided by the embodiment determines the session intention of the user according to the user session message, and then determines the feedback session message according to the session intention and the user portrait of the user. The feedback session message generated in this way not only meets the functional requirements of the current intention, but also penetrates the long-term behavior characteristics of the user, so that the recommendation result has both scene adaptability and personal preference fitting degree, and the user satisfaction is improved.

[0172] Next, the information interaction method provided by the embodiment of the application can be described with two specific examples.

[0173] Example one: the information interaction implementation process guided by the commodity acquisition intention. The user portrait information is: A region, male, 50 years old, hypertension history, daily taking A brand sustained-release tablets to lower blood pressure, X shopping platform new user, to find medicine. Fig. 8(a) shows the feedback session message schematic diagram after the user inputs the user session message. The user input user session message is "A brand sustained-release tablets". At this time, the feedback session message corresponding to the user session message can be obtained, and the feedback session message is displayed in the current session interface. The feedback session message includes demand analysis text information, at least one first medicine available for the user to obtain, and additional description prompt information for indicating the user to add questions. Fig. 8(b) shows the interface schematic diagram of displaying more commodities. When the user triggers the "view all" control in the feedback session message, all second objects can be displayed. Fig. 8(c) shows the information display page schematic diagram of the user triggering the additional description prompt information. When the user edits the additional description prompt information "D brand, 10 boxes" in the edit box in the feedback session message, the first medicine can be further filtered according to the additional description prompt information, and the first medicine in the feedback session message is updated. Subsequently, the user can view the detail page of the triggered first medicine in the current page, trigger the purchase control, and then the object acquisition progress can be displayed in real time. After purchasing the triggered medicine, information content related to the first medicine can be pushed to the user. Fig. 8(d) shows the display page schematic diagram of the pushed content, which is the remaining service items related to the user session content. Fig. 8(e) shows the display page schematic diagram of the pushed content, which is the medication guide related to the purchased medicine. Fig. 8(f) shows the display page schematic diagram of the pushed content, which is the price reduction reminder and the repurchase reminder related to the purchased medicine.

[0174] Example two: information interaction implementation process guided by service acquisition intention. User portrait information: A region, male, 33 years old, engaged in V professional, 5-year-old child at home, cough for 2 days, because of the recent children's disease intensive, worry about cross infection in the hospital, so did not go to the hospital, to the children today to 38 degrees, come for health consultation. Fig. 9(a) shows the feedback session message schematic diagram after the user inputs the user session message, the user input user session message is "5-year-old child, how to deal with sudden fever", at this time, the feedback session message corresponding to the user session message can be obtained and displayed in the current session interface. Fig. 9(b) shows the information display page schematic diagram of the user triggering the additional description prompt information, when the user triggers to edit the additional description prompt information "influenza detection" in the feedback session message, the user can further display the selectable service items according to the additional description prompt information. Subsequently, the user can purchase the triggered service item in the shopping platform. Fig. 9(c) shows the service progress information display page schematic diagram after the user purchases a certain service item, when the user triggers to edit the additional description prompt information in the feedback session message. Fig. 9(d) shows the display page schematic diagram of the detection report generation and subsequent suggestion.

[0175] As an optional embodiment of the above embodiment, the background execution logic of the information interaction method provided by the embodiment is shown in Figure 10 . As shown in Figure 10 , the background execution logic of the information interaction method is specifically: on the basis of obtaining the user portrait data, when the user inputs the demand text, the user session message is generated at this time, so that the user session message can be subjected to intention recognition, and the feedback session message can be generated according to the intention recognition result and the user portrait data. Before displaying the feedback session message, the reply loading state can be temporarily displayed in the current display page, and the generated content can be gradually displayed in the form of streaming display in the current display page during the generation of the feedback session message. Among them, the feedback session message includes demand analysis text information, at least one first object available for the user to acquire, and additional description prompt information for indicating the user to add questions. When the user triggers or edits the additional description prompt information, the new user session message is generated at this time, the new user session message and / or the historical session message can be used as the user session message, and the feedback session message can be determined based on the user session message. During the continuous generation of the feedback session message, the follow-up question loading state can be temporarily displayed in the current display page, and then the above steps can be repeatedly executed. Finally, the understanding of the user's demand can be gradually refined, each new interaction is based on the accumulation of the previous dialogue, the accuracy of the demand analysis result can be continuously optimized, and the goods or service items suitable for the user can be gradually screened out.

[0176] The following is an embodiment of the information interaction device provided by the embodiment of the application. The device and the information interaction method of each embodiment described above belong to the same inventive concept. Details not described in the embodiment of the information interaction device can be referred to the embodiment of the information interaction method

[0177] Figure 11 The structure diagram of the information interaction device provided by the embodiment of the application is shown in the figure. The device specifically includes a user session receiving module 510, a feedback session display module 520, and a new session processing module 530.

[0178] The user session receiving module 510 is configured to receive a user session message input by a user. The user session message corresponds to a user demand description. The feedback session display module 520 is configured to obtain a feedback session message corresponding to the user session message and display the feedback session message in a current session interface. The feedback session message is used to represent a demand analysis result of the user demand description. The demand analysis result at least includes demand analysis text information and additional description prompt information for indicating that the user adds a question. The new session processing module 530 is configured to, in response to receiving a new user session message input by the user according to the additional description prompt information, display a feedback session message corresponding to the new user session message in the session interface.

[0179] The technical solution provided by the embodiment of the application can obtain a feedback session message corresponding to a user session message when receiving a user session message input by a user and representing a user demand description. The feedback session message is displayed in a current session interface. The feedback session message is used to represent a demand analysis result of the user demand description. The demand analysis result at least includes demand analysis text information and additional description prompt information for indicating that the user adds a question. In response to receiving a new user session message input by the user according to the additional description prompt information, a feedback session message corresponding to the new user session message is displayed in the session interface. The technical problem of low matching efficiency between a large number of goods and user demand in an e-commerce scenario is effectively solved. The user demand description is analyzed and processed to be converted into structured data of implicit demand. The intention understanding accuracy is improved. The recommendation logic is intuitively displayed through the demand analysis text information. The recommendation direction is dynamically corrected in combination with the additional description prompt information. Gradual recommendation object optimization is achieved. Therefore, the adaptation degree between the display object and the user is improved.

[0180] On the basis of each technical solution described above, the feedback session display module 520 is further configured to display the demand analysis result in the feedback session message in a streaming form in the current session interface.

[0181] On the basis of each of the technical solutions above, optionally, the requirement analysis result of the feedback session message comprises at least one first object available for the user to acquire, and the feedback session display module 520 is further configured to display object information of at least one first object in a streaming manner after the text information of the requirement analysis is completed, and display object recommendation reason information of the first object at an associated position of the object information in the current session interface.

[0182] The object recommendation reason information is determined based on one or more of object comment information, object attribute information, user session messages, and the user portrait of the first object.

[0183] On the basis of each of the technical solutions above, optionally, in the case where the feedback session message comprises the first object, the feedback session message further comprises a requirement description prompt identifier, and the requirement description prompt identifier is used to represent an acquisition requirement of the user for the first object. The feedback session display module 520 is further configured to adjust the first object displayed in the feedback session message in response to a triggering operation on the requirement description prompt identifier.

[0184] On the basis of each of the technical solutions above, optionally, the feedback session message further comprises an object set filtered based on the user session messages, and the object set comprises at least an object total number of a filtered second object and a viewing identifier of the second object. The information interaction apparatus further comprises:

[0185] An object page display module, configured to display an object display page comprising the second object in response to a triggering operation on the viewing identifier. The second object comprises the first object.

[0186] On the basis of each of the technical solutions above, optionally, the apparatus further comprises:

[0187] An object panel display module, configured to display an object display panel in the current session interface in response to a triggering operation on any first object, so as to present object detail information and / or object acquisition information of the triggered first object based on the object display panel.

[0188] On the basis of each of the technical solutions above, optionally, the apparatus further comprises:

[0189] An additional session generation unit, configured to generate the additional user session message in response to a triggering operation of the user on the additional description prompt information and / or an event of inputting a session message in an edit box of the current session interface.

[0190] The first conversation processing unit is configured to determine the feedback conversation message corresponding to the new user conversation message by processing the new user conversation message and a previous feedback conversation message.

[0191] The second conversation processing unit is configured to determine the feedback conversation message corresponding to the new user conversation message by processing the new user conversation message and historical conversation messages in the current conversation interface.

[0192] The historical conversation messages at least include historical user conversation messages and / or historical feedback conversation messages.

[0193] On the basis of the above technical solutions, the device further includes:

[0194] The additional description adjustment module is configured to adjust at least one first selectable item in the additional description prompt information in response to a triggering operation of selecting the first selectable item in the additional description prompt information.

[0195] On the basis of the above technical solutions, the device further includes a feedback conversation determination module, and the feedback conversation determination module includes:

[0196] The conversation intention determination unit is configured to determine the conversation intention of the user according to the user conversation message.

[0197] The feedback conversation determination unit is configured to determine the feedback conversation message according to the conversation intention and a user portrait of the user.

[0198] On the basis of the above technical solutions, the feedback conversation determination unit is specifically configured to retrieve a target processing mode corresponding to the conversation intention, analyze and process the user conversation message and the user portrait based on the target processing mode, and determine the feedback conversation message.

[0199] On the basis of the above technical solutions, the conversation intention is an object acquisition intention, the object acquisition intention includes a commodity acquisition intention and / or a service acquisition intention, and the feedback conversation determination unit is specifically configured to perform rewritten understanding on the user conversation message to obtain first information, generate demand analysis text information, a first preset number of first objects, and an additional description prompt information based on the first information and the user portrait.

[0200] On the basis of the above technical solutions, the feedback conversation determination module further includes:

[0201] The additional description prompt information determination unit is configured to determine at least one selectable item based on object attribute information, comment information of a second object, and the user conversation message, so as to generate the additional description prompt information based on the at least one selectable item.

[0202] In the above technical solutions, optionally, the device further comprises:

[0203] The progress display module is configured to, in response to detecting the object acquisition event of acquiring the first object, record the object acquisition progress of the acquired first object, and display the object acquisition progress.

[0204] The information interaction device provided by the embodiments of the present application can execute the information interaction method provided by any of the embodiments of the present application, and has the corresponding function modules and beneficial effects of executing the information interaction method.

[0205] It should be noted that in the above embodiments of the information interaction device, each unit and module included is only divided according to the function logic, but is not limited to the above division, as long as the corresponding function can be realized; in addition, the specific names of each functional unit are only for the convenience of mutual differentiation, and do not serve to limit the protection scope of the present application.

[0206] Figure 12 A structural schematic diagram of an electronic device provided by the embodiments of the present application is provided. Figure 12 A block diagram of an exemplary electronic device 12 suitable for implementing embodiments of the present application is shown. Figure 12 The electronic device 12 shown is merely an example, and should not impose any limitation on the functions and use range of the embodiments of the present application.

[0207] As shown in Figure 12 The electronic device 12 is shown in the form of a general computing device. The components of the electronic device 12 can include, but are not limited to, one or more processors or processing units 16, system memory 28, and a bus 18 that connects the various system components, including the system memory 28 and the processing unit 16.

[0208] The bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration bus, a processor or local bus using any of a variety of bus architectures. By way of example, these architectures include Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.

[0209] The electronic device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the electronic device 12 and includes both volatile and nonvolatile media, removable and non-removable media.

[0210] The system memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The electronic device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be configured to read and write non-removable, non-volatile magnetic media ( Figure 12 Not shown, often called a "hard drive"). Although Figure 12 Not shown, a magnetic disk drive for reading and writing to a removable non-volatile magnetic disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 18 via one or more data media interfaces. System memory 28 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of various embodiments of the present invention.

[0211] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28. Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may include an implementation of a network environment. Program modules 42 generally perform the functions and / or methods of the embodiments described herein.

[0212] The electronic device 12 can also communicate with one or more external devices 14 (e.g., a keyboard, pointing device, display 24, etc.), one or more devices that enable a user to interact with the electronic device 12, and / or any device that enables the electronic device 12 to communicate with one or more other computing devices (e.g., a network card, a modem, etc.). Such communication can occur via an input / output (I / O) interface 22. Furthermore, the electronic device 12 can communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network such as the Internet) via a network adapter 20. As shown, the network adapter 20 communicates with other modules of the electronic device 12 via a bus 18. It should be understood that, although not shown, other hardware and / or software modules can be used in conjunction with the electronic device 12, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0213] The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, such as implementing the information interaction method steps provided in the first embodiment of the present invention, which includes:

[0214] receiving a user session message input by a user; wherein the user session message corresponds to a user demand description;

[0215] obtaining a feedback session message corresponding to the user session message, and displaying the feedback session message in a current session interface; wherein the feedback session message is used to represent a demand analysis result of the user demand description, and the demand analysis result at least includes demand analysis text information and additional description prompt information used to indicate that the user adds a question;

[0216] in response to receiving a new user session message input by the user according to the additional description prompt information, displaying a feedback session message corresponding to the new user session message in the session interface.

[0217] Of course, those skilled in the art can understand that the processor can also implement the technical solutions of the information interaction method provided by any embodiment of the application.

[0218] The embodiment provides a computer readable storage medium, and a computer program is stored on the computer readable storage medium. The program is executed by a processor to implement the steps of the information interaction method provided by the foregoing embodiments of the application. The method comprises the following steps:

[0219] receiving a user session message input by a user; wherein the user session message corresponds to a user demand description;

[0220] obtaining a feedback session message corresponding to the user session message, and displaying the feedback session message in a current session interface; wherein the feedback session message is used to represent a demand analysis result of the user demand description, and the demand analysis result at least includes demand analysis text information and additional description prompt information used to indicate that the user adds a question;

[0221] in response to receiving a new user session message input by the user according to the additional description prompt information, displaying a feedback session message corresponding to the new user session message in the session interface.

[0222] The computer storage medium of the embodiments of the present application can adopt any combination of one or more computer readable media. The computer readable medium can be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium may, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination thereof. More specific examples (non-exhaustive list) of the computer readable storage medium include an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, the computer readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus or device.

[0223] The computer readable signal medium can include a data signal propagated in baseband or propagated as a carrier wave, in which computer readable program code is embodied. Such propagated data signals can take a wide variety of forms, including but not limited to electro-magnetic signals, optical signals, or any suitable combination thereof. The computer readable signal medium can also be any computer readable medium that is not a storage medium, that is capable of storing the program for use by or in connection with the instruction execution system, apparatus or device.

[0224] The program code embodied on the computer readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wire line, optical fiber cable, RF, etc., or any suitable combination of the above.

[0225] The computer program code for carrying out operations of the embodiments of the present application can be written in one or more programming languages or combinations of languages including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages such as "C" or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0226] Note that the above merely describes preferred embodiments of the present application and the principles of the technology applied. Those skilled in the art will understand that the present application is not limited to the specific embodiments described herein, and that various obvious changes, modifications and substitutions can be made without departing from the scope of the present application. Therefore, although the present application has been described in detail through the above embodiments, the present application is not limited to the above embodiments, and can include more other equivalent embodiments without departing from the concept of the present application, and the scope of the present application is determined by the scope of the claims.

Claims

1. An information interaction method, characterized in that: include: Receiving a user conversation message input by a user; wherein the user conversation message corresponds to a user demand description; Obtaining a feedback conversation message corresponding to the user conversation message and displaying the feedback conversation message in the current conversation interface; wherein the feedback conversation message is used to represent the demand analysis results of the user demand description, and the demand analysis results include at least demand analysis text information and additional description prompt information for instructing the user to ask additional questions; In response to receiving a new user conversation message input by the user according to the additional description prompt information, a feedback conversation message corresponding to the new user conversation message is displayed in the conversation interface.

2. The method according to claim 1, characterized in that The displaying of the feedback session message in the current session interface includes: The demand analysis results in the feedback session messages are sequentially displayed in a streaming manner in the current session interface.

3. The method according to claim 1, characterized in that The demand analysis result of the feedback conversation message includes at least one first object available to the user, and the displaying of the feedback conversation message in the current conversation interface includes: After the demand analysis text information is displayed in a streaming form in the current conversation interface, object information of at least one first object is displayed in sequence, and object recommendation reason information of the first object is displayed at a location associated with the object information; The object recommendation reason information is determined based on one or more of the object comment information, object attribute information, user conversation message and the user portrait of the first object.

4. The method according to claim 1, wherein When the feedback session message includes the first object, the feedback session message also includes a demand description prompt identifier, where the demand description prompt identifier is used to represent the user's demand for obtaining the first object. Displaying the feedback session message in the current session interface includes: In response to a triggering operation on the requirement description prompt identifier, a first object displayed in the feedback session message is adjusted.

5. The method according to claim 1, wherein The feedback conversation message also includes a set of objects filtered out based on the user conversation message, the set of objects including at least the total number of the filtered out second objects and a viewing identifier for viewing the second objects. The method further includes: In response to a triggering operation on the viewing identifier, displaying an object display page including the second object; The second object includes the first object.

6. The method according to any one of claims 1 to 5, characterized in that: The method further comprises: In response to a triggering operation on any first object, an object display panel is displayed in the current session interface to present object detail information and / or object acquisition information of the triggered first object based on the object display panel.

7. The method according to any one of claims 1 to 5, characterized in that In response to receiving the new user conversation message input by the user according to the additional description prompt information, displaying the feedback conversation message corresponding to the new user conversation message in the conversation interface includes: In response to the user triggering the additional description prompt information and / or inputting a conversation message in the edit box of the current conversation interface, generating the newly added user conversation message; Determine the feedback session message corresponding to the new user session message by processing the new user session message and the previous feedback session message; or, Determining a feedback session message corresponding to the new user session message by comparing the new user session message and the historical session messages in the current session interface; The historical session messages at least include historical user session messages and / or historical feedback session messages.

8. The method according to any one of claims 1 to 5, characterized in that: The method further comprises: In response to a triggering operation of replacing a first optional item in the additional description prompt information, at least one first optional item in the additional description prompt information is adjusted.

9. The method according to claim 1, characterized in that The feedback session message is determined based on the following method: Determining the user's conversation intention based on the user conversation message; The feedback conversation message is determined according to the conversation intention and the user profile of the user.

10. The method according to claim 9, characterized in that The determining the feedback conversation message according to the conversation intention and the user profile of the user includes: Retrieve a target processing method corresponding to the conversation intent; The user conversation message and the user portrait are analyzed and processed based on the target processing mode to determine the feedback conversation message.

11. The method according to claim 10, characterized in that The conversation intention is an object acquisition intention, which includes a product acquisition intention and / or a service acquisition intention. The analyzing and processing the user conversation message and the user portrait based on the target processing method to determine the feedback conversation message includes: Rewriting and understanding the user conversation message to obtain first information; Based on the first information and the user portrait, demand analysis text information, a first preset number of first objects, and additional description prompt information are generated.

12. The method according to claim 11, characterized in that The additional description prompt information is determined based on the following method: At least one optional item is determined based on the object attribute information, comment information, and the user conversation message of the second object, so as to generate the additional description prompt information based on the at least one optional item.

13. The method according to claim 1, wherein The method further comprises: In response to detecting an object acquisition event for acquiring a first object, recording an object acquisition progress of the acquired first object; Displays the progress of obtaining the object.

14. An information interaction device, characterized in that: include: A user conversation receiving module, configured to receive a user conversation message input by a user; wherein the user conversation message corresponds to a user demand description; A feedback session display module is configured to obtain feedback session messages corresponding to the user session messages and display the feedback session messages in the current session interface; wherein the feedback session messages are used to represent the demand analysis results of the user demand description, and the demand analysis results include at least demand analysis text information and additional description prompt information for instructing the user to ask additional questions; The new session processing module is configured to, in response to receiving a new user session message input by the user according to the additional description prompt information, display a feedback session message corresponding to the new user session message in the session interface.

15. An electronic device, characterized in that: The electronic device comprises: one or more processors; a memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the information interaction method according to any one of claims 1 to 13.

16. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the information interaction method according to any one of claims 1 to 13 is implemented.

17. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the computer program implements the information interaction method according to any one of claims 1 to 13.

Images