Article search method and device, computer device, and storage medium

Abstract

The present disclosure provides an article search method and device, computer equipment and a storage medium, and belongs to the technical field of networks. The technical scheme provided by the present disclosure can analyze the historical interaction behavior of the object by the search agent through a large model to infer the current possible search intention of the object when the object's stay behavior on the article card meets the preset condition, that is, the object has a certain article search intention. The search information that meets the search intention is displayed on the search page of the search agent. The search information can meet the search intention of the object, so that the object does not need to jump to the article search page and manually input search words to search, which can reduce the interaction operation. When the object's stay behavior on the article card meets the preset condition, the search page of the search agent is displayed, and the search page of the search agent is displayed without the object's click operation and other interaction operations, which further reduces the interaction operation and improves the human-computer interaction efficiency.

Description

Technical Field

[0001] This disclosure relates to the field of network technology, and in particular to a method, apparatus, computer device, and storage medium for searching items. Background Technology

[0002] With the development of network technology, more and more applications are able to provide users with video browsing functions, and can even recommend items in the form of item cards within the video browsing function. When an item card inspires a user to want to view more items, the user can then perform a search. This process can be as follows: the user needs to jump from the currently viewed item card to the item search page, manually enter search terms on the item search page, and view the search results on the search results page. However, the above-described item search process is cumbersome and has low human-computer interaction efficiency. Summary of the Invention

[0003] This disclosure provides a method, apparatus, computer device, and storage medium for searching items, which can improve the efficiency of human-computer interaction. The technical solution of this disclosure is as follows: According to one aspect of the embodiments of this disclosure, an item search method is provided, including: The playback page displays item cards, which provide item information. If the object's dwelling behavior on the item card meets preset conditions, based on the object's search intent, the search agent's search page is displayed. The search page is used by the search agent to provide search information that matches the search intent, which is obtained by the search agent calling a large model to process the object's historical interaction behavior.

[0004] According to one aspect of the embodiments of this disclosure, an item search device is provided, comprising: The playback page display module is configured to display item cards on the playback page, the item cards being used to provide item information; The search page display module is configured to display the search page of the search agent based on the search intent of the object if the object's dwell behavior on the item card meets preset conditions. The search page is used by the search agent to provide search information that matches the search intent, which is obtained by the search agent calling a large model to process the object's historical interaction behavior.

[0005] In some embodiments, the object's dwelling behavior on the item card satisfies a preset condition, which means any of the following situations: The object stays on the item card for a first preset time and does not perform any operation during the stay; The object's dwell time on the item card reaches the countdown time; The object stays on the item card and the content progress of the item card reaches the target progress.

[0006] In some embodiments, the progress of the item card reaching the target progress includes any of the following situations: The video playback progress of the item card has reached the target progress; The browsing progress of the text and image content in the item card has reached the target progress.

[0007] In some embodiments, the playback page display module is configured to perform at least one of the following: A smart search entry is displayed on the item card. In response to the object's trigger operation on the smart search entry, the search page of the search agent is displayed based on the object's search intent. The smart search entry is used to invoke the search agent. In response to the object's pause operation on the video in the item card, the search page of the search agent is displayed based on the object's search intent; In response to a sliding operation of the object in a preset direction, the search page of the search agent is displayed based on the search intent of the object; In response to the terminal's motion characteristics matching preset motion characteristics, the search agent's search page is displayed based on the object's search intent.

[0008] In some embodiments, the smart search entry is displayed as a prompt bar in an item card, a top floating button, a bottom navigation bar, or a content page corner mark.

[0009] In some embodiments, the playback page display module is configured as follows: Search suggestions are displayed on the item card. These suggestions are generated by the search agent by processing the historical interaction behavior of the object using a large model.

[0010] In some embodiments, the search page display module is configured as follows: Based on the search intent of the object, a search suggestion page of the search agent is displayed, the search suggestion page includes search suggestion information that matches the search intent; In response to a confirmation operation of the search suggestion information, a search session page of the search agent is displayed, the search session page including search results obtained based on the search intent.

[0011] In some embodiments, the search suggestion information includes at least one of the following: The search suggestion text displayed on the item card is obtained by the search agent calling a large model to process the historical interaction behavior of the object; The search keywords are obtained by the search agent calling the large model to process the historical interaction behavior of the object.

[0012] In some embodiments, the search page display module is configured to perform any of the following: In response to the editing operation of the search suggestion information on the search suggestion page, the edited search suggestion information is obtained; in response to the confirmation operation of the edited search suggestion information, the search session page of the search agent is displayed. In response to an input operation on the search suggestion page, the input search information is obtained; in response to a confirmation operation on the search information, the search session page of the search agent is displayed.

[0013] In some embodiments, the search suggestion page further includes representative images of multiple recommended categories, each representative image being linked to the recommendation page of the corresponding recommended category, and the recommended categories being obtained by the search agent calling a large model to process the historical interaction behavior of the object; The device further includes: The recommendation page display module is configured to display recommended items belonging to the recommended category on the recommendation page of the recommended category in response to a trigger operation for a representative image of any recommended category.

[0014] In some embodiments, the search suggestion page is displayed in a half-screen format, a side-sliding panel format, a bottom pop-up format, or a full-screen overlay format.

[0015] In some embodiments, the apparatus further includes: The aggregation display page module is configured to display the platform's aggregation display page in response to a back operation on the search session page. The aggregation display page is used to display multiple items provided by the platform.

[0016] In some embodiments, the search page display module is configured as follows: If the object's behavior of staying on the item card meets the preset conditions and the object meets the search suggestion conditions, the search suggestion page will be displayed. The search suggestion conditions being met refer to: The object did not close the search suggestion page; Alternatively, the time interval between the current time and the last time the search suggestions page was closed reaches the target time interval.

[0017] In some embodiments, the target time interval is determined based on the activity level of the object, which is determined based on the object's behavior on the platform.

[0018] In some embodiments, the search page display module is configured as follows: Based on the search intent of the object, a search results page of the search agent is displayed, the search results page including search results obtained based on the search intent.

[0019] In some embodiments, the object’s historical interaction behavior includes at least one of browsing behavior, search behavior, and item conversion behavior.

[0020] According to another aspect of the embodiments of the present disclosure, a computer device is provided, the computer device comprising: One or more processors; Memory used to store the executable program code of the processor; The processor is configured to execute the program code to implement the aforementioned item search method.

[0021] According to another aspect of the present disclosure, a computer-readable storage medium is provided that, when program code in the computer-readable storage medium is executed by a processor of a computer device, enables the computer device to perform the above-described item search method.

[0022] According to another aspect of the present disclosure, a computer program product is provided, including a computer program / instructions that, when executed by a processor, implement the above-described item search method.

[0023] The technical solution provided in this disclosure, if the object's behavior of lingering on the item card meets preset conditions, that is, the object has a certain item search intention, the search agent analyzes the object's historical interaction behavior through a large model to infer the object's current possible search intention, and displays search information that matches the search intention on the search agent's search page. This search information can satisfy the object's search intention, so that the object does not need to jump to the item search page and manually enter search terms to search, which can reduce interaction operations. Moreover, the search agent's search page is displayed as soon as the object's behavior of lingering on the item card meets the preset conditions, without the need for the object to click or perform other interaction operations, further reducing interaction operations and improving human-computer interaction efficiency.

[0024] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0025] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure, and are not intended to unduly limit this disclosure.

[0026] Figure 1 This is a schematic diagram illustrating the implementation environment of an item search method according to an exemplary embodiment.

[0027] Figure 2 This is a flowchart illustrating an item search method according to an exemplary embodiment.

[0028] Figure 3 This is a schematic diagram illustrating a playback page according to an exemplary embodiment.

[0029] Figure 4 This is a schematic diagram illustrating a playback page according to an exemplary embodiment.

[0030] Figure 5 This is a schematic diagram illustrating a playback page according to an exemplary embodiment.

[0031] Figure 6 This is a schematic diagram illustrating a search suggestion page according to an exemplary embodiment.

[0032] Figure 7 This is a schematic diagram illustrating a search session page according to an exemplary embodiment.

[0033] Figure 8 This is a block diagram illustrating an item search device according to an exemplary embodiment.

[0034] Figure 9 This is a block diagram illustrating a terminal according to an exemplary embodiment. Detailed Implementation

[0035] To enable those skilled in the art to better understand the technical solutions of this disclosure, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0036] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0037] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, data stored, data displayed, etc.) and signals involved in this disclosure are all authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0038] Figure 1 This is a schematic diagram illustrating an implementation environment for an item search method according to an exemplary embodiment. See also: Figure 1 The implementation environment specifically includes: terminal 101 and server 102. Terminal 101 and server 102 can be connected directly or indirectly through wired or wireless communication, which is not limited herein.

[0039] Terminal 101 is one implementation of a computer device, and can be at least one of the following: a smartphone, a smartwatch, a desktop computer, a laptop, an MP3 player, an MP4 player, and a laptop computer. Terminal 101 runs an application that supports item search. Users can log in to the application through Terminal 101 to access the services provided by the application. For example, users can perform item searches through the application on Terminal 101.

[0040] Terminal 101 generally refers to one of a plurality of terminals; this embodiment uses terminal 101 as an example. Those skilled in the art will understand that the number of terminals can be more or less. For example, there may be several terminals, or dozens or hundreds of terminals, or even more. This disclosure does not limit the number of terminals or the type of device.

[0041] Server 102 is at least one of a single server, multiple servers, a cloud computing platform, and a virtualization center. Server 102 provides backend services for the item search application. Server 102 can connect to terminal 101 via a wireless or wired network, and server 102 can provide the terminal with item search services and database services, etc. In some embodiments, the number of servers may be more or fewer, and this disclosure does not limit this. Of course, server 102 may also include other functional servers to provide more comprehensive and diversified services.

[0042] Figure 2 This is a flowchart illustrating an item search method according to an exemplary embodiment, see [link to flowchart]. Figure 2 Taking a computer device as an example, the method includes the following steps: In step 201, item cards are displayed on the playback page. Item cards are used to provide item information.

[0043] The online platform can provide a playback page for playing videos and displaying item cards. In some optional methods, this approach can, after a preset browsing time on the playback page, determine the item the user prefers from the candidate items based on their attributes and historical interaction behavior (such as clicks, browsing, and conversions), and display the item card for that item. The item card can display item information, including the item's video, images, description, price, and service guarantees. The item card can also display details, a "Search More" control, and a "Not Interested" control. The terminal can respond to triggering the details control by displaying the item's details page; respond to triggering the "Search More" control by displaying an aggregated display page showcasing multiple items offered by the platform; and respond to triggering the "Not Interested" control by closing the item card.

[0044] In step 202, if the object's behavior of staying on the item card meets the preset conditions, the search page of the search agent is displayed based on the object's search intent. The search page is used by the search agent to provide search information that matches the search intent. The search intent is obtained by the search agent calling the large model to process the object's historical interaction behavior.

[0045] The object's historical interaction behavior can include one or more of browsing behavior, search behavior, and item conversion behavior. Browsing behavior can include watching video content on the platform, watching live streams on the platform, browsing item cards on the playback page, and browsing the search page of the search agent. Search behavior can include searching on the search list page and searching on the search agent's search page. It should be noted that the aforementioned historical interaction behavior is obtained after full user authorization. The terminal can provide authorization to the user through methods such as pop-up windows; this disclosure does not limit this approach.

[0046] The search intent can be obtained by the search agent calling the large model to process the object's historical interaction behavior, or it can be obtained by the search agent calling the large model to process the category corresponding to the item card and the object's historical interaction behavior. Here, the category corresponding to the item card refers to the category to which the item in the item card belongs. Since the object's dwelling behavior on the item card meets preset conditions, that is, the object has a certain willingness to learn about the item card, the large model can infer the object's search intent based on the category corresponding to the item card. This search intent can refer to the object's intent tag inferred by the large model based on the object's historical interaction behavior, and can include information such as the item category, brand, price, and style that the object wants to search for. Of course, the content included in this intent tag is not limited to these.

[0047] This large model is an artificial intelligence model. In some embodiments, it can be a multimodal large model, which refers to a large model capable of simultaneously processing and understanding multiple different modalities. It can be pre-trained on a large amount of multimodal data. In an optional implementation, the large model may include an encoder, a decoder, a self-attention layer, and a feed-forward neural network. The encoder is mainly used to convert input data (usually in sequence form) into vector representations. This process can capture the semantic features of the input data. The decoder is responsible for converting the intermediate representations generated by the encoder into output data (usually in sequence form). The self-attention layer is a mechanism that allows the model to pay attention to other positions in the sequence to better encode the information of the current position. The feed-forward neural network can perform nonlinear transformations on the output of the self-attention layer to enhance the model's expressive power. The various parts work together, enabling the model built on them to perform well in various complex processing tasks, such as natural language processing, computer vision, speech recognition, machine translation, text summarization, and intelligent question answering.

[0048] The technical solution provided in this disclosure, if the object's behavior of staying on the item card meets preset conditions, that is, the object has a certain item search intention, the search agent analyzes the object's historical interaction behavior through a large model to infer the object's current possible search intention, and displays search information that matches the search intention on the search agent's search page. This search information can satisfy the object's search intention, so that the object does not need to jump to the item search page and manually enter search terms to search, which can reduce interaction operations. Moreover, the search agent's search page is displayed as soon as the object's behavior of staying on the item card meets the preset conditions, without the need for the object to click or perform other interaction operations, which further reduces interaction operations and improves human-computer interaction efficiency.

[0049] In some embodiments, the object's dwelling behavior on an item card satisfies a preset condition if any of the following conditions are met: the object dwells on the item card for a duration equal to a first preset duration and there is no operation during the dwelling period; the object dwells on the item card for a duration equal to a countdown timer; or the object dwells on the item card and the content progress of the item card reaches a target progress. The object dwelling on the item card for a duration equal to the first preset duration and there is no operation during the dwelling period means: the object dwells on the item card for a duration equal to the first preset duration without any interactive operation. If the object performs any interactive operation, the dwelling timer is interrupted, the dwelling time is reset to zero, and dwelling time is only counted again when the object has no interactive operation. The object dwelling on the item card for a duration equal to the countdown timer can mean: regardless of whether the object has any interactive operation, the dwelling time on the item card reaches the countdown timer, and this dwelling time is the total duration from displaying the item card to the object leaving the item card. The countdown timer can be displayed on the item card itself. This can be done through the smart search entry on the item card, either as a numerical countdown or a progress bar. This provides a psychological buffer for the user, avoiding the abruptness of a sudden jump. The content of the item card can include text, images, and videos. Reaching the target progress for the item card's content includes either: the video playback progress reaching the target progress; or the browsing progress of the text and images on the item card reaching the target progress. Reaching the target progress for video playback means the video on the item card has finished playing. Reaching the target progress for browsing text and images means the browsing of the text and images on the item card has been completed.

[0050] In this embodiment of the disclosure, when the object's dwelling behavior meets the preset conditions, no additional triggering operation such as clicking is required from the object. The search page is proactively provided in a seamless and context-appropriate manner, achieving the goal of reaching the object during the critical window period when the object is dwelling on the item card and has a certain intention to search for items. This significantly reduces the object's operational threshold and cognitive burden, thereby improving the object's user experience and immersion, effectively increasing the exposure rate of the search page, helping to encourage the object to further search for items, and improving the item conversion rate.

[0051] In addition, as an alternative approach, this method can display a sliding prompt message after the object reaches the target progress of the item card content. In response to the object's preset sliding operation, the search agent's search page is displayed based on the object's search intent. The sliding prompt message can enhance the object's trust and help the object decide autonomously when it needs the search agent's service, thereby improving the object's operational certainty and interface predictability.

[0052] In some embodiments, the first preset duration and the countdown duration may be the same or different. The first preset duration or countdown duration can be personalized based on the object's historical interaction behavior. For example, based on different objects' historical dwell time on certain item cards, countdown completion rate, click operations, whether there was any active exit behavior, and historical dwell time on the search agent's search page, click operations, whether there was any active exit behavior, etc., a preset rule or machine learning model can be used to determine the patience level or waiting willingness intensity of different objects, and a countdown duration or first preset duration matching the patience level or waiting willingness intensity of each object can be determined. For example, for objects that have historically spent a short time on item cards but a long time on the search page, a shorter countdown duration can be set on the item cards to quickly display the search page. Compared with a fixed countdown duration, the embodiments of this disclosure that personalize the duration based on the object's historical interaction behavior can effectively reduce the waiting anxiety of objects with search intentions, improve the smoothness of the jump from item cards to the search page, avoid object churn due to uniformly long countdowns, and improve object retention rate.

[0053] In some embodiments, besides displaying the search agent's search page in response to other interactive operations of the object, the terminal may also display the search agent's search page in response to other interactive operations of the object. The method further includes at least one of the following: displaying a smart search entry on the item card; displaying the search agent's search page based on the object's search intent in response to the object's triggering operation on the smart search entry, wherein the smart search entry is used to invoke the search agent; displaying the search agent's search page based on the object's search intent in response to the object's pause operation on the video in the item card; displaying the search agent's search page based on the object's search intent in response to the object's swipe operation in a preset direction; and displaying the search agent's search page based on the object's search intent in response to the terminal's motion characteristics conforming to preset motion characteristics. The intelligent search entry can be displayed as a prompt bar in an item card, a top floating button, a bottom navigation bar, or a content page corner mark. The intelligent search entry can display a smart icon and search prompt text. This embodiment provides an intelligent search entry for an object to trigger, thereby displaying the search agent's search page, allowing the object to autonomously decide when it needs the search agent's service, improving the object's operational certainty and interface predictability. The search agent's search page is displayed when the video introducing the item in the item card is paused. At this time, the object is in a psychologically idle state, with a higher acceptance and openness to additional information, which helps increase the object's attention to the search page. The object's swiping operation in a preset direction can refer to swiping from the edge of the screen inwards. The system displays the search page of the search agent in response to the user's swipe gesture in a preset direction. This allows the user to actively access the search page by intuitively swiping when needed. This low-intrusion, high-autonomy approach significantly reduces the user's aversion to the search page. The system also displays the search page in response to the user's motion characteristics matching the preset motion characteristics. This leverages the immediacy and fun of physical actions, allowing the user to access the search page through a specific action without visual positioning or precise clicking. This novel, game-like interaction stimulates the user's curiosity and willingness to try, effectively attracting their attention.

[0054] In some embodiments, the terminal can display personalized search suggestions for different objects to stimulate their search interest. The method may further include displaying search suggestions on item cards, where the suggestions are generated by a search agent calling a large model to process the object's historical interaction behavior. The search suggestions can be displayed on the intelligent search entry point. The large model can process the object's historical interaction behavior to obtain the object's search intent, or it can process the category corresponding to the item card and the object's historical interaction behavior to obtain the object's search intent. The large model then generates search suggestions based on the search intent, reflecting the object's search intent, for example, a search suggestion like "Want to try a new wool sweater this fall / winter?" In this embodiment, different objects have different historical interaction behaviors, reflecting different preferences. Generating different search suggestions based on the different historical interaction behaviors of different objects can improve the targeting of the search suggestions, enhance their matching degree and attractiveness, encourage further item searches, and promote item conversion.

[0055] In some embodiments, displaying the search page in step 202 may refer to first displaying a search suggestion page to prompt the object to perform a search, and then displaying a search session page based on the object's confirmation of the search suggestion information on the search suggestion page. Correspondingly, displaying the search agent's search page based on the object's search intent in step 202 may include: displaying the search agent's search suggestion page based on the object's search intent, the search suggestion page including search suggestion information that matches the search intent; and displaying the search agent's search session page in response to the confirmation of the search suggestion information, the search session page including search results obtained based on the search intent. In this embodiment of the disclosure, a progressive path is formed by providing search suggestion information for the object to preview on the search suggestion page, and then displaying the search session page providing search results after the object confirms the search suggestion information. This progressive path from preview to object confirmation to display of results helps the object focus on the currently displayed page at each step, strengthens the object's sense of confirmation of its own operation, creates a controllable and coherent interaction rhythm, and allows the object to obtain a deeper level of immersive experience.

[0056] The search suggestion page can be displayed in a half-screen, side-sliding panel, bottom pop-up, or full-screen overlay format. Half-screen, side-sliding panel, and bottom pop-up formats do not completely obscure the playback page, reducing distraction, while the full-screen overlay enhances immersion. A close control can be displayed on the search suggestion page. The device can respond to the user's action of closing the control on the search suggestion page, returning to the playback page. This provides a clear path for the user to exit the search suggestion page, allowing them to close it at any time as needed, reducing distraction and improving interaction smoothness.

[0057] The search session page can be displayed in either half-screen or full-screen format. Search results on the search session page can include: recommended items determined by the search agent from multiple candidate items in the item library based on at least one of the search intent and search suggestions confirmed by the object, invoked by the large model. Furthermore, the large model can also determine recommended items from multiple candidate items in the item library based on at least one of the search intent, search suggestions confirmed by the object, and item card information. If the recommended item matches the object's search intent, and there are multiple recommended items, the search session page can display thumbnails of all recommended items, as well as service guarantee information, price, description, and purchase controls for the first recommended item. For example, the terminal can respond to a trigger action on the purchase controls by displaying the parameter selection page for the first recommended item, which is the item with the highest recommendation priority among the multiple items. Alternatively, the terminal can respond to a trigger action on the thumbnail of any recommended item by displaying the item details page for that recommended item. In addition, the search session page can provide objects with the ability to communicate with the search agent. Objects can input their search requirements on the search session page, and the search session page can display the search results generated by the search agent invoking the large model based on the search requirements input by the object.

[0058] In some embodiments, the method may further include: in response to a rollback operation on a search session page, displaying an aggregated display page of the platform, the aggregated display page being used to display multiple items provided by the platform, which helps to switch the recommendation scenario when the recommended items on the search session page are not attractive to the object, and provides the object with more items that match potential preferences on the aggregated display page. Alternatively, the method may also display a playback page in response to a rollback operation on a search session page.

[0059] In some embodiments, the search suggestion information may include at least one of the following: search suggestion text displayed on the item card; search keywords obtained by the search agent calling a large model to process the historical interaction behavior of the object. The search suggestion page may display a smart search control, the search suggestion text from the item card may be filled back into the search input box on the search suggestion page, and the confirmation operation of the search suggestion text may refer to the triggering operation of the smart search control.

[0060] In some embodiments, the search prompt information may further include search query information obtained by the search agent calling a large model to process the historical interaction behavior of the object. The search query information is used to prompt the object to ask the search agent, such as "Do you have any new wool sweaters to recommend?" The terminal can respond to the trigger operation of the search query information and display a search session page. The search results displayed on the search session page are generated by the search agent calling the large model based on the search query information.

[0061] In some embodiments, in addition to providing search suggestion information for object confirmation on the search suggestion page, the terminal may also provide editing functions on the search suggestion page, such as displaying a soft keyboard on the search suggestion page, allowing the object to edit the entered search suggestion information or re-enter the search information. The method may also include any of the following: in response to an editing operation on the search suggestion information on the search suggestion page, obtaining the edited search suggestion information; in response to a confirmation operation on the edited search suggestion information, displaying the search session page of the search agent; in response to an input operation on the search suggestion page, obtaining the input search information; in response to a confirmation operation on the search information, displaying the search session page of the search agent. In this embodiment, in response to a confirmation operation of the edited search suggestion information, the terminal displays a search session page for the search agent. The search results displayed on this page may include: recommended items determined by the search agent from multiple candidate items in the item library based on at least one of the search intent and the edited search suggestion information, using a large model. Similarly, in response to a confirmation operation of the search information, the terminal displays a search session page for the search agent. The search results displayed on this page may include: recommended items determined by the search agent from multiple candidate items in the item library based on at least one of the search intent and the input search information, using a large model. This embodiment provides an editing function to meet the personalized search needs of objects, improving the flexibility and coverage of the search, and making the search results determined based on the search information input by the object more consistent with the object's intent.

[0062] To more intuitively display recommended categories that match the search intent of an object, in some embodiments, the search suggestion page also includes representative images of multiple recommended categories. Each representative image is linked to the recommendation page of the corresponding recommended category. The recommended categories are obtained by the search agent calling a large model to process the object's historical interaction behavior. The method may further include: in response to a trigger operation on a representative image of any recommended category, displaying recommended items belonging to that category on the recommendation page of that category. The search suggestion page may also display category description information such as "beauty and sunscreen." This recommended category can be determined by the large model processing the object's historical interaction behavior to obtain the object's search intent, and then determined from multiple candidate categories in the item database based on the search intent. Alternatively, the search intent can be obtained by the large model processing the object's historical interaction behavior and the category corresponding to the item card. In this embodiment, images, as visual elements, can stimulate the object's interest and purchasing desire. The object only needs to click on the representative image of the recommended category to directly view the recommended items, eliminating the need for inputting search information, thereby reducing input costs and improving the browsing experience and conversion rate.

[0063] In some embodiments, if an object's behavior of lingering on an item card meets preset conditions, displaying the search agent's search page based on the object's search intent may include: displaying a search suggestion page if the object's behavior of lingering on the item card meets preset conditions and the object meets search suggestion conditions. Meeting search suggestion conditions means that: the object has not closed the search suggestion page; or, the time interval between the current time and the last time the search suggestion page was closed reaches a target time interval. This target time interval can be determined based on the object's activity level, which is determined based on the object's behavior on the platform. The object's activity level can be determined based on factors such as login frequency, browsing duration, page visit depth, number of clicks and interactions, and number of content publications on the platform. Higher activity levels indicate more frequent interactions between the object and the platform. For example, for highly active objects that have closed the search suggestion page, the target time interval can be appropriately extended to avoid interference during multiple browsing sessions on the platform; while for less active objects, the target time interval can be appropriately shortened to display the search suggestion page at a reasonable frequency, improving the object's experience and the effective reach of search suggestion information.

[0064] Unlike the above embodiments where the search page includes a search suggestion page initially displayed and a search session page subsequently displayed, in some embodiments, the search page displayed in step 202 may refer to a search results page that provides search results. The step 202, displaying the search agent's search page based on the object's search intent, may include: displaying the search agent's search results page based on the object's search intent, where the search results page includes search results obtained based on the search intent. These search results may include: recommended items determined by the search agent from multiple candidate items in the item library based on the search intent using a large model, or recommended items determined by the search agent from multiple candidate items in the item library based on the search intent and item card information using a large model. The search results page may be in the form of a session page, where the object can engage in a conversation with the search agent. In this embodiment, when the object's dwell behavior meets preset conditions, the search results are directly displayed on the search results page without requiring the object to confirm the search information. This reduces the object's decision-making burden and clicking steps, improves operational efficiency and interaction fluency, and provides the object with a lightweight and fast user experience.

[0065] The following is an illustrative example of the specific implementation of the item search method.

[0066] Figure 3 This is a schematic diagram of a playback page according to an exemplary embodiment. The playback page 300 displays item cards 301, which are located above the bottom navigation bar. Item cards 301 can display item videos, images, descriptions, prices, and service guarantees. They can also display a smart search entry 302, which can display search suggestions such as "Try a new wool sweater this fall / winter?". These search suggestions are obtained through the following steps: a search agent calls a large model to process the category corresponding to the item card and the object's historical interaction behavior to obtain the object's search intent; the large model then generates the search suggestion based on the search intent. Figure 3 The countdown timer is not displayed. For example... Figure 4 As shown, the countdown timer is displayed in digital format in the intelligent search entry 302. Figure 5 As shown, the countdown timer is displayed as a progress bar in the intelligent search entry 302.

[0067] The terminal can respond to a click operation by an object targeting the intelligent search entry 302, displaying, for example... Figure 6 The search suggestion page 600 shown, or if the terminal detects that the object's dwell time on the item card has reached the countdown timer, will display something like this. Figure 6The search suggestion page 600 shown includes a search input box 601 that displays search suggestion text from the intelligent search entry. The search input box 601 allows users to edit the search suggestion text or re-enter search information. The search suggestion page 600 can also display an intelligent search control 602. The terminal can respond to the user's triggering operation on the intelligent search control 602, that is, to the user's confirmation operation on the information in the search input box 601, and display, as shown... Figure 7 The search session page 700 shown can contain search results that can be recommended items determined by the search agent from multiple candidate items based on information in the search input box 601, item card information, and search intent, using a large model.

[0068] The search suggestion page 600 can also display search keywords such as "sweater" and "coat" obtained by the search agent calling a large model to process the historical interaction behavior of the object. The terminal can respond to the object's triggered operation for any search keyword and display such as Figure 7 The search session page 700 shown can contain search results that can be recommended items determined from multiple candidate items by the search agent calling a large model based on the search keywords selected by the object, information of the item card, and the search intent. Figure 6 Below the search input box 601, representative images and category descriptions for multiple recommended product categories can be displayed. The terminal can respond to a trigger operation on a representative image of any recommended product category and display recommended items belonging to that category on the recommended product category's recommendation page. Figure 6 The screen also displays a close control 603, which the terminal can respond to by triggering the close control 603 and returning to the playback page 300.

[0069] Figure 7 The search session page 700 can display search results, which may include multiple recommended items, such as... Figure 7 The search session page 700 can display the image of the first recommended item, service guarantee information, price, description, and purchase control. It can also display thumbnails of multiple candidate items. The search session page 700 can also display an exit control 701. In response to the triggering operation of the exit control 701, the terminal displays the aggregated display page or returns to the playback page 300.

[0070] Figure 8 This is a block diagram illustrating an item search device according to an exemplary embodiment. See also: Figure 8 The device includes: The playback page display module 801 is configured to display item cards on the playback page, the item cards being used to provide item information; The search page display module 802 is configured to display the search page of the search agent based on the search intent of the object if the object's dwell behavior on the item card meets preset conditions. The search page is used by the search agent to provide search information that matches the search intent, which is obtained by the search agent calling a large model to process the object's historical interaction behavior.

[0071] In some embodiments, the object's dwelling behavior on the item card satisfies a preset condition, which means any of the following situations: The object stays on the item card for a first preset time and does not perform any operation during the stay; The object's dwell time on the item card reaches the countdown time; The object stays on the item card and the content progress of the item card reaches the target progress.

[0072] In some embodiments, the progress of the item card reaching the target progress includes any of the following situations: The video playback progress of the item card has reached the target progress; The browsing progress of the text and image content in the item card has reached the target progress.

[0073] In some embodiments, the playback page display module is configured to perform at least one of the following: A smart search entry is displayed on the item card. In response to the object's trigger operation on the smart search entry, the search page of the search agent is displayed based on the object's search intent. The smart search entry is used to invoke the search agent. In response to the object's pause operation on the video in the item card, the search page of the search agent is displayed based on the object's search intent; In response to a sliding operation of the object in a preset direction, the search page of the search agent is displayed based on the search intent of the object; In response to the terminal's motion characteristics matching preset motion characteristics, the search agent's search page is displayed based on the object's search intent.

[0074] In some embodiments, the smart search entry is displayed as a prompt bar in an item card, a top floating button, a bottom navigation bar, or a content page corner mark.

[0075] In some embodiments, the playback page display module is configured as follows: Search suggestions are displayed on the item card. These suggestions are generated by the search agent by processing the historical interaction behavior of the object using a large model.

[0076] In some embodiments, the search page display module is configured as follows: Based on the search intent of the object, a search suggestion page of the search agent is displayed, the search suggestion page includes search suggestion information that matches the search intent; In response to a confirmation operation of the search suggestion information, a search session page of the search agent is displayed, the search session page including search results obtained based on the search intent.

[0077] In some embodiments, the search suggestion information includes at least one of the following: The search suggestion text displayed on the item card is obtained by the search agent calling a large model to process the historical interaction behavior of the object; The search keywords are obtained by the search agent calling the large model to process the historical interaction behavior of the object.

[0078] In some embodiments, the search page display module is configured to perform any of the following: In response to the editing operation of the search suggestion information on the search suggestion page, the edited search suggestion information is obtained; in response to the confirmation operation of the edited search suggestion information, the search session page of the search agent is displayed. In response to an input operation on the search suggestion page, the input search information is obtained; in response to a confirmation operation on the search information, the search session page of the search agent is displayed.

[0079] In some embodiments, the search suggestion page further includes representative images of multiple recommended categories, each representative image being linked to the recommendation page of the corresponding recommended category, and the recommended categories being obtained by the search agent calling a large model to process the historical interaction behavior of the object; The device further includes: The recommendation page display module is configured to display recommended items belonging to the recommended category on the recommendation page of the recommended category in response to a trigger operation for a representative image of any recommended category.

[0080] In some embodiments, the search suggestion page is displayed in a half-screen format, a side-sliding panel format, a bottom pop-up format, or a full-screen overlay format.

[0081] In some embodiments, the apparatus further includes: The aggregation display page module is configured to display the platform's aggregation display page in response to a back operation on the search session page. The aggregation display page is used to display multiple items provided by the platform.

[0082] In some embodiments, the search page display module is configured as follows: If the object's behavior of staying on the item card meets the preset conditions and the object meets the search suggestion conditions, the search suggestion page will be displayed. The search suggestion conditions being met refer to: The object did not close the search suggestion page; Alternatively, the time interval between the current time and the last time the search suggestions page was closed reaches the target time interval.

[0083] In some embodiments, the target time interval is determined based on the activity level of the object, which is determined based on the object's behavior on the platform.

[0084] In some embodiments, the search page display module is configured as follows: Based on the search intent of the object, a search results page of the search agent is displayed, the search results page including search results obtained based on the search intent.

[0085] In some embodiments, the object’s historical interaction behavior includes at least one of browsing behavior, search behavior, and item conversion behavior.

[0086] It should be noted that the item search device provided in the above embodiments is only illustrated by the division of the above functional units during item search. In practical applications, the above functions can be assigned to different functional units as needed, that is, the internal structure of the computer device can be divided into different functional units to complete all or part of the functions described above. In addition, the item search device and the item search method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0087] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0088] Computer equipment can be implemented as a terminal. Figure 9 This is a block diagram illustrating a terminal 900 according to an exemplary embodiment. Figure 9A structural block diagram of a terminal 900 provided in an exemplary embodiment of this disclosure is shown. The terminal 900 may be a smartphone, tablet computer, MP3 player (Moving Picture Experts Group Audio Layer III), MP4 player (Moving Picture Experts Group Audio Layer IV), laptop computer, or desktop computer. The terminal 900 may also be referred to as a user device, portable terminal, laptop terminal, desktop terminal, or other names.

[0089] Typically, terminal 900 includes a processor 901 and a memory 902.

[0090] Processor 901 may include one or more processing cores, such as a quad-core processor or an octa-core processor. Processor 901 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). Processor 901 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 901 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, processor 901 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0091] The memory 902 may include one or more computer-readable storage media, which may be non-transitory. The memory 902 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory 902 are used to store at least one computer program, which is executed by the processor 901 to implement the item search method provided in the method embodiments of this application.

[0092] In some embodiments, the terminal 900 may also optionally include a peripheral device interface 903 and at least one peripheral device. The processor 901, memory 902, and peripheral device interface 903 can be connected via a bus or signal line. Each peripheral device can be connected to the peripheral device interface 903 via a bus, signal line, or circuit board. Specifically, the peripheral device includes at least one of the following: a radio frequency circuit 904, a display screen 905, a camera assembly 906, an audio circuit 907, and a power supply 908.

[0093] Peripheral device interface 903 can be used to connect at least one I / O (Input / Output) related peripheral device to processor 901 and memory 902. In some embodiments, processor 901, memory 902 and peripheral device interface 903 are integrated on the same chip or circuit board; in some other embodiments, any one or two of processor 901, memory 902 and peripheral device interface 903 can be implemented on separate chips or circuit boards, which is not limited in this embodiment.

[0094] The radio frequency (RF) circuit 904 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The RF circuit 904 communicates with communication networks and other communication devices via electromagnetic signals. The RF circuit 904 converts electrical signals into electromagnetic signals for transmission, or converts received electromagnetic signals back into electrical signals. In some embodiments, the RF circuit 904 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a user identity module card, etc. The RF circuit 904 can communicate with other terminals through at least one wireless communication protocol. This wireless communication protocol includes, but is not limited to: the World Wide Web, metropolitan area networks, intranets, various generations of mobile communication networks (2G, 3G, 4G, and 5G), wireless local area networks, and / or WiFi (Wireless Fidelity) networks. In some embodiments, the RF circuit 904 may also include circuitry related to NFC (Near Field Communication), which is not limited in this application.

[0095] Display screen 905 is used to display a UI (User Interface). This UI may include graphics, text, icons, videos, and any combination thereof. When display screen 905 is a touch display screen, it also has the ability to collect touch signals on or above its surface. These touch signals can be input as control signals to processor 901 for processing. In this case, display screen 905 can also be used to provide virtual buttons and / or a virtual keyboard, also known as soft buttons and / or a soft keyboard. In some embodiments, there may be one display screen 905, disposed on the front panel of terminal 900; in other embodiments, there may be at least two display screens 905, disposed on different surfaces of terminal 900 or in a folded design; in other embodiments, display screen 905 may be a flexible display screen, disposed on a curved or folded surface of terminal 900. Furthermore, display screen 905 may be configured as a non-rectangular, irregular shape, i.e., a non-rectangular screen. Display screen 905 may be made of materials such as LCD (Liquid Crystal Display) or OLED (Organic Light-Emitting Diode).

[0096] The camera assembly 906 is used to acquire images or videos. In some embodiments, the camera assembly 906 includes a front-facing camera and a rear-facing camera. Typically, the front-facing camera is located on the front panel of the terminal, and the rear-facing camera is located on the back of the terminal. In some embodiments, there are at least two rear-facing cameras, which are any one of a main camera, a depth-sensing camera, a wide-angle camera, and a telephoto camera, to achieve background blurring by fusion of the main camera and the depth-sensing camera, panoramic shooting by fusion of the main camera and the wide-angle camera, VR (Virtual Reality) shooting, or other fusion shooting functions. In some embodiments, the camera assembly 906 may also include a flash. The flash can be a single-color temperature flash or a dual-color temperature flash. A dual-color temperature flash is a combination of a warm-light flash and a cool-light flash, which can be used for light compensation at different color temperatures.

[0097] The audio circuit 907 may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, converting them into electrical signals that are input to the processor 901 for processing, or to the radio frequency circuit 904 for voice communication. For stereo sound acquisition or noise reduction purposes, multiple microphones may be used, each positioned at a different location on the terminal 900. The microphone may also be an array microphone or an omnidirectional microphone. The speaker is used to convert electrical signals from the processor 901 or the radio frequency circuit 904 into sound waves. The speaker may be a conventional diaphragm speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, it can convert electrical signals not only into audible sound waves but also into inaudible sound waves for purposes such as distance measurement. In some embodiments, the audio circuit 907 may also include a headphone jack.

[0098] Power supply 908 is used to power the various components in terminal 900. Power supply 908 can be AC ​​power, DC power, a disposable battery, or a rechargeable battery. When power supply 908 includes a rechargeable battery, the rechargeable battery can be a wired rechargeable battery or a wireless rechargeable battery. A wired rechargeable battery is a battery that is charged via a wired line, while a wireless rechargeable battery is a battery that is charged via a wireless coil. The rechargeable battery can also be used to support fast charging technology.

[0099] Those skilled in the art will understand that Figure 9 The structure shown does not constitute a limitation on terminal 900, and may include more or fewer components than shown, or combine certain components, or use different component arrangements.

[0100] In an exemplary embodiment, a computer-readable storage medium is also provided, such as a memory including a computer program that can be executed by a processor to perform the methods in the above embodiments. For example, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, and optical data storage device, etc.

[0101] In an exemplary embodiment, a computer-readable storage medium including instructions is also provided, such as a memory including instructions, which can be executed by a processor of a computer device, such as a terminal, to perform the above-described item search method. Optionally, the computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0102] This disclosure also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the above-described item search method.

[0103] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0104] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for searching for items, characterized in that, The method includes: The playback page displays item cards, which provide item information. If the object's dwelling behavior on the item card meets preset conditions, based on the object's search intent, the search agent's search page is displayed. The search page is used by the search agent to provide search information that matches the search intent, which is obtained by the search agent calling a large model to process the object's historical interaction behavior.

2. The item search method according to claim 1, characterized in that, The object's dwelling behavior on the item card satisfies any of the following preset conditions: The object stays on the item card for a first preset time and does not perform any operation during the stay; The object's dwell time on the item card reaches the countdown time; The object stays on the item card and the content progress of the item card reaches the target progress.

3. The item search method according to claim 2, characterized in that, The progress of the item card's content reaching the target progress includes any of the following situations: The video playback progress of the item card has reached the target progress; The browsing progress of the text and image content in the item card has reached the target progress.

4. The item search method according to claim 1, characterized in that, The method further includes at least one of the following: A smart search entry is displayed on the item card. In response to the object's trigger operation on the smart search entry, the search page of the search agent is displayed based on the object's search intent. The smart search entry is used to invoke the search agent. In response to the object's pause operation on the video in the item card, the search page of the search agent is displayed based on the object's search intent; In response to a sliding operation of the object in a preset direction, the search page of the search agent is displayed based on the search intent of the object; In response to the terminal's motion characteristics matching preset motion characteristics, the search page of the search agent is displayed based on the search intent of the object.

5. The item search method according to claim 4, characterized in that, The intelligent search entry is displayed as a prompt bar in the item card, a top floating button, a bottom navigation bar, or a content page corner mark.

6. The item search method according to claim 1, characterized in that, The method further includes: Search suggestions are displayed on the item card. These suggestions are generated by the search agent by processing the historical interaction behavior of the object using a large model.

7. The item search method according to claim 1, characterized in that, The search page displayed by the search agent based on the search intent of the object includes: Based on the search intent of the object, a search suggestion page of the search agent is displayed, the search suggestion page includes search suggestion information that matches the search intent; In response to a confirmation operation of the search suggestion information, a search session page of the search agent is displayed, the search session page including search results obtained based on the search intent.

8. The item search method according to claim 7, characterized in that, The search suggestion information includes at least one of the following: The search suggestion text displayed on the item card is obtained by the search agent calling a large model to process the historical interaction behavior of the object; The search keywords are obtained by the search agent calling the large model to process the historical interaction behavior of the object.

9. The item search method according to claim 7, characterized in that, The method further includes any of the following: In response to the editing operation of the search suggestion information on the search suggestion page, the edited search suggestion information is obtained; in response to the confirmation operation of the edited search suggestion information, the search session page of the search agent is displayed. In response to an input operation on the search suggestion page, the input search information is obtained; in response to a confirmation operation on the search information, the search session page of the search agent is displayed.

10. The item search method according to claim 7, characterized in that, The search suggestion page also includes representative images of multiple recommended categories. Each representative image is linked to the recommendation page of the corresponding recommended category. The recommended categories are obtained by the search agent calling a large model to process the historical interaction behavior of the object. The method further includes: In response to a triggering operation for a representative image of any recommended category, recommended items belonging to that recommended category are displayed on the recommendation page of that recommended category.

11. The item search method according to claim 7, characterized in that, The search suggestion page can be displayed in a half-screen format, a side-sliding panel format, a bottom pop-up format, or a full-screen overlay format.

12. The item search method according to claim 7, characterized in that, The method further includes: In response to a rollback operation on the search session page, an aggregated display page of the platform is displayed, which is used to display multiple items offered by the platform.

13. The item search method according to claim 7, characterized in that, If the object's dwelling behavior on the item card meets preset conditions, the search page of the search agent is displayed based on the object's search intent, including: If the object's behavior of staying on the item card meets the preset conditions and the object meets the search suggestion conditions, the search suggestion page will be displayed. The search suggestion conditions being met refer to: The object did not close the search suggestion page; Alternatively, the time interval between the current time and the last time the search suggestions page was closed reaches the target time interval.

14. The item search method according to claim 13, characterized in that, The target time interval is determined based on the activity level of the object, which is based on the object's behavior on the platform.

15. The item search method according to claim 1, characterized in that, The search page displayed by the search agent based on the search intent of the object includes: Based on the search intent of the object, a search results page of the search agent is displayed, the search results page including search results obtained based on the search intent.

16. The item search method according to any one of claims 1 to 15, characterized in that, The object's historical interaction behavior includes at least one of browsing behavior, search behavior, and item conversion behavior.

17. An item search device, characterized in that, include: The playback page display module is configured to display item cards on the playback page, the item cards being used to provide item information; The search page display module is configured to display the search page of the search agent based on the search intent of the object if the object's dwell behavior on the item card meets preset conditions. The search page is used by the search agent to provide search information that matches the search intent, which is obtained by the search agent calling a large model to process the object's historical interaction behavior.

18. A computer device, characterized in that, The computer device includes: One or more processors; Memory used to store the executable program code of the processor; The processor is configured to execute the program code to implement the item search method as described in any one of claims 1 to 16.

19. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by the processor of a computer device, the computer device is enabled to perform the item search method as described in any one of claims 1 to 16.

20. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the item search method according to any one of claims 1 to 16.

Images