Information recommendation method and device, storage medium and computer equipment

Abstract

The invention discloses an information recommendation method and device, a storage medium and computer equipment, and relates to the technical field of Internet. The method comprises the steps that in response to an information recommendation operation, an information recommendation request is sent, and the information recommendation request is used for obtaining at least one piece of first topic information; in response to the received first topic information, the first topic information and / or an information input box are / is displayed, and the information input box is used for receiving second topic information input by the user; in response to a trigger operation of the first topic information or an input operation of the second topic information, determining target topic information, and sending the target topic information; at least one piece of recommendation information is received and displayed, the recommendation information is obtained based on matching of the target topic information, and the recommendation information comprises recommended merchants and / or recommended commodities. According to the method, the extensibility and freshness of the recommendation information can be improved, the requirement of the user for checking new commodities is met, and the method has good interactivity.

Description

Technical Field

[0001] The present invention relates to the field of Internet technologies, and in particular, to an information recommendation method, apparatus, storage medium, and computer device. Background Art

[0002] With the continuous development of Internet technologies, users' demands for online shopping are becoming more and more diverse. Currently, in order to increase the click-through rate and conversion rate of products, e-commerce platforms will recommend various products or merchants to users when they browse products, so that users can view a variety of products or merchants at any time, and thus can select suitable products for subsequent order placement operations according to their own needs.

[0003] In the prior art, the information recommendation solutions of e-commerce platforms mainly recommend products or merchants based on information such as users' past historical order information and historical behavior information. Compared with the product information that users have purchased in the past, the repetition rate and similarity rate of the recommended information generated by this recommendation method are relatively high. The demand for users to try new products cannot be well met, so that the recommended products cannot meet the needs of some users who want to try new products, resulting in limited breadth of the information browsed by users and low experience. Summary of the Invention

[0004] In view of this, the present application provides an information recommendation method, apparatus, storage medium, and computer device, mainly aiming to solve the technical problems of limited breadth of recommended information and low user experience.

[0005] According to a first aspect of the present invention, an information recommendation method is provided, and the method includes:

[0006] The client responds to an information recommendation operation and sends an information recommendation request to the server;

[0007] The server responds to receiving the information recommendation request, obtains at least one first topic information, and sends the first topic information to the client;

[0008] The client responds to receiving the first topic information and displays the first topic information and / or an information input box, where the information input box is used to receive second topic information input by the user;

[0009] The client responds to a trigger operation on the first topic information or an input operation on the second topic information, determines target topic information, and sends the target topic information to the server;

[0010] The server responds to receiving the target topic information, matches at least one piece of recommended information based on the target topic information, and sends the recommended information to the client;

[0011] The client receives and displays the recommended information, where the recommended information includes recommended merchants and / or recommended products.

[0012] Optionally, the method further includes: the client displays an information recommendation entry on a preset page through a first preset component and / or a second preset component, where the information recommendation entry is used to send the information recommendation request to the server in response to an information recommendation operation, the first preset component is displayed at a preset position on the preset page, and / or the second preset component is displayed in the information flow of the preset page; and / or the client displays at least one first topic information on the preset page through a third preset component, where the first topic information is used to determine the first topic information as the target topic information in response to a trigger operation and send the target topic information to the server, and the third preset component is displayed at a preset position on the preset page and / or in the information flow of the preset page.

[0013] Optionally, the client displays at least one first topic information on the preset page through a third preset component, including: the client displays a plurality of the first topic information on the preset page through the third preset component, where the plurality of the first topic information scrolls and displays in a preset direction, and / or the plurality of the first topic information scrolls and displays in the direction corresponding to the sliding operation in response to a sliding operation.

[0014] Optionally, the server obtains at least one first topic information, including: the server randomly selects at least one first topic information from a pre-established first topic database, where a plurality of preset topic information is pre-stored in the first topic database; and / or the server selects at least one first topic information from the first topic database according to the trigger frequency of each topic information in the first topic database; and / or the server selects at least one first topic information from the first topic database according to the relevance between user information and / or scenario information and each topic information in the first topic database; and / or the server generates at least one first topic information according to user information and / or scenario information.

[0015] Optionally, the client displays the first topic information and / or an information input box, including: the client uses a fourth preset component to display the first topic information and / or an icon corresponding to the first topic information; and / or the client displays the information input box and / or a prompt message corresponding to the information input box.

[0016] Optionally, in response to a trigger operation on the first topic information or an input operation on the second topic information, the client determines target topic information and sends the target topic information to the server, including: in response to the trigger operation on the first topic information, the client determines the triggered first topic information as the target topic information; or in response to the input operation on the second topic information, the client determines the input second topic information as the target topic information; the client sends the target topic information to the server and displays the target topic information, and / or dynamically displays a loading component corresponding to the target topic information.

[0017] Optionally, the server matches at least one piece of recommended information based on the target topic information, including: the server obtains a semantic recognition result according to the target topic information, recalls multiple pieces of recommended information according to the semantic recognition result, and sorts the multiple pieces of recommended information.

[0018] Optionally, the server obtains a semantic recognition result according to the target topic information, recalls multiple pieces of recommended information according to the semantic recognition result, and sorts the multiple pieces of recommended information, including: the server obtains at least one keyword based on a pre-trained semantic recognition model according to the target topic information, where the keyword includes at least one of a category keyword, a specification keyword, a scenario keyword, and a product description keyword; the server filters out multiple pieces of recommended information from a pre-established recommended information library based on a pre-trained information recall model, at least according to the category keyword; the server sorts the multiple pieces of recommended information according to the relevance between the recommended information and at least one of the specification keyword, the scenario keyword, and the product description keyword, and determines at least one piece of recommended information and the arrangement order of the recommended information according to the sorting result.

[0019] Optionally, the client receives and displays the recommended information, including: the client receives at least one piece of the recommended information and displays at least one piece of the recommended information in a list or information flow manner; and / or the client displays a recommended information switching component, where the recommended information switching component is used to display recommended information corresponding to the recommended information type in response to a selection operation of the recommended information type, and the recommended information type includes a product type and a merchant type.

[0020] Optionally, the method further includes: the server, in response to receiving the information recommendation request, obtains at least one third topic information, and sends the third topic information to the client; the client, in response to receiving the third topic information, displays at least one of the third topic information within a preset range of the information input box, and / or displays preset topic prompt information, where the third topic information is used to, in response to a trigger operation, determine the third topic information as the target topic information and send the target topic information to the server, and the topic prompt information is used to, in response to a trigger operation, display at least one of the first topic information.

[0021] Optionally, the server obtaining at least one third topic information includes: the server randomly selects at least one third topic information from a pre-established second topic database, where the second topic database pre-stores a plurality of preset topic information and / or at least one second topic information input by a user; and / or the server selects at least one third topic information from the second topic database according to the trigger frequency of each topic information in the second topic database; and / or the server selects at least one third topic information from the second topic database according to the relevance between user information and / or scenario information and each topic information in the second topic database; and / or the server generates at least one third topic information according to user information and / or scenario information.

[0022] Optionally, the method further includes: the client, in response to a trigger operation on the third topic information, determines the triggered third topic information as the target topic information and sends the target topic information to the server; the client cancels the display of the triggered third topic information, and / or updates the display of the third topic information on the current page.

[0023] Optionally, the method further includes: the client, in response to a backtracking operation on the recommendation information, displays at least one of the first topic information, and / or displays at least one target topic information and the recommendation information corresponding to the target topic information.

[0024] According to a second aspect of the present invention, an information recommendation method is provided, and the method includes:

[0025] In response to an information recommendation operation, sending an information recommendation request, where the information recommendation request is used to obtain at least one first topic information;

[0026] In response to receiving the first topic information, displaying the first topic information and / or an information input box, where the information input box is used to receive a second topic information input by a user;

[0027] In response to the triggering operation of the first topic information or the input operation of the second topic information, determine the target topic information, and send the target topic information;

[0028] Receive and display at least one recommended information, where the recommended information is obtained by matching based on the target topic information, and the recommended information includes recommended merchants and / or recommended products.

[0029] Optionally, the method further includes: displaying an information recommendation entry on a preset page through a first preset component and / or a second preset component, where the information recommendation entry is used to send the information recommendation request to the server in response to an information recommendation operation request, the first preset component is displayed at a preset position on the preset page, and / or, the second preset component is displayed in the information flow of the preset page; and / or displaying at least one first topic information on the preset page through a third preset component, where the first topic information is used to determine the first topic information as the target topic information in response to a triggering operation, and send the target topic information to the server, and the third preset component is displayed at a preset position on the preset page and / or in the information flow of the preset page.

[0030] Optionally, the displaying at least one first topic information on the preset page through a third preset component includes: displaying a plurality of the first topic information on the preset page through the third preset component, where the plurality of the first topic information are scrolled and displayed in a preset direction, and / or, the plurality of the first topic information respond to a sliding operation and scroll and display in the direction corresponding to the sliding operation.

[0031] Optionally, the displaying the first topic information and / or the information input box includes: displaying the first topic information and / or an icon corresponding to the first topic information by using a fourth preset component; and / or displaying the information input box and / or a prompt message corresponding to the information input box.

[0032] Optionally, the in response to the triggering operation of the first topic information or the input operation of the second topic information, determine the target topic information, and send the target topic information includes: in response to the triggering operation of the first topic information, determine the triggered first topic information as the target topic information, or, the client determines the input second topic information as the target topic information in response to the input operation of the second topic information; send the target topic information, and display the target topic information, and / or, dynamically display a loading component corresponding to the target topic information.

[0033] Optionally, receiving and displaying at least one recommended information includes: receiving at least one piece of the recommended information and displaying at least one piece of the recommended information in the form of a list or an information stream; and / or displaying a recommended information switching component, where the recommended information switching component is configured to display the recommended information corresponding to the type of recommended information in response to a selection operation of the type of recommended information, and the type of recommended information includes a product type and a merchant type.

[0034] Optionally, the method further includes: in response to receiving third topic information, displaying at least one piece of the third topic information within a preset range of the information input box, and / or displaying preset topic prompt information, where the third topic information is configured to be determined as the target topic information in response to a triggering operation and send the target topic information, and the topic prompt information is configured to display at least one piece of the first topic information in response to a triggering operation.

[0035] Optionally, the method further includes: in response to a triggering operation on the third topic information, determining the triggered third topic information as the target topic information and sending the target topic information; eliminating the display of the triggered third topic information, and / or updating the display of the third topic information on the current page.

[0036] Optionally, the method further includes: in response to a backtracking operation on the recommended information, displaying at least one piece of the first topic information, and / or displaying at least one piece of the target topic information and the recommended information corresponding to the target topic information.

[0037] According to a third aspect of the present invention, there is provided an information recommendation method, the method including:

[0038] In response to receiving an information recommendation request, obtaining at least one piece of first topic information and sending the first topic information to the client;

[0039] In response to receiving the target topic information, matching at least one piece of recommended information based on the target topic information and sending the recommended information to the client, where the target topic information is determined based on a triggering operation on the first topic information or an input operation on second topic information, the second topic information is input through an information input box, and the recommended information includes recommended merchants and / or recommended products.

[0040] Optionally, the obtaining of at least one first topic information includes: randomly selecting at least one first topic information from a pre-established first topic database, where a plurality of preset topic information is pre-stored in the first topic database; and / or selecting at least one first topic information from the first topic database according to the triggering frequency of each topic information in the first topic database; and / or selecting at least one first topic information from the first topic database according to the relevance between user information and / or scenario information and each topic information in the first topic database; and / or generating at least one first topic information according to user information and / or scenario information.

[0041] Optionally, the matching of at least one recommended information based on the target topic information includes: obtaining a semantic recognition result according to the target topic information, recalling a plurality of recommended information according to the semantic recognition result, and sorting the plurality of recommended information.

[0042] Optionally, the obtaining a semantic recognition result according to the target topic information, recalling a plurality of recommended information according to the semantic recognition result, and sorting the plurality of recommended information includes: based on a pre-trained semantic recognition model, obtaining at least one keyword according to the target topic information, where the keyword includes at least one of a category keyword, a specification keyword, a scenario keyword, and a product description keyword; based on a pre-trained information recall model, screening out a plurality of recommended information from a pre-established recommended information database at least according to the category keyword; sorting the plurality of recommended information according to the relevance between the recommended information and at least one of the specification keyword, the scenario keyword, and the product description keyword, and determining at least one recommended information and the arrangement order of the recommended information according to the sorting result.

[0043] Optionally, the method further includes: in response to receiving the information recommendation request, obtaining at least one third topic information and sending the third topic information to the client.

[0044] Optionally, the obtaining of at least one third topic information includes: randomly selecting at least one third topic information from a pre-established second topic database, where a plurality of preset topic information and / or at least one second topic information input by the user is pre-stored in the second topic database; and / or selecting at least one third topic information from the second topic database according to the triggering frequency of each topic information in the second topic database; and / or selecting at least one third topic information from the second topic database according to the relevance between user information and / or scenario information and each topic information in the second topic database; and / or generating at least one third topic information according to user information and / or scenario information.

[0045] According to the fourth aspect of the present invention, an information recommendation device is provided, and the device includes:

[0046] A recommendation request response module, configured to send an information recommendation request in response to an information recommendation operation, where the information recommendation request is used to obtain at least one first topic information;

[0047] A topic information display module, configured to display the first topic information and / or an information input box in response to receiving the first topic information, where the information input box is used to receive second topic information input by a user;

[0048] A target topic determination module, configured to determine target topic information in response to a trigger operation on the first topic information or an input operation on the second topic information, and send the target topic information;

[0049] A recommended information display module, configured to receive and display at least one recommended information, where the recommended information is obtained by matching based on the target topic information, and the recommended information includes recommended merchants and / or recommended products.

[0050] Optionally, the device further includes a recommendation entry display module, where the recommendation entry display module is configured to display an information recommendation entry on a preset page through a first preset component and / or a second preset component, where the information recommendation entry is used to send the information recommendation request to the server in response to an information recommendation operation request, the first preset component is displayed at a preset position on the preset page, and / or, the second preset component is displayed in the information flow on the preset page; and / or display at least one first topic information on the preset page through a third preset component, where the first topic information is used to determine the first topic information as the target topic information in response to a trigger operation, and send the target topic information to the server, and the third preset component is displayed at a preset position on the preset page and / or in the information flow on the preset page.

[0051] Optionally, the recommendation entry display module is specifically configured to display a plurality of the first topic information on the preset page through a third preset component, where the plurality of first topic information are scrolled and displayed in a preset direction, and / or, the plurality of first topic information respond to a sliding operation and scroll and display in the direction corresponding to the sliding operation.

[0052] Optionally, the topic information display module is specifically configured to display the first topic information and / or an icon corresponding to the first topic information by using a fourth preset component; and / or display the information input box and / or a prompt message corresponding to the information input box.

[0053] Optionally, the target topic determination module is specifically configured to, in response to a triggering operation of the first topic information, determine the triggered first topic information as the target topic information; or the client determines the input second topic information as the target topic information in response to an input operation of the second topic information; send the target topic information, and display the target topic information, and / or dynamically display a loading component corresponding to the target topic information.

[0054] Optionally, the recommended information display module is specifically configured to receive at least one piece of the recommended information, and display at least one piece of the recommended information in a list or information flow manner; and / or display a recommended information switching component, where the recommended information switching component is configured to display recommended information corresponding to the recommended information type in response to a selection operation of the recommended information type, and the recommended information type includes a product type and a merchant type.

[0055] Optionally, the topic information display module is further configured to, in response to receiving third topic information, display at least one piece of the third topic information within a preset range of the information input box, and / or display preset topic prompt information, where the third topic information is configured to be determined as the target topic information in response to a triggering operation and send the target topic information, and the topic prompt information is configured to display at least one piece of the first topic information in response to a triggering operation.

[0056] Optionally, the target topic determination module is further configured to, in response to a triggering operation of the third topic information, determine the triggered third topic information as the target topic information and send the target topic information; cancel the display of the triggered third topic information, and / or update the display of the third topic information on the current page.

[0057] Optionally, the recommended information display module is further configured to, in response to a backtracking operation of the recommended information, display at least one piece of the first topic information, and / or display at least one piece of the target topic information and the recommended information corresponding to the target topic information.

[0058] According to a fifth aspect of the present invention, an information recommendation device is provided, and the device includes:

[0059] A recommendation request response module, configured to, in response to receiving an information recommendation request, obtain at least one piece of first topic information and send the first topic information to the client;

[0060] A recommended information determination module, configured to, in response to receiving the target topic information, match at least one piece of recommended information based on the target topic information, and send the recommended information to the client, where the target topic information is determined based on a trigger operation of the first topic information or an input operation of the second topic information, the second topic information is input through an information input box, and the recommended information includes recommended merchants and / or recommended products.

[0061] Optionally, the recommended request response module is specifically configured to randomly select at least one piece of first topic information from a pre-established first topic database, where a plurality of preset topic information is pre-stored in the first topic database; and / or select at least one piece of first topic information from the first topic database according to the trigger frequency of each piece of topic information in the first topic database; and / or select at least one piece of first topic information from the first topic database according to the relevance between user information and / or scenario information and each piece of topic information in the first topic database; and / or generate at least one piece of first topic information according to user information and / or scenario information.

[0062] Optionally, the recommended information determination module is specifically configured to obtain a semantic recognition result according to the target topic information, recall a plurality of pieces of recommended information according to the semantic recognition result, and sort the plurality of pieces of recommended information.

[0063] Optionally, the recommended information determination module is specifically configured to, based on a pre-trained semantic recognition model, obtain at least one keyword according to the target topic information, where the keyword includes at least one of a category keyword, a specification keyword, a scenario keyword, and a product description keyword; based on a pre-trained information recall model, filter out a plurality of pieces of recommended information from a pre-established recommended information database at least according to the category keyword; sort the plurality of pieces of recommended information according to the relevance between the recommended information and at least one of the specification keyword, the scenario keyword, and the product description keyword, and determine at least one piece of recommended information and the arrangement order of the recommended information according to the sorting result.

[0064] Optionally, the recommended request response module is further configured to, in response to receiving the information recommendation request, obtain at least one piece of third topic information, and send the third topic information to the client.

[0065] Optionally, the recommended request response module is further specifically configured to randomly select at least one third topic information from a pre-established second topic database, where a plurality of preset topic information and / or at least one second topic information input by a user are pre-stored in the second topic database; and / or select at least one third topic information from the second topic database according to the trigger frequency of each topic information in the second topic database; and / or select at least one third topic information from the second topic database according to the relevance between user information and / or scenario information and each topic information in the second topic database; and / or generate at least one third topic information according to user information and / or scenario information.

[0066] According to a sixth aspect of the present invention, there is provided a storage medium having a computer program stored thereon, and when the program is executed by a processor, the above information recommendation method is implemented.

[0067] According to a seventh aspect of the present invention, there is provided a computer device including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the program, the above information recommendation method is implemented.

[0068] An information recommendation method, device, storage medium, and computer device provided by the present invention. First, in response to an information recommendation operation, a client sends an information recommendation request to a server. The server can obtain at least one first topic information according to the request. Then, the client can receive and display the first topic information and / or display an information input box, where the first topic information can be triggered, and the information input box is used to receive the second topic information input by the user. Further, when the first topic information is triggered or the second topic information is input, the client can determine target topic information according to the triggered first topic information or the input second topic information, and send the target topic information to the server. In response to receiving the target topic information, the server can match at least one recommended information based on the target topic information and push it to the client for display. The above information recommendation method can determine target topic information based on the first topic information selected by the user or the second topic information input by the user, and match at least one recommended information based on the target topic information and push it to the user, so that the recommended information is not limited to the user's past historical order information and user behavior information, and at the same time, it is not limited to the search keywords input by the user. Thus, it can support the user to flexibly select a recommended topic or set a new topic according to the interest point, and randomly expand the information search scope through the topic, thereby improving the breadth and freshness of the recommended information, meeting the user's need to view new products. In addition, the above method has good interactivity and can effectively improve the user experience of viewing search information.

[0069] The above description is only an overview of the technical solution of the present application. In order to better understand the technical means of the present application, it can be implemented according to the content of the specification. Moreover, in order to make the above and other objects, features, and advantages of the present application more obvious and understandable, the specific embodiments of the present application are hereinafter given. Brief Description of the Drawings

[0070] The drawings described herein are used to provide a further understanding of the present invention, and constitute a part of the present application. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0071] Figure 1 A flowchart showing a method for information recommendation provided by an embodiment of the present invention is shown;

[0072] Figure 2 A schematic diagram of a scenario of a method for information recommendation provided by an embodiment of the present invention is shown;

[0073] Figure 3 A schematic diagram of a scenario of a method for information recommendation provided by an embodiment of the present invention is shown;

[0074] Figure 4 A schematic diagram of a scenario of a method for information recommendation provided by an embodiment of the present invention is shown;

[0075] Figure 5 A schematic diagram of a scenario of a method for information recommendation provided by an embodiment of the present invention is shown;

[0076] Figure 6 A schematic diagram of a scenario of a method for information recommendation provided by an embodiment of the present invention is shown;

[0077] Figure 7 A schematic diagram of the structure of an information recommendation device provided by an embodiment of the present invention is shown;

[0078] Figure 8 A schematic diagram of the structure of another information recommendation device provided by an embodiment of the present invention is shown. Detailed Embodiments

[0079] The present invention will be described in detail below with reference to the drawings and in conjunction with the embodiments. It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments may be combined with each other.

[0080] In one embodiment, as Figure 1 shown, a method for information recommendation is provided. Taking the application of this method to a client and a server as an example, the method includes the following steps:

[0081] 101. In response to an information recommendation operation, the client sends an information recommendation request to the server.

[0082] Specifically, a user may initiate an information recommendation operation in various ways, such as by viewing a certain application, refreshing the current page, clicking on an information recommendation entry in a certain page, etc. In response to this information recommendation operation, the client may send an information recommendation request to the server. In this embodiment, the specific manner in which the user initiates the information recommendation operation is not specifically limited herein.

[0083] 102. In response to receiving the information recommendation request, the server obtains at least one first topic information and sends the first topic information to the client.

[0084] Among them, the interaction method between the client and the server may be direct interaction or indirect interaction. For example, the client may directly send a request or information to the server and receive the response data returned by the server, or may indirectly send a request or information to the server through a third-party client and / or a third-party server and receive the response data returned by the server through the third-party client and / or the third-party server. It can be understood that the interaction method between the client and the server is not limited in this embodiment.

[0085] Specifically, in response to receiving the information recommendation request, the server may obtain at least one first topic information based on a preset topic information acquisition strategy and send the obtained first topic information to the client. In this embodiment, the acquisition strategy of the first topic information may include at least one of various selection methods such as random selection, selection according to topic click-through rate, selection according to the relevance between topic information and user information, etc. The selection strategy and the number of selections of the first topic information may be set according to the actual situation and are not specifically limited herein. In addition, in response to receiving the information recommendation request, the server may also generate at least one first topic information based on user information and / or scenario information, etc., and send the generated first topic information to the client.

[0086] In this embodiment, the topic information refers to any statement related to the search theme, and a word count limit may be set. Among them, as a kind of copywriting information, compared with traditional keyword hint information such as product names, product categories, and discount types, the topic information has better text freedom, more flexible and novel setting methods, and is easier to meet the requirement of pushing new product information, thereby improving the information recommendation effect.

[0087] 103. In response to receiving the first topic information, the client displays the first topic information and / or an information input box, where the information input box is used to receive the second topic information input by the user.

[0088] Specifically, in response to receiving the first topic information pushed by the server, the client can display the first topic information through a preset component. At the same time, the client can also display an information input box, where the information input box can be used to receive the second topic information input by the user. In this embodiment, by displaying the first topic information and / or the information input box, it is convenient for users to input their interested topics by selection and input, thereby improving the breadth of information recommendation. Compared with the traditional keyword search method, the information recommendation method provided in this embodiment can provide a wider range of information recommendations for users. Users can select or input any topic information based on their own needs and view the recommended information corresponding to the topic information. Its information recommendation form is novel and the range of recommended information is wide.

[0089] 104. In response to the trigger operation of the first topic information or the input operation of the second topic information, the client determines the target topic information and sends the target topic information to the server.

[0090] Specifically, the client can, in response to the trigger operation of the first topic information, send the triggered first topic information to the server as the target topic information, or, in response to the input operation of the second topic information, send the input second topic information to the server as the target topic information. In this embodiment, the trigger operation of the first topic information and the input operation of the second topic information usually cannot be completed simultaneously, and there will be a sequential execution order when the client responds to the two operations. For example, when the user "simultaneously" sends the first topic information and the second topic information, the client can, based on the tiny time difference between the two operations, respond to the two operations successively and send the two target topic information to the server respectively; or, in the case where the time difference between the two operations is very tiny, select the previous operation to respond and send the topic information corresponding to the previous operation to the server as the target topic information.

[0091] 105. In response to receiving the target topic information, the server matches at least one piece of recommended information based on the target topic information and sends the recommended information to the client.

[0092] Specifically, after receiving the target topic information, the server can match at least one piece of recommended information in the recommended information library based on the association relationship between the target topic information and the recommended information. For example, the server can perform word segmentation on the target topic information to obtain at least one segmented keyword, then calculate the relevance between each segmented keyword and the recommended information, and screen out at least one piece of recommended information according to the relevance. For another example, the server can also perform semantic recognition based on the target topic information to obtain a semantic recognition result corresponding to the recommended information, and then calculate the relevance between the semantic recognition result and each piece of recommended information in the recommended information library, and screen out at least one piece of recommended information according to the relevance. Further, after screening out multiple pieces of recommended information, the server can also sort each piece of recommended information based on the relevance between the segmented keyword or semantic recognition result and each piece of recommended information, and send the sorted recommended information to the client.

[0093] 106. The client receives and displays the recommended information, where the recommended information includes recommended merchants and / or recommended products.

[0094] Specifically, after receiving the recommended information sent by the server, the client can display each piece of recommended information in the order of arrangement of the recommended information. Among them, the recommended information can include various types such as recommended merchants and recommended products. For example, the client can display multiple recommended merchants, or display multiple recommended products, or cross-display recommended merchants and recommended products, etc. In this embodiment, the client can preset the information type, display method, and display quantity of the recommended information, or can also receive the operations of the user to select or set the information type, display method, and display quantity of the recommended information.

[0095] The information recommendation method provided in this embodiment is as follows. First, in response to an information recommendation operation, the client sends an information recommendation request to the server. The server can obtain at least one first topic information according to this request. Then, the client can receive and display the first topic information and / or display an information input box. Among them, the first topic information can be triggered, and the information input box can be used to receive the second topic information input by the user. Further, when the first topic information is triggered or the second topic information is input, the client can determine the target topic information according to the triggered first topic information or the input second topic information, and send the target topic information to the server. In response to receiving the target topic information, the server can match at least one recommended information based on the target topic information and push it to the client for display. The above information recommendation method can determine the target topic information based on the first topic information selected by the user or the second topic information input by the user, and match at least one recommended information based on the target topic information and push it to the user, so that the recommended information is not limited to the user's past historical order information and user behavior information, and at the same time, it is not limited to the search keywords input by the user. Thus, it can support the user to flexibly select the recommended topic or set a new topic according to the interest points, and can freely expand the scope of information search through the topic, so as to improve the breadth and freshness of the recommended information, meet the user's need to view new products. In addition, the above method has good interactivity and can effectively improve the user experience of viewing search information.

[0096] In one embodiment, the above information recommendation method may further include the following steps:

[0097] 201. The client displays an information recommendation entry on a preset page through a first preset component and / or a second preset component.

[0098] Among them, the information recommendation entry can be used to send an information recommendation request to the server in response to an information recommendation operation. The first preset component is displayed at a preset position on the preset page, and / or the second preset component is displayed in the information flow of the preset page.

[0099] Specifically, when a user enters a certain application to view information, they can view any page in the application. On at least one preset page, the client can display an information recommendation entry through a first preset component and / or a second preset component. When the user clicks on the information recommendation entry, an information recommendation request can be initiated, thereby entering the subsequent recommendation process. Among them, the preset page can be any page of the application software, such as the main page, the theme venue page, the information recommendation page, etc. The first preset component and the second preset component can be components in various forms such as label components and bubble components. In this embodiment, there are no specific restrictions on the category of the preset page and the design form of the first preset component and the second preset component. By separately setting the first preset component and the second preset component in the preset position and the information flow of the preset page in this embodiment to display the information recommendation entry, the interaction performance of the page and the convenience of the user to initiate an information recommendation request can be improved, thereby enhancing the user experience.

[0100] For example, as Figure 2 shown, a main page of an application is displayed. Referring to Figure 2 (1), at the preset position on this main page, an information recommendation entry can be displayed through the label component "Intelligent Ordering". When the user clicks on this label component, an information recommendation request can be initiated and enter the subsequent information recommendation process. Further, referring to Figure 2 (2), in the information flow of this main page, an information recommendation entry can be displayed through the bubble component "Don't know what to eat?". When the user clicks on this bubble component, an information recommendation request can also be initiated and enter the subsequent information recommendation process.

[0101] 202. The client displays at least one first topic information through a third preset component on the preset page.

[0102] Among them, the first topic information is used to, in response to a trigger operation, determine the first topic information as the target topic information and send the target topic information to the server. The third preset component is displayed at the preset position on the preset page and / or in the information flow of the preset page. In this embodiment, the third preset component can be associated and displayed with the second preset component or the first preset component.

[0103] Specifically, in at least one preset page, the client can also display at least one first topic information through a third preset component. When the user clicks on a certain first topic information, the triggered first topic information can be determined as the target topic information. Then, the target topic information can be sent to the server, and this topic can be used as the first search topic to enter the subsequent recommendation process. Among them, the preset page can be any page of the application software, such as the main page, the theme venue page, the information recommendation page, etc. The third preset component can be various forms of components such as a label component, a bubble component, etc. In this embodiment, there are no specific restrictions on the category of the preset page and the design form of the third preset component. By setting the third preset component at the preset position and in the information flow of the preset page to display the first topic information, this embodiment can improve the interaction performance of the page and the convenience for the user to initiate an information recommendation request, enabling the user to more easily understand the information search function, thereby improving the usability of the information search function.

[0104] For example, referring to Figure 2 (2), in the information flow of the main page of a certain application, multiple first topic information can be displayed through a bubble component, such as displaying topics like "Fat Loss and Muscle Gain Package Plan" and "What to Eat for Dinner with Friends". When the user clicks on a certain topic, the triggered topic information can be determined as the target topic information and sent to the server, so that the server can match and recommend information based on the triggered topic information, thereby entering the subsequent information recommendation process.

[0105] Through the above embodiments, by setting the first preset component, the second preset component, and the third preset component at the preset position and in the information flow of the preset page to display the information recommendation entry and the first topic information, the interaction performance of the page and the convenience for the user to initiate an information recommendation request can be improved, enabling the user to more easily understand the information search function, thereby improving the usability and user experience of the information search function.

[0106] In one embodiment, the above step 202 can be specifically implemented through the following steps: The client displays multiple first topic information through the third preset component on the preset page, where the multiple first topic information scrolls and displays along a preset direction, and / or, the multiple first topic information scrolls and displays along the direction corresponding to the swipe operation in response to the swipe operation.

[0107] Specifically, the third preset component can display multiple first topic information at the same time. Among them, the multiple topic information can be set to an active and / or passive scrolling display effect, so that more first topic information can be displayed on the current page. In this embodiment, the third preset component can be used in conjunction with the first preset component and / or the second preset component. Through the above method, the convenience of users viewing topic information can be improved, facilitating the users to use the information recommendation function. At the same time, through the scrolling display effect, it is also convenient for users to notice the information search function, so that users can view more new products that have not been tried through the information recommendation function during the information browsing process, thereby improving the user experience.

[0108] For example, referring to Figure 2 (2), in the information flow on the main page of a certain application, multiple first topic information can be displayed through a bubble component, such as displaying topics like "Fat Loss and Muscle Gain Package Plan" and "What to Eat for Friends' Dinner at Night". When there are too many topics to be fully displayed in the limited position on the current page, a scrolling display effect can be set so that multiple first topic information can be automatically scrolled, or, alternatively, it can be passively scrolled in response to the user's sliding operation, or, it can also be scrolled by combining the automatic and passive scrolling effects.

[0109] In one embodiment, the method of responding to the information recommendation operation in step 101 can be implemented through the following steps: The client generates an information recommendation request in response to the trigger operation of the information recommendation entry displayed on the preset page, and sends the information recommendation request to the server, and / or, the client generates an information recommendation request in response to the trigger operation of the first topic information displayed on the preset page, determines the target topic information based on the triggered first topic information, and then sends the information recommendation request and the target topic information to the server.

[0110] Specifically, the user can initiate an information recommendation operation through the information recommendation entry displayed in the first preset component and / or the second preset component on the preset page, or can also initiate an information recommendation operation through any first topic information displayed in the third preset component on the preset page. In this embodiment, when initiating an information recommendation operation through the first topic information in the third preset component, it can be understood as directly jumping from step 101 to step 105, thereby performing related functions such as matching recommended information based on the target topic information, or this process can also be understood as a simplified process from step 101 to step 104.

[0111] For example, referring to Figure 2 (1), the user can initiate an information recommendation operation through the information recommendation entry corresponding to the label component "Intelligent Ordering" displayed on the main page to enter the subsequent information recommendation process; referring toFigure 2 (2), the user can also initiate an information recommendation operation through the information recommendation entry corresponding to the bubble component "Don't know what to eat?" displayed on the main page to enter the subsequent information recommendation process; refer to Figure 2 (2), the user can also initiate an information recommendation operation through the first topic corresponding to the bubble component "Fat loss and muscle gain meal plan" displayed on the main page, and use this topic as the first search topic to enter the subsequent information recommendation process.

[0112] By responding to the trigger operation of the information recommendation entry and / or the first topic information displayed on the preset page, the above embodiments can initiate an information recommendation request, improve the interaction performance of the page and the convenience of the user to initiate an information recommendation request, make it easier for the user to understand the information search function, and thus improve the usability and user experience of the information search function.

[0113] In one embodiment, the method for obtaining the first topic information in step 102 can be implemented through the following steps: the server randomly selects at least one first topic information from the pre-established first topic database, where multiple preset topic information is pre-stored in the first topic database, and / or, according to the trigger frequency of each topic information in the first topic database, at least one first topic information is selected from the first topic database, and / or, the server selects at least one first topic information from the first topic database according to the relevance between the user information and / or scenario information and each topic information in the first topic database, and / or, the server generates at least one first topic information according to the user information and / or scenario information.

[0114] Specifically, after receiving the information recommendation request, the server can select at least one of multiple strategies such as random selection, according to the click-through rate of each topic information, and according to the relevance between the user information and / or scenario information and the topic information, to select a preset number of first topic information and send it to the client for display. Selecting the first topic information according to the relevance between the user information and / or scenario information and each topic information in the first topic database means selecting at least one first topic information from the first topic database according to the relevance between at least one of the following multiple information: the topics the user has ever input and / or selected, the information the user has browsed, the historical order information, the current time and season, and the business district where the user is located, and each topic information.

[0115] For example, when a user initiates an information recommendation request, the current scenario information is a weekday afternoon in August in a certain business district, and the user information is that the user has previously selected topics related to fat loss. At this time, the server can, based on the above user information and scenario information, select multiple topics such as "fat-loss-friendly milk tea" and "summer heat-relieving herbal tea" from the first topic database, and select the topic with a higher click-through rate as the first topic information to be pushed to the client, thereby improving the accuracy of the first topic information recommendation. In addition, the server can also randomly select several topics as the first topic information to be pushed to the client for display, thereby improving the breadth and interest of information recommendation.

[0116] In addition, after receiving the information recommendation request, the server can also generate a preset number of first topic information according to the user information and / or scenario information. Specifically, the server can generate multiple first topic information based on at least one of various information such as the topics that the user has previously input and / or selected, the information that the user has browsed, the historical order information, the current time and season, and the business district where the user is located, through a pre-trained topic generation model, and then send the generated first topic information to the client for display. Among them, the topic generation model can be trained based on a large natural language model.

[0117] The above embodiments select the first topic information through a combination strategy of multiple strategies, or generate the first topic information through the user information and / or scenario information, which can improve the accuracy of the first topic information recommendation or improve the breadth of information recommendation, thereby improving the effect of information recommendation.

[0118] In one embodiment, the method of displaying the first topic information and / or the information input box in step 103 can be implemented through the following steps: The client uses the fourth preset component to display the first topic information, and / or display the icon corresponding to the first topic information, and / or the client displays the information input box, and / or display the prompt information corresponding to the information input box.

[0119] Specifically, when the client displays the first topic information, it can also display the icon corresponding to the first topic information. For example, when the first topic information is "fat-loss-friendly milk tea", it can also display the milk tea icon corresponding to this topic. In addition, when the client displays the information input box, it can also display the prompt information corresponding to the information input box. For example, when displaying the information input box, it can also display "You can ask me questions related to food" in the information input box. Through the above method, the interaction performance of the page and the interest and usability of the information recommendation function can be improved, thereby improving the information recommendation effect.

[0120] For example, as Figure 3 shown, it shows an information recommendation page of an application. In this example, when the user passes through such asFigure 2 After initiating an information recommendation request from the information recommendation entry on the displayed main page, the client can display an information recommendation page as shown in Figure 3 shown. Referring to Figure 3 (1) and Figure 3 (2), the client can display multiple first topic information and the corresponding icons for each first topic information through a preset component, and can also display an information input box at a preset position on the information recommendation page, and display the prompt information corresponding to the information input box in the information input box, so as to improve the page interaction performance.

[0121] In one embodiment, step 104 can be implemented through the following steps: The client determines the triggered first topic information as the target topic information in response to the trigger operation of the first topic information, or the client determines the input second topic information as the target topic information in response to the input operation of the second topic information. Then, the client sends the target topic information to the server and displays the target topic information, and / or dynamically displays the loading component corresponding to the target topic information.

[0122] Specifically, the client can determine the triggered first topic information or the input second topic information as the target topic information and send the target topic information to the server, so that the target topic information can be used as the basic information for information recommendation. Further, during the process of the server performing semantic recognition and searching for recommended information, the client can display the target topic information and dynamically display the loading component corresponding to the target topic information, so as to improve the interest of the information recommendation function and thus improve the user experience.

[0123] For example, referring to Figure 3 , when the user clicks on any first topic information in the information recommendation page, such as when the user clicks on the topic "milk tea that can be drunk for weight loss", the client can determine this triggered topic as the target topic information; or, referring to Figure 4 , when the user inputs any custom second topic information in the information input box on the current page, such as when the user inputs the topic "refreshing herbal tea to drink in summer", the client can determine this input topic as the target topic information. Then, the client can send the determined target topic information to the server and display the target topic information "milk tea that can be drunk for weight loss" or display "refreshing herbal tea to drink in summer" in the information recommendation page, and then dynamically display the loading component corresponding to the target topic information "I'm thinking..." and the corresponding progress bar.

[0124] In one embodiment, step 105 can be implemented through the following steps: The server can obtain a semantic recognition result based on the target topic information, then recall multiple recommended information according to the semantic recognition result, and finally sort the multiple pieces of recommended information and send the sorted recommended information to the client. In this embodiment, the target topic information can be semantically recognized by a pre-trained semantic recognition model. Among them, the semantic recognition result can include various keywords corresponding to the recommended information. For example, the recognized keywords can include category keywords, specification keywords, product description keywords, product evaluation keywords, scenario keywords, and so on. Based on the keywords recognized by semantics, it is convenient to accurately search, filter, and sort the recommended information, thereby improving the accuracy of information recommendation.

[0125] Compared with the way of finding recommended information according to the keywords contained in the search statement in the traditional technology, the information recommendation method provided in the above embodiment can find recommended information closer to the target topic information through semantic recognition. The recommended information is not limited to the keywords in the search statement, and the recommended information can be filtered and sorted according to the semantic information implied in the target topic information. Therefore, the accuracy of information recommendation can be effectively improved.

[0126] In one embodiment, step 105 can be implemented through the following steps: The server first obtains at least one keyword based on the pre-trained semantic recognition model according to the target topic information. Among them, the keyword includes at least one of category keywords, specification keywords, scenario keywords, and product description keywords. Then, based on the pre-trained information recall model, at least according to the category keywords, multiple recommended information is filtered out from the pre-established recommended information library. Finally, according to the relevance between the recommended information and at least one of the specification keywords, scenario keywords, and product description keywords, the multiple recommended information is sorted, and at least one recommended information and the arrangement order of the recommended information are determined according to the sorting result.

[0127] In the above embodiments, the server can optimize and train a pre-trained large model (i.e., a large language model or a natural language large model) through a large number of samples in a preset domain, so as to obtain a semantic recognition model. The input of the semantic recognition model is target topic information, and the output of the semantic recognition model is at least one keyword. In this embodiment, the keyword can be at least one of a category keyword, a specification keyword, a scenario keyword, and a product description keyword. Through these keywords, operations such as searching, screening, and sorting of recommended information can be performed, so that multiple recommended information corresponding to the target topic information can be found through semantic recognition, including product information, merchant information, and so on. Specifically, the server can, based on the pre-trained information recall model, screen out multiple recommended information from the pre-established recommended information library at least according to the category keyword. Among them, the information recall model can also be obtained through training of the large model. Further, the multiple recommended information can be sorted according to the relevance between the recommended information and keywords such as the specification keyword, the scenario keyword, and the product description keyword, and at least one recommended information and the arrangement order of the recommended information can be determined according to the sorting result.

[0128] For example, assume that the target topic information is "milk tea that can be drunk for weight loss". By performing semantic recognition on the target topic information, keywords such as "low fat", "low sugar", "sugar free", "no sugar added", and "milk tea" can be obtained. Among them, the keyword "milk tea" is a category keyword. According to this keyword and the above other keywords, multiple milk tea products related to the target topic information can be screened out. Then, the relevance between the specification keyword and the product description keywords "low fat", "low sugar", "sugar free", "no sugar added" and each milk tea product can be further calculated, and the screened milk tea products can be sorted according to the relevance calculation result. Finally, the milk tea products ranked in the preset top few are selected as recommended information and sent to the client according to the relevance. In the prior art, generally, the recommended information is searched and matched according to the keywords included in the search statement. For the keywords not included in the search statement, the search effect is much worse, or the relevant recommended information is ranked relatively low, and it is difficult for users to notice, resulting in poor accuracy of information recommendation. In contrast, in this embodiment, by performing semantic recognition on the target topic information, at least one keyword associated with the recommended information is obtained, and information recommendation is performed according to the keyword, which can effectively improve the accuracy of information recommendation.

[0129] Through semantic recognition of the target topic information, keywords corresponding to the recommended information are obtained, and the recommended information is searched, filtered, and sorted according to the keywords, which can improve the relevance between the recommended information and the target topic information, thereby improving the accuracy of information recommendation. In addition, by performing semantic recognition on the arbitrarily selected or input target topic information to obtain the recommended information, the user's search content can have a high degree of freedom. As long as the user inputs topic information related to the preset field, the corresponding recommended information can be obtained. Compared with the traditional keyword search, the information recommendation method provided in this embodiment can improve the relevance between the recommended information and the user's needs, thereby improving the information recommendation effect.

[0130] In one embodiment, step 106 can be implemented through the following steps: The client receives at least one piece of recommended information and displays at least one piece of recommended information in the form of a list or information stream, and / or the client displays a recommended information switching component, where the recommended information switching component is used to display the recommended information corresponding to the type of recommended information in response to the selection operation of the type of recommended information. The types of recommended information include product types and merchant types.

[0131] Specifically, after receiving the recommended information sent by the server, the client can display the recommended information in various ways such as a list or information stream. Among them, the recommended information can be merchant information and / or product information, etc. In addition, the client can also display a recommended information switching component for switching the type of recommended information. This component can be set in various forms such as labels, dropdown boxes, icons, etc. For example, the recommended information switching component can be set to two labels, "Merchant" and "Product". When the user clicks on the label corresponding to "Merchant", the client can display multiple pieces of merchant information in a list; when the user clicks on the label corresponding to "Product", the client can display multiple pieces of product information in a list. Through the display of various types of recommended information in the above embodiment, the richness of the recommended information can be improved. At the same time, it can also meet the user's viewing and selection needs for various types of recommended information to enhance the information recommendation effect.

[0132] For example, as Figure 5 shown, it shows an information recommendation page of an application. In this example, when the user selects any first topic information or inputs any second topic information through the information recommendation page as Figure 3 or as Figure 4 shown, the client can display the corresponding recommended information in the information recommendation page as Figure 5 shown. Refer to Figure 5 (1) and Figure 5(2) The client can display multiple recommended merchants or multiple recommended products in the form of a list or information stream, or display a recommended information switching component, so that the page can switch the display of merchants and products according to the user's selection operation.

[0133] In one embodiment, the above information recommendation method may further include the following steps:

[0134] 107. The server, in response to receiving an information recommendation request, obtains at least one third topic information and sends the third topic information to the client.

[0135] 108. The client, in response to receiving the third topic information, displays at least one third topic information within a preset range of the information input box, and / or displays preset topic prompt information.

[0136] Among them, the third topic information is used to determine the third topic information as the target topic information in response to a trigger operation and send the target topic information to the server, and the topic prompt information is used to display at least one first topic information in response to a trigger operation.

[0137] Specifically, the server, in response to receiving an information recommendation request, may also obtain at least one third topic information and send the third topic information to the client. After the client receives the third topic information, it can display the third topic information within the preset range of the information input box. In this embodiment, the third topic information can be used as an effective supplement to the first topic information to provide users with more selectable topics. In addition, the third topic information can be displayed within the preset range of the information input box, so as to facilitate users to refer to or select when inputting the second topic information. In this way, the page interaction performance can be improved, and the information recommendation efficiency and the breadth of information recommendation can be improved. In addition, the client can also display topic prompt information. By interacting with the topic prompt information, for example, clicking on the topic prompt information, at least one first topic information can be displayed. The display method can be that the page automatically scrolls to the display position corresponding to the first topic information or updates the first topic information in the current page, etc. By setting the topic prompt information, the page interaction performance can be improved and the information recommendation effect can be improved.

[0138] In this embodiment, the client can set the display timing of the third topic information according to the actual situation. For example, after the user initiates an information recommendation request, the first topic information and the third topic information can be displayed on the same page to enrich the user's selection range of topic information; or at least one third topic information can be displayed after the user selects a first topic information to prompt the user that more topics can be selected or more topics can be custom-input. It can be understood that this embodiment does not specifically limit the display timing and display position of the third topic information.

[0139] For example, as Figure 6 shown, an information recommendation page of an application is displayed. In this example, when the user selects any first topic information or inputs any second topic information through the information recommendation page as Figure 3 or as Figure 4 shown, the client can display the recommended information on the information recommendation page as Figure 5 shown. At the same time, at least one third topic information and the preset topic prompt information can also be displayed on the information recommendation page as Figure 6 shown. For example, the client can display multiple third topic information such as "Which is the best brand of snail rice noodles?" and "Which stores have the most repeat customers nearby?", and at the same time, the preset topic prompt information such as "What everyone is asking" can also be displayed. Among them, when there are too many third topic information displayed on the page and all of them cannot be fully displayed in the limited position on the current page, a scrolling display effect can be set so that multiple third topic information can be automatically scrolled, or, alternatively, it can be passively scrolled in response to the user's sliding operation, or, further, the automatic and passive scrolling effects can be combined for scrolling display.

[0140] In one embodiment, the method for obtaining the third topic information in step 107 can be implemented through the following steps: The server randomly selects at least one third topic information from the pre-established second topic database, where the second topic database pre-stores multiple preset topic information and / or at least one second topic information input by the user, and / or, the server selects at least one third topic information from the second topic database according to the trigger frequency of each topic information in the second topic database, and / or, the server selects at least one third topic information from the second topic database according to the relevance between the user information and / or scenario information and each topic information in the second topic database, and / or, the server generates at least one third topic information according to the user information and / or scenario information.

[0141] Specifically, after receiving the information recommendation request, the server can also select a preset number of third topic information and send them to the client for display through at least one of a variety of strategies such as random selection, based on the click-through rate of each topic information, and based on the correlation between user information and / or scene information and the topic information. Among them, selecting the third topic information based on the correlation between user information and / or scene information and each topic information in the third topic database means selecting at least one third topic information in the second topic database based on the correlation between at least one of a variety of information such as topics that the user has previously input and / or selected, information that the user has browsed, historical order information, current time and season, and the business district where the user is located and each topic information. In this embodiment, the topics in the second topic database may overlap with or be completely different from the topics in the first topic database, wherein the topic information in the second topic database can be set according to actual conditions.

[0142] For example, when a user initiates an information recommendation request, the current scene information is an afternoon on a weekday in August in a certain business district, and the user information is that the user has entered topics related to fat loss. At this time, the server can select multiple topics such as "nearby delicious low-fat milk tea" and "afternoon tea brand with the most repeat customers" from the second topic database based on the above user information and scene information, and select topics with higher click-through rates as third topic information to push to the client, thereby improving the accuracy of the third topic information recommendation. In addition, the server can also randomly select several topics as third topic information and push them to the client for display, thereby improving the breadth of information recommendation.

[0143] In addition, after receiving the information recommendation request, the server can also generate a preset number of third topic information based on user information and / or scene information. Specifically, the server can generate multiple third topic information based on at least one of the topics that the user has previously input and / or selected, the information that the user has browsed, historical order information, the current time and season, and the business district where the user is located, through a pre-trained topic generation model, and then send the generated third topic information to the client for display. Among them, the topic generation model can be pre-trained based on a large natural language model.

[0144] The above-mentioned embodiments select the third topic information through a combination of multiple strategies, or generate the third topic information based on user information and / or scenario information, so that the third topic information can be used as an effective supplement to the first topic information, thereby providing users with more topics to choose from, thereby improving the breadth and interest of information recommendations, thereby improving the information recommendation effect.

[0145] In one embodiment, the information recommendation method may further include the following steps:

[0146] 109. The client determines the triggered third topic information as the target topic information in response to the trigger operation of the third topic information, and sends the target topic information to the server.

[0147] 110. The client eliminates the display of the triggered third topic information, and / or updates the display of the third topic information on the current page.

[0148] Specifically, when any third topic information is triggered, the client can determine the triggered third topic information as the target topic information and send it to the server, so that the server can perform semantic recognition and push recommended information based on the target topic information. Then, the client can display the recommended information, and eliminate the display of the triggered third topic information, and / or update the display of the third topic information on the current page. By eliminating the display of the triggered third topic information and / or updating the display of the third topic information on the current page, the untriggered topic information can be automatically filled in the current page for the user to view, thereby improving the display efficiency and trigger probability of the third topic information, and thus improving the information recommendation effect.

[0149] For example, referring to Figure 6 (1), when the user clicks on the third topic information "Which is the best brand of snail noodles", the client can determine this topic as the target topic information and send the target topic information to the server. Then, referring to Figure 6 (2), the client can eliminate the display of the third topic information "Which is the best brand of snail noodles" and automatically fill it with the subsequent topics, or the client can also update the display of the third topic information on the current page, that is, replace a new batch of third topic information for the user to view and select.

[0150] In one embodiment, the above information recommendation method may further include the following steps:

[0151] 111. The client displays at least one first topic information in response to the backtracking operation of the recommended information, and / or displays at least one target topic information and the recommended information corresponding to the target topic information.

[0152] Specifically, when the user performs operations such as flipping up and down or swiping left and right on the page displaying the recommended information, the client can respond to the backtracking operation of the recommended information and display the historical recommended information and / or at least one first topic information, so that the user can view all the historical recommended information. In this embodiment, a component for viewing the historical recommended information can also be displayed on the page. When the user clicks on this component, a backtracking operation of the recommended information can also be initiated. It can be understood that the initiation method of the backtracking operation can be set according to the actual situation, and this embodiment does not make specific limitations here. In addition, the information backtracking range can be limited to all the information displayed in the current information recommendation page, or can be limited to all the recommended information browsed by the user within a preset time period, etc., and this embodiment does not make specific limitations either.

[0153] For example, referring to Figure 3 、 Figure 5 and Figure 6 , when the user selects the topic "Which is the best brand of snail rice noodles", the recommended information corresponding to this topic can be viewed. Further, when the user performs an information backtracking operation such as flipping up on the current page, the recommended information corresponding to the historical search topic "What kind of milk tea can be drunk during weight loss" as shown in Figure 5 can be viewed. When the user further performs an information backtracking operation such as flipping up, multiple first topic information such as "What kind of milk tea can be drunk during weight loss" and "What to eat for a friends' dinner at night" as shown in Figure 3 can also be viewed, so as to realize the backtracking view of all the recommended information and topic information.

[0154] In the above embodiment, by responding to the backtracking operation of the recommended information, the client displays the historical recommended information and / or the first topic information, which can facilitate the user to view the past search records, thereby improving the page interaction performance and the convenience of viewing the recommended information, and further improving the user experience.

[0155] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. In addition, the labels corresponding to the respective steps in the above embodiments only serve as identification and do not limit the execution order of the steps. The execution order of the steps in each embodiment can be set according to the actual situation.

[0156] Further, as a specific implementation of the method shown in Figures 1 to 6 , this embodiment provides an information recommendation device as shown in Figure 7 . The device includes: a recommendation request response module 31, a topic information display module 32, a target topic determination module 33, and a recommended information display module 34, where:

[0157] A recommended request response module 31, which can be used to send an information recommendation request in response to an information recommendation operation, where the information recommendation request is used to obtain at least one first topic information;

[0158] A topic information display module 32, which can be used to display the first topic information and / or an information input box in response to receiving the first topic information, where the information input box is used to receive second topic information input by a user;

[0159] A target topic determination module 33, which can be used to determine target topic information in response to a trigger operation on the first topic information or an input operation on the second topic information, and send the target topic information;

[0160] A recommended information display module 34, which can be used to receive and display at least one recommended information, where the recommended information is obtained by matching based on the target topic information, and the recommended information includes recommended merchants and / or recommended products.

[0161] In a specific application scenario, the device further includes a recommended entry display module 35. The recommended entry display module 35 can be used to display an information recommendation entry on a preset page through a first preset component and / or a second preset component, where the information recommendation entry is used to send the information recommendation request to the server in response to an information recommendation operation. The first preset component is displayed at a preset position on the preset page, and / or the second preset component is displayed in the information flow of the preset page; and / or display at least one first topic information on the preset page through a third preset component, where the first topic information is used to determine the first topic information as the target topic information in response to a trigger operation and send the target topic information to the server. The third preset component is displayed at a preset position on the preset page and / or in the information flow of the preset page.

[0162] In a specific application scenario, the recommended entry display module 35 can be specifically used to display a plurality of the first topic information on the preset page through a third preset component, where the plurality of the first topic information are scrolled and displayed in a preset direction, and / or the plurality of the first topic information are scrolled and displayed in the direction corresponding to a sliding operation in response to the sliding operation.

[0163] In a specific application scenario, the topic information display module 32 can be specifically used to display the first topic information and / or an icon corresponding to the first topic information by using a fourth preset component; and / or display the information input box and / or a prompt message corresponding to the information input box.

[0164] In a specific application scenario, the target topic determination module 33 can be specifically used to determine the triggered first topic information as the target topic information in response to the triggering operation of the first topic information, or the client determines the input second topic information as the target topic information in response to the input operation of the second topic information; send the target topic information, and display the target topic information, and / or dynamically display the loading component corresponding to the target topic information.

[0165] In a specific application scenario, the recommended information display module 34 can be specifically used to receive at least one piece of the recommended information and display at least one piece of the recommended information in the form of a list or an information stream; and / or display a recommended information switching component, where the recommended information switching component is used to display the recommended information corresponding to the recommended information type in response to the selection operation of the recommended information type, and the recommended information type includes a product type and a merchant type.

[0166] In a specific application scenario, the topic information display module 32 can also be used to display at least one piece of the third topic information within a preset range of the information input box in response to receiving the third topic information, and / or display a preset topic prompt information, where the third topic information is used to determine the third topic information as the target topic information in response to the triggering operation and send the target topic information, and the topic prompt information is used to display at least one piece of the first topic information in response to the triggering operation.

[0167] In a specific application scenario, the target topic determination module 33 can also be used to determine the triggered third topic information as the target topic information in response to the triggering operation of the third topic information and send the target topic information; cancel the display of the triggered third topic information, and / or update the display of the third topic information on the current page.

[0168] In a specific application scenario, the recommended information display module 34 can also be used to display at least one piece of the first topic information in response to the backtracking operation of the recommended information, and / or display at least one piece of the target topic information and the recommended information corresponding to the target topic information.

[0169] It should be noted that for other corresponding descriptions of each functional unit involved in the information recommendation device provided in this embodiment, reference can be made to Figures 1 to 6 the corresponding description therein, which will not be elaborated here.

[0170] Further, as Figures 1 to 6 a specific implementation of the method shown, this embodiment provides an information recommendation device, as Figure 8As shown, the device includes: a recommendation request response module 41 and a recommendation information determination module 42, where:

[0171] The recommendation request response module 41 is configured to, in response to receiving an information recommendation request, obtain at least one first topic information and send the first topic information to the client;

[0172] The recommendation information determination module 42 is configured to, in response to receiving the target topic information, match at least one recommendation information based on the target topic information and send the recommendation information to the client, where the target topic information is determined based on a trigger operation of the first topic information or an input operation of a second topic information, the second topic information is input through an information input box, and the recommendation information includes recommended merchants and / or recommended products.

[0173] In a specific application scenario, the recommendation request response module 41 is specifically configured to randomly select at least one first topic information from a pre-established first topic database, where a plurality of preset topic information is pre-stored in the first topic database; and / or select at least one first topic information from the first topic database according to the trigger frequency of each topic information in the first topic database; and / or select at least one first topic information from the first topic database according to the relevance between user information and / or scenario information and each topic information in the first topic database; and / or generate at least one first topic information according to user information and / or scenario information.

[0174] In a specific application scenario, the recommendation information determination module 42 is specifically configured to obtain a semantic recognition result according to the target topic information, recall a plurality of recommendation information according to the semantic recognition result, and sort the plurality of recommendation information.

[0175] In a specific application scenario, the recommendation information determination module 42 is specifically configured to, based on a pre-trained semantic recognition model, obtain at least one keyword according to the target topic information, where the keyword includes at least one of a category keyword, a specification keyword, a scenario keyword, and a product description keyword; based on a pre-trained information recall model, screen out a plurality of recommendation information from a pre-established recommendation information database at least according to the category keyword; sort the plurality of recommendation information according to the relevance between the recommendation information and at least one of the specification keyword, the scenario keyword, and the product description keyword, and determine at least one recommendation information and the arrangement order of the recommendation information according to the sorting result.

[0176] In a specific application scenario, the recommendation request response module 41 may further be configured to, in response to receiving the information recommendation request, obtain at least one third topic information and send the third topic information to the client.

[0177] In a specific application scenario, the recommendation request response module 41 may specifically further be configured to randomly select at least one third topic information from a pre-established second topic database, where the second topic database pre-stores a plurality of preset topic information and / or at least one second topic information input by a user; and / or select at least one third topic information from the second topic database according to the trigger frequency of each topic information in the second topic database; and / or select at least one third topic information from the second topic database according to the relevance between user information and / or scenario information and each topic information in the second topic database; and / or generate at least one third topic information according to user information and / or scenario information.

[0178] It should be noted that for other corresponding descriptions of each functional unit involved in the information recommendation device provided in this embodiment, reference may be made to Figures 1 to 6 the corresponding description in, which will not be elaborated herein.

[0179] Based on the method as described above Figures 1 to 6 shown, correspondingly, this embodiment further provides a storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the information recommendation method as described above Figures 1 to 6 shown.

[0180] Based on such an understanding, the technical solution of the present application can be embodied in the form of a software product. The software product to be recognized can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various implementation scenarios of the present application.

[0181] Based on the method as described above Figures 1 to 6 shown, and Figure 7 and Figure 8 shown in the information recommendation device embodiment, in order to achieve the above object, this embodiment further provides a computer device for information recommendation, which may specifically be a personal computer, a server, a smart phone, a tablet computer, a smart watch, or other network devices, etc. The computer device includes a storage medium and a processor; the storage medium is used to store a computer program and an operating system; the processor is used to execute the computer program to implement the method as described above Figures 1 to 6 shown.

[0182] Optionally, the computer device may further include an internal memory, a communication interface, a network interface, a camera, a Radio Frequency (RF) circuit, sensors, an audio circuit, a WI-FI module, a display screen (Display), an input device such as a keyboard (Keyboard), etc. Optionally, the communication interface may further include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (such as a WI-FI interface), etc.

[0183] Those skilled in the art can understand that the structure of the computer device for identifying an operation action provided in this embodiment does not limit the computer device, and it may include more or fewer components, or combine certain components, or have different component arrangements.

[0184] The storage medium may further include an operating system and a network communication module. The operating system is a program for managing the hardware and software resources to be recognized of the above computer device, and supports the operation of the information processing program and other software and / or programs to be recognized. The network communication module is used to implement communication between components inside the storage medium, and communication between other hardware and software in the information processing computer device.

[0185] Through the description of the above embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus a necessary general hardware platform, or can also be implemented by hardware. By applying the technical solution of this application, first, in response to an information recommendation operation, the client sends an information recommendation request to the server, and the server can obtain at least one first topic information according to the request. Then, the client can receive and display the first topic information and / or display an information input box, where the first topic information can be triggered, and the information input box can be used to receive the second topic information input by the user. Further, when the first topic information is triggered or the second topic information is input, the client can determine the target topic information according to the triggered first topic information or the input second topic information, and send the target topic information to the server. In response to receiving the target topic information, the server can match at least one recommended information based on the target topic information and push it to the client for display. Compared with the prior art, the above information recommendation method supports users to flexibly select the recommended topics or set new topics according to their interests, and can arbitrarily expand the scope of information search through topics, so as to improve the breadth and freshness of the recommended information, meet the user's need to view new products. In addition, the above method has good interactivity and can effectively improve the user experience of viewing search information.

[0186] Those skilled in the art can understand that the accompanying drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the drawings are not necessarily essential for implementing the present application. Those skilled in the art can understand that the modules in the devices in the implementation scenario can be distributed in the devices of the implementation scenario according to the description of the implementation scenario, or can be correspondingly changed and located in one or more devices different from the present implementation scenario. The modules in the above implementation scenario can be combined into one module, or can be further split into multiple sub-modules.

[0187] The above serial numbers of the present application are only for description and do not represent the advantages or disadvantages of the implementation scenarios. The above disclosure is only several specific implementation scenarios of the present application. However, the present application is not limited thereto, and any changes that can be conceived by those skilled in the art should fall within the protection scope of the present application.

Claims

1. An information recommendation method, characterized in that, the method includes: The client sends an information recommendation request to the server in response to an information recommendation operation; The server obtains at least one first topic information in response to receiving the information recommendation request, and sends the first topic information to the client; The client displays the first topic information and / or an information input box in response to receiving the first topic information, wherein the information input box is used to receive a second topic information input by the user; The client determines target topic information in response to a trigger operation on the first topic information or an input operation on the second topic information, and sends the target topic information to the server; The server matches at least one recommended information based on the target topic information in response to receiving the target topic information, and sends the recommended information to the client; The client receives and displays the recommended information, wherein the recommended information includes recommended merchants and / or recommended products.

2. The method according to claim 1, characterized in that, the method further includes: The client displays an information recommendation entry on a preset page through a first preset component and / or a second preset component, wherein the information recommendation entry is used to send the information recommendation request to the server in response to an information recommendation operation, the first preset component is displayed at a preset position on the preset page, and / or, the second preset component is displayed in the information flow of the preset page; and / or The client displays at least one first topic information on a preset page through a third preset component, wherein the first topic information is used to determine the first topic information as the target topic information in response to a trigger operation, and send the target topic information to the server, the third preset component is displayed at a preset position on the preset page and / or in the information flow of the preset page.

3. The method according to claim 2, characterized in that, The client displays at least one first topic information on a preset page through a third preset component, including: The client displays a plurality of the first topic information on the preset page through a third preset component, wherein the plurality of first topic information are scrolled and displayed in a preset direction, and / or, the plurality of first topic information respond to a sliding operation and scroll and display in the direction corresponding to the sliding operation.

4. The method according to claim 1, characterized in that, The server obtains at least one first topic information, including: The server randomly selects at least one first topic information from a pre-established first topic database, wherein a plurality of preset topic information are pre-stored in the first topic database; and / or The server selects at least one first topic information from the first topic database according to the trigger frequency of each topic information in the first topic database; and / or The server selects at least one first topic information from the first topic database according to the relevance between the user information and / or scenario information and each topic information in the first topic database; and / or The server generates at least one first topic information according to the user information and / or scenario information.

5. An information recommendation method Characterized in that The method includes In response to an information recommendation operation, sending the information recommendation request, where the information recommendation request is used to obtain at least one first topic information; In response to receiving the first topic information, displaying the first topic information and / or an information input box, where the information input box is used to receive a second topic information input by the user; In response to a trigger operation on the first topic information or an input operation on the second topic information, determining the target topic information and sending the target topic information; Receiving and displaying at least one recommended information, where the recommended information is matched based on the target topic information, and the recommended information includes recommended merchants and / or recommended products.

6. An information recommendation method Characterized in that The method includes In response to receiving an information recommendation request, obtaining at least one first topic information and sending the first topic information to the client; In response to receiving the target topic information, matching at least one recommended information based on the target topic information and sending the recommended information to the client, where the target topic information is determined based on a trigger operation on the first topic information or an input operation on a second topic information, the second topic information is input through an information input box, and the recommended information includes recommended merchants and / or recommended products.

7. An information recommendation device Characterized in that The device includes A recommendation request response module, configured to send the information recommendation request in response to an information recommendation operation, where the information recommendation request is used to obtain at least one first topic information; A topic information display module, configured to display the first topic information and / or an information input box in response to receiving the first topic information, where the information input box is used to receive a second topic information input by the user; A target topic determination module, configured to determine the target topic information and send the target topic information in response to a trigger operation on the first topic information or an input operation on the second topic information; A recommended information display module, configured to receive and display at least one recommended information, where the recommended information is matched based on the target topic information, and the recommended information includes recommended merchants and / or recommended products.

8. An information recommendation device Characterized in that The device includes A recommendation request response module, configured to obtain at least one first topic information in response to receiving an information recommendation request and send the first topic information to the client; A recommended information determination module, configured to, in response to receiving the target topic information, match at least one piece of recommended information based on the target topic information, and send the recommended information to the client, where the target topic information is determined based on a trigger operation on the first topic information or an input operation on the second topic information, the second topic information is input through an information input box, and the recommended information includes recommended merchants and / or recommended products.

9. A storage medium, on which a computer program is stored, characterized in that, when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images