Content recommendation method, apparatus, device, storage medium, and program product

Abstract

The application discloses a content recommendation method and device, equipment, storage medium and program product, and relates to the technical field of computers. The method comprises the following steps: receiving an interactive operation on a content browsing interface in a target application program; in response to the interactive operation meeting rearrangement conditions, obtaining rearrangement recommendation content corresponding to the interactive operation; replacing the cached recommendation content to be displayed with the rearrangement recommendation content; and displaying the rearrangement recommendation content on the content browsing interface in connection with the displayed recommendation content. The matching of the rearrangement conditions is performed according to the real-time interactive operation of the user, the rearrangement of the recommendation content is performed according to the interactive operation when the rearrangement conditions are met, the rearrangement recommendation content is obtained, and the rearrangement recommendation content is replaced with the cached and unloaded content for subsequent display, thereby improving the real-time relevance between the recommendation content and the interactive operation, converting the real-time operation of the user into the condition for recalling the recommendation content, and improving the accuracy and effectiveness of the content recommendation.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a content recommendation method, apparatus, device, storage medium, and program product. Background Technology

[0002] Content recommendation is commonly used in various application scenarios such as video content recommendation, news content recommendation, and product content recommendation. For example, after obtaining user authorization, the system obtains the user's static attribute data and historical operation data, and then matches the content that matches the user's interests from the content pool based on the static attribute data and historical operation data, and displays the content to the user.

[0003] In related technologies, the server sends recommended content to the terminal in batches. For example, the server sends 50 recommended items to the terminal for display each time. The terminal divides the 50 recommended items into 5 batches, displays 10 recommended items each time, and after all 50 recommended items have been displayed, it retrieves another 50 recommended items from the server for display.

[0004] However, in the above content recommendation process, if a batch of content recommended to a user at one time fails to hit the user's interests in a timely manner, it will lead to low content recommendation efficiency and low accuracy, resulting in a poor user experience when browsing recommended content. Summary of the Invention

[0005] This application provides a content recommendation method, apparatus, device, storage medium, and program product, which can improve the accuracy of content recommendation. The technical solution is as follows.

[0006] On the one hand, a content recommendation method is provided, the method comprising:

[0007] Receive interactive actions on the content browsing interface of the target application, the interactive actions being used to indicate real-time interaction with the displayed recommended content, the target application also including cached recommended content to be displayed;

[0008] In response to the interaction operation meeting the reordering conditions, the reordered recommended content corresponding to the interaction operation is obtained. The reordering conditions are conditions configured for interaction tendencies, and the reordered recommended content is content rearranged based on the interaction tendencies corresponding to the interaction operation.

[0009] Replace the cached recommended content to be displayed with the rearranged recommended content;

[0010] The rearranged recommended content is then linked to the already displayed recommended content and displayed on the content browsing interface.

[0011] On the other hand, a content recommendation method is provided, the method comprising:

[0012] A reordering request is sent by a receiving terminal, the terminal having a target application installed thereon. The reordering request includes an interactive operation and cached recommended content not displayed in the target application. The interactive operation is a real-time operation received by the terminal on the content browsing interface of the target application.

[0013] Determine the interaction tendency corresponding to the interactive operation, wherein the interaction tendency includes positive interaction tendency and negative interaction tendency, the positive interaction tendency is used to indicate the interest tendency on the displayed recommended content, and the negative interaction tendency is used to indicate the tendency to ignore the displayed recommended content;

[0014] Based on the interaction preference, the cached recommended content and the recommended content in the content pool are rearranged to obtain rearranged recommended content;

[0015] The rearranged recommended content is fed back to the terminal, and the rearranged recommended content is used to replace the cached recommended content for display.

[0016] On the other hand, a content recommendation device is provided, the device comprising:

[0017] A receiving module is used to receive interactive operations on the content browsing interface of a target application, the interactive operations being used to indicate the real-time interaction with the displayed recommended content, the target application also including cached recommended content to be displayed;

[0018] The acquisition module is used to acquire rearranged recommended content corresponding to the interaction operation in response to the interaction operation meeting the rearrangement conditions. The rearrangement conditions are conditions configured for interaction tendencies, and the rearranged recommended content is content rearranged based on the interaction tendencies corresponding to the interaction operation.

[0019] The update module is used to replace the cached recommended content to be displayed with the rearranged recommended content;

[0020] The display module is used to connect the rearranged recommended content with the already displayed recommended content and display it on the content browsing interface.

[0021] On the other hand, a content recommendation device is provided, the device comprising:

[0022] A receiving module is used to receive a reordering request sent by a terminal, wherein a target application is installed in the terminal, and the reordering request includes an interactive operation and cached recommended content not displayed in the target application, wherein the interactive operation is a real-time operation received by the terminal on the content browsing interface of the target application;

[0023] The determination module is used to determine the interaction tendency corresponding to the interactive operation, wherein the interaction tendency includes positive interaction tendency and negative interaction tendency. The positive interaction tendency is used to indicate the interest tendency towards the displayed recommended content, and the negative interaction tendency is used to indicate the tendency to ignore the displayed recommended content.

[0024] The sorting module is used to rearrange the cached recommended content and the recommended content in the content pool based on the interaction tendency to obtain rearranged recommended content;

[0025] The sending module is used to send the rearranged recommended content back to the terminal, and the rearranged recommended content is used to replace the cached recommended content for display.

[0026] On the other hand, a computer device is provided, the computer device including a processor and a memory, the memory storing at least one instruction, at least one program, code set or instruction set, the at least one instruction, the at least one program, the code set or instruction set being loaded and executed by the processor to implement the content recommendation method as described in any of the embodiments of this application above.

[0027] On the other hand, a computer-readable storage medium is provided, wherein at least one instruction, at least one program, code set, or instruction set is stored therein, wherein the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by a processor to implement the content recommendation method as described in any of the embodiments of this application above.

[0028] On the other hand, a computer program product or computer program is provided, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform any of the content recommendation methods described in the above embodiments.

[0029] The beneficial effects of the technical solutions provided in this application include at least the following:

[0030] The system matches reordering conditions based on real-time user interactions. When these conditions are met, the recommended content is reordered according to the interactions, resulting in reordered recommended content. This reordered recommended content then replaces cached but unloaded content for subsequent display. This improves the real-time correlation between recommended content and interactions, transforming real-time user actions into conditions for recalling recommended content, thereby enhancing the accuracy and effectiveness of content recommendations. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 This is a schematic diagram illustrating content recommendations in the related technologies provided by an exemplary embodiment of this application;

[0033] Figure 2 This is a schematic diagram illustrating the overall process of a content recommendation method provided in an exemplary embodiment of this application;

[0034] Figure 3 This is a schematic diagram of an implementation environment provided by an exemplary embodiment of this application;

[0035] Figure 4 This is a schematic diagram of the cloud-client interaction process provided in an exemplary embodiment of this application;

[0036] Figure 5 This is a flowchart of a content recommendation method provided in an exemplary embodiment of this application;

[0037] Figure 6 Based on Figure 5 The illustrated embodiment provides a schematic diagram of the division of interaction tendencies;

[0038] Figure 7 Based on Figure 5 The illustrated embodiment provides a schematic diagram of the content recommendation principle;

[0039] Figure 8 Based on Figure 5 The illustrated embodiment provides a schematic diagram of the client module division;

[0040] Figure 9 This is a flowchart of a content recommendation method provided in another exemplary embodiment of this application;

[0041] Figure 10 Based on Figure 9 The illustrated embodiment provides a schematic diagram of the content rearrangement process;

[0042] Figure 11 This is a flowchart of a content recommendation method provided in another exemplary embodiment of this application;

[0043] Figure 12 This is a structural block diagram of a content recommendation apparatus provided in an exemplary embodiment of this application;

[0044] Figure 13 This is a structural block diagram of a content recommendation apparatus provided in another exemplary embodiment of this application;

[0045] Figure 14 This is a structural block diagram of a content recommendation apparatus provided in another exemplary embodiment of this application;

[0046] Figure 15 This is a structural block diagram of a computer device provided in an exemplary embodiment of this application. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0048] Content recommendation is commonly used in various application scenarios such as video content recommendation, news content recommendation, and product content recommendation. Typically, personalized content recommendations are delivered to users using paginated requests. That is, in order to reduce the system's query per second (QPS) pressure and conserve machine resources, after the client sends a recommendation request to the server (such as a cloud server), the server returns a fixed batch of content elements (items) as the recommendation results for each refresh.

[0049] In related technologies, the server obtains static attribute data (such as gender, age, interest tags, etc.) and historical operation data (such as historical browsing data, historical click data, etc.) of the user account, and generates a certain amount of recommended content to display to the user each time based on the static attribute data and historical operation data. It is understood that in the specific implementation of this application, the static attribute data, historical operation data and other related data of the user account are involved. When the above embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0050] Indicative, such as Figure 1As shown, in the relevant technology, the client 110 retrieves 50 recommended content items from the cloud 120 in each batch for display, and the 50 recommended content items are displayed in five batches. Each refresh yields 10 recommended content items. After refreshing 5 times, the client 110 sends a recommendation request to the cloud 120 again and retrieves another 50 recommended content items from the cloud 120 for display.

[0051] In this embodiment, a collaborative recommendation approach is adopted, whereby the client monitors the user's real-time behavior and feeds the user's real-time behavior back to the server, which then adjusts the recommended content based on the user's actual behavior.

[0052] Indicative, such as Figure 2 As shown, client 210 retrieves 50 recommended content items from cloud 220 in each batch for display, and these 50 recommended content items are displayed sequentially in five batches, with 10 recommended content items obtained each time. After displaying 10 items, if it is determined that content reordering is necessary based on user interaction, a reordering request is sent to cloud 220. This reordering request includes cached recommended content that has been retrieved but not yet displayed in the current client, as well as the interaction operation. Cloud 220 rearranges the cached recommended content and the content in the content pool according to the interaction tendency, resulting in 40 or 50 recommended content items. Figure 2 (Taking 50 items as an example) The new rearranged recommended content is fed back to client 210 for display. Client 210 displays the rearranged recommended content in batches according to the order of the rearranged recommended content.

[0053] Secondly, the implementation environment involved in the embodiments of this application will be described, for illustrative purposes only. Please refer to [the relevant documentation]. Figure 3 The implementation environment involves a terminal 310 and a server 320, which are connected via a communication network 330.

[0054] In some embodiments, the terminal 310 is equipped with a target application with content browsing functionality. This target application includes video playback programs, music playback programs, news browsing programs, shopping programs, short video programs, etc., and this application embodiment does not limit this to any particular application. Based on the user's interactive operations on the content browsing interface, the terminal 310 sends a rearrangement request to the server 320. The server 320 then rearranges and generates recommended content based on the rearrangement request, replacing the currently retrieved but not displayed recommended content in the terminal 310 for display.

[0055] After receiving the reordering request reported by the terminal 310, the server 320 can determine the interaction tendency corresponding to the interaction operation based on the operation identifier of the interaction operation received by the terminal 310, as well as the identifier of the cached recommended content that the terminal 310 has stored but not displayed. Based on the interaction tendency, the server 320 re-scored and reordered the cached recommended content that has been stored but not displayed and the content in the content pool, and fed back to the terminal 310 for display.

[0056] The aforementioned terminal can be various forms of terminal devices such as mobile phones, tablets, desktop computers, portable laptops, and smart TVs, and this application embodiment does not limit this to any particular type.

[0057] It is worth noting that the aforementioned servers can be independent physical servers, server clusters or distributed systems composed of multiple physical servers, or cloud servers that provide basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0058] Cloud technology refers to a hosting technology that unifies hardware, software, and network resources within a wide area network (WAN) or local area network (LAN) to achieve data computation, storage, processing, and sharing. Based on the cloud computing business model, cloud technology encompasses network technology, information technology, integration technology, management platform technology, and application technology. It can form resource pools, providing flexible and convenient on-demand access. Cloud computing technology will become a crucial support. Backend services of technical network systems require substantial computing and storage resources, such as video websites, image websites, and many portal websites. With the rapid development and application of the internet industry, every item may have its own identification mark in the future, requiring transmission to backend systems for logical processing. Data at different levels will be processed separately, and various industry data will require robust system support, which can only be achieved through cloud computing.

[0059] In some embodiments, the server described above can also be implemented as a node in a blockchain system. Blockchain is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and cryptographic algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked together using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include a blockchain underlying platform, a platform product service layer, and an application service layer.

[0060] In this embodiment, the above-mentioned server 320 is implemented as a cloud server as an example for illustration.

[0061] Taking the server implementation as a cloud server as an example, Figure 4 This is a schematic diagram of the cloud-client interaction process provided in an exemplary embodiment of this application, such as... Figure 4 As shown, the client first sends a recommendation request 410 to the cloud. The cloud responds to the recommendation request based on the recommendation service 420 and sends the recommendation results back to the client. The recommendation results include the recommended content displayed on the client.

[0062] The client exposes the recommended content to the client (430) and accepts user consumption (440). User consumption (440) refers to the user's reception and feedback of the displayed recommended content. The user's consumption result is analyzed by the behavior trigger decision module (450) to determine whether the current user consumption (440) triggers a reordering, that is, whether the user consumption (440) meets the reordering conditions. When the reordering conditions are met, a recommendation interface (460) is requested to send a recommendation request to the cloud. The cloud executes the reordering process through the reordering service (470) and feeds back the reordering result list (480) to the client. The client adjusts the playlist (490) according to the reordering result list (480) and continues to expose the recommended content to the client (430).

[0063] Based on the above description, the content recommendation method provided in this application will be explained. This method can be executed by a server or a terminal, or by both a server and a terminal. In this embodiment, the method is illustrated by being executed by a terminal. Figure 5 As shown, the method includes the following steps.

[0064] Step 501: Receive interactive operations on the content browsing interface of the target application, which also includes cached recommended content to be displayed.

[0065] Interactive actions are used to indicate the real-time interaction with the displayed recommended content.

[0066] That is, after the terminal launches the target application and displays the content browsing interface, it monitors the user's interactive operations on the content browsing interface in real time based on user authorization, such as: swiping operations, clicking operations, long-pressing operations, and hovering operations. The interactive operations also include the operation object, such as: clicking the like control, or long-pressing the target content. Optionally, the interactive operations can also be implemented as continuous operations, such as: long-pressing the target content followed by clicking the like control. This application embodiment does not limit this.

[0067] In the target application's content browsing interface, cached recommended content is updated and displayed in batches. For example, after the terminal retrieves 50 recommended items from the cloud, it first displays the first 10, and stores the remaining 40 in the cache. Once the first 10 items are displayed and the user requests further content, the first 10 are retrieved from the remaining 40 and displayed. When all 50 items are displayed, or the last 10 begin loading and displaying on the content browsing interface, another 50 items are retrieved from the cloud and cached for later display. It is worth noting that the above values ​​are merely illustrative examples; this application does not limit the number of recommended items retrieved. For instance, recommended content can be loaded and displayed one item at a time in the content browsing interface.

[0068] Cached recommended content refers to the recommended content that the aforementioned terminal has retrieved from the cloud but has not yet displayed or loaded into the content browsing interface.

[0069] It is worth noting that the above-mentioned recommended content in the browsing interface is displayed side by side in the content browsing interface; or, it is displayed sequentially in the content browsing interface in the form of swiping and switching; or, it is displayed in the content browsing interface in the form of a waterfall flow by swiping in and out of the screen. This application embodiment does not limit this.

[0070] To illustrate, let's take a short video app as an example. The client retrieves 10 short videos from the cloud and caches them. The first short video among the 10 short videos is played in the current content browsing interface. When a "like" action is received on the first short video in the content browsing interface, the "like" action is treated as an interaction.

[0071] Step 502: In response to the interaction operation meeting the reordering conditions, obtain the reordered recommended content corresponding to the interaction operation.

[0072] Among them, the rearrangement conditions are conditions configured for interaction preferences, and the rearranged recommended content is the content obtained by rearranging the content based on the interaction preferences corresponding to the interaction operations.

[0073] In other words, when an interactive action meets the reordering criteria, it indicates that the interactive action reflects an interactive tendency that requires the recommended content to be reordered, such as being interested in the currently displayed content or not being interested in the currently displayed content.

[0074] In some embodiments, obtaining the re-ranked recommended content corresponding to the interactive operation includes at least one of the following methods:

[0075] First, in response to an interactive action meeting the reordering criteria, a reordering request is sent to the server. The server then reorders the recommended content based on the interactive tendencies of the interactive action and sends the reordered recommended content back to the terminal.

[0076] In one scenario, the server reorders the recommended content in the content pool; or, the server reorders the content in the content pool and the cached recommended content.

[0077] Second, in response to interactive operations that meet the reordering conditions, the cached recommended content to be displayed is reordered based on the interaction tendency of the interactive operation, resulting in reordered recommended content.

[0078] The above-described method for obtaining rearranged recommended content is merely an illustrative example, and this application does not limit it. In this embodiment, the example of the server executing the rearrangement process to obtain rearranged recommended content and feeding it back to the terminal is used for illustration.

[0079] In some embodiments, the interactive action is matched with actions in a preset action set, which is a set of actions set based on interaction preferences; in response to the matching of the interactive action with the actions in the preset action set, the reordered recommended content corresponding to the interactive action is obtained.

[0080] In some embodiments, the interaction tendency can be implemented as a specific interaction tendency, such as: a positive interaction tendency; or, a negative interaction tendency;

[0081] Alternatively, interaction tendency can be divided into at least two types of interaction tendency, such as positive interaction tendency and negative interaction tendency, where positive interaction tendency is used to indicate interest in the displayed recommended content, and negative interaction tendency is used to indicate the tendency to ignore the displayed recommended content.

[0082] For example: illustrative, such as Figure 6 As shown, interaction tendencies are divided into explicit positive interaction tendencies 610, implicit positive interaction tendencies 620, explicit negative interaction tendencies 630, and implicit negative interaction tendencies 640. Explicit positive interaction tendencies 610 indicate a clear interest tendency in the displayed recommended content, implicit positive interaction tendencies 620 indicate an implicit interest tendency in the displayed recommended content, explicit negative interaction tendencies 630 indicate a clear tendency to ignore the displayed recommended content, and implicit negative interaction tendencies 640 indicate an implicit tendency to ignore the displayed recommended content.

[0083] Among them, explicit positive interaction tendencies include user behaviors such as liking, commenting, sharing, and following.

[0084] Implicit positive interaction tendency: For example, a high completion rate in short video recommendation scenarios indicates that users have an implicit positive interest in the recommended videos.

[0085] Explicit negative interaction tendencies: such as clicking "not interested" or other user engagement behaviors.

[0086] Implicit negative interaction tendency: For example, in a short video recommendation scenario, if a user swipes through N videos in a row, it means that the user has an implicit negative behavior towards the recommended videos.

[0087] Taking the server's reordering of recommended content based on the interaction tendency of interactive operations as an example, in some embodiments, after the terminal determines the interaction tendency corresponding to the interactive operation, it sends a reordering request to the server indicating the interaction tendency; or, the terminal sends an operation identifier of the interactive operation to the server, and the server determines the interaction tendency corresponding to the interactive operation based on the operation identifier.

[0088] When the terminal determines the interaction tendency corresponding to the interactive operation, the terminal obtains a preset operation set corresponding to at least two interaction tendencies, and matches the interactive operation with the operations in the preset operation sets corresponding to at least two interaction tendencies, thereby: 1. determining whether the interactive operation meets the rearrangement conditions; 2. when the rearrangement conditions are met, determining the interaction tendency corresponding to the interactive operation, that is, the interaction tendency corresponding to the preset operation set matched with the interactive operation.

[0089] When the server determines the interaction tendency corresponding to the interactive operation, the terminal sends a rearrangement request to the server, which includes the operation identifier of the interactive operation. The server includes an interest level assessment module. The operation identifier is input into the interest level assessment module, and the interaction tendency of the interactive operation is analyzed by the interest level assessment module to obtain the interaction tendency corresponding to the interactive operation.

[0090] Step 503: Replace the cached recommended content to be displayed with the rearranged recommended content.

[0091] Optionally, the stored cached recommended content can be cleared from the cache area, and the rearranged recommended content can be stored sequentially in the cache area based on the order of the rearranged recommended content.

[0092] In other words, the stored cached recommended content is deleted, and the rearranged recommended content is stored in the cache area and displayed after the recommended content in the current content browsing interface has been displayed.

[0093] In some embodiments, the client retrieves a list of rearranged results from the cloud. The list of rearranged results includes sequentially arranged rearranged recommended content. After retrieving the rearranged recommended content from the list of rearranged results, the client overwrites the cached recommended content for storage.

[0094] Step 504: Rearrange the recommended content and connect it with the already displayed recommended content on the content browsing interface.

[0095] In some embodiments, the recommended content is rearranged based on its order in the cache, and then displayed sequentially on the content browsing interface.

[0096] Optionally, the rearranged recommended content is divided according to the order of arrangement to obtain at least two sequentially arranged content batches; the content batch at the top of the list is then linked with the already displayed recommended content and displayed on the content browsing interface.

[0097] This is illustrative; please refer to it. Figure 7 The content browsing interface 700 displays recommended content, and the current content browsing interface 700 includes 10 loaded recommended items, while the client also caches 40 cached recommended items. For example... Figure 7 As shown, when a user likes recommended content 710, the recommended content is rearranged based on the like, resulting in 50 rearranged recommended content items, which then replace the original 40 cached recommended content items in the cache area for subsequent display.

[0098] like Figure 8 As shown, in this embodiment, the client 800 mainly includes the following modules:

[0099] Behavior Trigger Decision Module 810: Used to decide whether to trigger cloud-based reordering service based on whether user behavior meets the set business rules;

[0100] Reordering Request Module 820: If the triggering conditions for cloud reordering are met, this module is responsible for initiating a reordering service request (i.e., the above reordering request) and reporting all items that have not been exposed to users to the cloud.

[0101] Playlist Adjustment Module 830: This module will replace the cached regular recommendations with the rearranged results for display.

[0102] In summary, the content recommendation method provided in this application matches reordering conditions based on the user's real-time interactive operations. When the reordering conditions are met, the recommended content is reordered based on the interactive operations to obtain reordered recommended content. The reordered recommended content replaces the cached but unloaded content for subsequent display, improving the real-time correlation between recommended content and interactive operations. It transforms the user's real-time operations into conditions for recalling recommended content, thereby improving the accuracy and effectiveness of content recommendation.

[0103] In an optional embodiment, after the client sends a reordering request to the cloud, the cloud reorders the recommended content through a reordering service. (Illustrative example) Figure 9This is a flowchart of a content recommendation method provided in another exemplary embodiment of this application. This method can be executed by a server or a terminal, or jointly by both. In this embodiment, the method is described using an example of execution by a server (e.g., a cloud server). Figure 9 As shown, the method includes the following steps.

[0104] Step 901: Receive a rearrangement request sent by the terminal. The rearrangement request includes interactive operations and cached recommended content not displayed in the target application.

[0105] Optionally, the target application is an application installed on the terminal, wherein the interactive operation is a real-time operation received by the terminal on the content browsing interface of the target application.

[0106] In some embodiments, when a real-time interactive operation is received on the terminal, the terminal matches the interactive operation with preset rearrangement conditions, such as matching the interactive operation with operations in a preset operation set, where the preset operation set is an operation set set based on interactive preferences. When the interactive operation matches an operation in the preset operation set, it is determined that the interactive operation meets the rearrangement conditions, and a rearrangement request is sent.

[0107] In some embodiments, the reordering request includes an operation identifier for the interactive operation and a content identifier for the cached recommended content. For example, the interactive operation received by the terminal is a click operation on the like control of content 0, with the corresponding operation identifier being a0. If the cached recommended content not displayed in the current target application includes "1, 2, 3, 4, 5", then the reordering request includes "a0; 1 2 3 4 5".

[0108] Step 902: Determine the interaction tendency corresponding to the interactive operation.

[0109] Interaction tendency includes positive interaction tendency and negative interaction tendency. Positive interaction tendency indicates interest in the displayed recommended content, while negative interaction tendency indicates the tendency to ignore the displayed recommended content.

[0110] In some embodiments, the server stores an interest assessment module, which evaluates the interaction tendency of interactive operations based on the interest assessment module. Optionally, the interest assessment module includes a pre-trained neural network model, which evaluates the interactive operation to obtain the interaction tendency of the interactive operation. The neural network model is trained using sample interactive operations labeled with reference interaction tendencies.

[0111] In other embodiments, a preset set of operations corresponding to at least two interaction tendencies is obtained; the interaction operation is matched with the operations in the preset set of operations corresponding to at least two interaction tendencies to obtain the interaction tendency corresponding to the interaction operation.

[0112] In some embodiments, positive interaction tendency includes explicit positive interaction tendency and implicit positive interaction tendency; negative interaction tendency includes explicit negative interaction tendency and implicit negative interaction tendency.

[0113] Step 903: Based on interaction preferences, the cached recommended content and the recommended content in the content pool are rearranged to obtain rearranged recommended content.

[0114] Optionally, the rearrangement methods differ depending on whether the interaction tendency is positive or negative. For example, [illustrative example follows]. Figure 10 As shown, after interest assessment 1020 in the reordering service 1010, the interaction tendency corresponding to the interaction operation is determined, and it is determined whether the interaction tendency corresponding to the interaction operation is a positive tendency 1030. If the interaction tendency of the interaction operation is a positive tendency, then positive interest chasing 1040 is performed to reorder the recommended content; if the interaction tendency of the interaction operation is a negative tendency, then negative interest filtering 1050 is performed to reorder the recommended content.

[0115] In some embodiments, in response to an interaction tendency corresponding to an interaction operation being a positive interaction tendency, the cached recommended content and the recommended content in the content pool are rearranged based on a first rearrangement rule corresponding to the positive interaction tendency to obtain rearranged recommended content; wherein, the first rearrangement rule is used to prioritize recommending similar content to the recommended content targeted by the interaction operation.

[0116] In response to the interaction tendency corresponding to the interaction operation being negative, the cached recommended content and the recommended content in the content pool are rearranged based on the second rearrangement rule corresponding to the negative interaction tendency to obtain rearranged recommended content. The second rearrangement rule is used to filter and recommend similar content to the recommended content targeted by the interaction operation.

[0117] Step 904: Feedback the rearranged recommended content to the terminal.

[0118] After obtaining the rearranged recommended content, the rearranged recommended content is fed back to the target application on the terminal. The rearranged recommended content is used to replace the cached recommended content that is not displayed in the target application.

[0119] In summary, the content recommendation method provided in this application matches reordering conditions based on the user's real-time interactive operations. When the reordering conditions are met, the recommended content is reordered based on the interactive operations to obtain reordered recommended content. The reordered recommended content replaces the cached but unloaded content for subsequent display, improving the real-time correlation between recommended content and interactive operations. It transforms the user's real-time operations into conditions for recalling recommended content, thereby improving the accuracy and effectiveness of content recommendation.

[0120] The method provided in this embodiment reorders recommended content based on the interactive tendencies of interactive operations. This reordering incorporates the interactive interests reflected in the interactive operations themselves into the reordering of recommended content, thereby improving the ranking accuracy of recommended content.

[0121] The method provided in this embodiment, through data interaction between the terminal and the cloud, allows the cloud to rearrange the recommended content in the content pool and the cached recommended content not displayed on the terminal, ensuring content coverage and avoiding the problem of low content recall caused by the content originally recommended to the terminal being ignored during the rearrangement process.

[0122] In an optional embodiment, the interaction between the terminal and the server is used as an example for illustration. The aforementioned positive and negative interaction tendencies are both divided into explicit and implicit. Figure 11 This is a flowchart of a content recommendation method provided in another exemplary embodiment of this application. For example... Figure 11 As shown, the method includes:

[0123] Step 1101: The terminal receives interactive operations on the content browsing interface of the target application.

[0124] The target application also includes cached recommendations to be displayed. Interactive actions are used to indicate real-time interaction with the displayed recommendations.

[0125] Step 1102: In response to the interactive operation meeting the rearrangement conditions, the terminal sends a rearrangement request to the server.

[0126] Among them, the rearrangement condition is a condition configured based on the interaction tendency.

[0127] In some embodiments, an interactive action is matched with an action in a preset action set, which is an action set based on interaction preferences; in response to the matching of an interactive action with an action in the preset action set, it is determined that the interactive action meets the rearrangement conditions.

[0128] Optionally, if the interactive action does not conform to the reordered recommendation, the cached recommended content will continue to be displayed to the user in order.

[0129] Step 1103: The server receives the rearrangement request sent by the terminal.

[0130] The reordering request includes interactive actions and cached recommendations not displayed in the target application.

[0131] Step 1104: The server determines the interaction tendency corresponding to the interactive operation.

[0132] Interaction tendency includes positive interaction tendency and negative interaction tendency. Positive interaction tendency indicates interest in the displayed recommended content, while negative interaction tendency indicates the tendency to ignore the displayed recommended content.

[0133] In some embodiments, a preset set of operations corresponding to at least two interaction tendencies is obtained; the interaction operation is matched with the operations in the preset set of operations corresponding to at least two interaction tendencies to obtain the interaction tendency corresponding to the interaction operation.

[0134] In some embodiments, positive interaction tendency includes explicit positive interaction tendency and implicit positive interaction tendency; negative interaction tendency includes explicit negative interaction tendency and implicit negative interaction tendency.

[0135] Then, obtain the first set of operations corresponding to explicit positive interaction tendency; the second set of operations corresponding to implicit positive interaction tendency; the third set of operations corresponding to explicit negative interaction tendency; and the fourth set of operations corresponding to implicit negative interaction tendency. After matching the interaction operations with the first set of operations, the second set of operations, the third set of operations, and the fourth set of operations, obtain the interaction tendency corresponding to the interaction operation.

[0136] Step 1105: The server reorders the cached recommended content and the recommended content in the content pool based on interaction preferences to obtain reordered recommended content.

[0137] Optionally, in this embodiment, the interaction tendency includes explicit positive interaction tendency, implicit positive interaction tendency, explicit negative interaction tendency, and implicit negative interaction tendency as examples for illustration.

[0138] one, Explicit positive interaction tendency In response to the interaction tendency corresponding to the interactive operation being to display a positive interaction tendency, a first similarity is determined between the cached recommended content and the recommended content corresponding to the interactive operation, and a second similarity is determined between the recommended content in the content pool and the recommended content corresponding to the interactive operation. Based on the first similarity and the second similarity, the cached recommended content and the recommended content in the content pool are rearranged to obtain rearranged recommended content.

[0139] For explicit positive interaction tendencies, the re-ranking recommendation module uses the item that triggered the interaction to recall similar items. These recalled items are then merged and scored with items not yet displayed to the client to obtain the final re-ranking result, which is then sent to the client. The re-ranking result ensures that items with explicit positive interaction tendencies are displayed more prominently in subsequent screen positions. During the re-ranking process, after arranging the recommended content based on a first similarity and a second similarity, a first proportion of trending content or other content is randomly incorporated into the ordered recommended content to meet diversity requirements; for example, the first proportion is 20%.

[0140] two, Implicit positive interaction tendency In response to the implicit positive interaction tendency corresponding to the interactive operation, the first similarity between the cached recommended content and the recommended content corresponding to the interactive operation is determined, as well as the second similarity between the recommended content in the content pool and the recommended content corresponding to the interactive operation. After a preset number of recommended content items are added, the cached recommended content and the similar recommended content in the content pool are arranged based on the first and second similarities to obtain the rearranged recommended content.

[0141] For implicit positive interaction tendencies, the re-ranking recommendation module uses the item that triggered the interaction to recall similar items. The recalled items are then merged and scored with items not exposed to the client to obtain the final re-ranking result, which is then sent to the client. In the re-ranking result, similar items in the implicit positive interests will be displayed after a certain delay, such as displaying the item with the highest similarity after a delay of n recommended content intervals, where n is a preset positive integer.

[0142] It is worth noting that when scoring items corresponding to the aforementioned positive interaction tendency, the scoring is based on the similarity between the item and the interactive operation, as well as other preset scoring rules, such as diversity rules and interest tag rules.

[0143] Among them, the diversity rule refers to the rule of mixing hot content, topic content and ranking results to achieve the diversity requirement; the interest tag rule refers to the rule of improving the scoring results of content related to the user's own interest tags.

[0144] three, Explicit negative interaction tendency In response to the interaction tendency corresponding to the interactive operation being to display a negative interaction tendency, the first similarity between the cached recommended content and the recommended content corresponding to the interactive operation is determined, as well as the second similarity between the recommended content in the content pool and the recommended content corresponding to the interactive operation. Based on the first similarity and the second similarity, the cached recommended content and the recommended content in the content pool are filtered and rearranged to obtain the rearranged recommended content.

[0145] For explicit negative interaction tendencies, the negative interest filtering module will filter out similar items from the unexposed item list and content pool, then score the remaining items, obtain the reordering results, and send them to the client. The reordering results ensure that the subsequent filtering of similar items in explicit negative interests takes effect for a longer period than that of implicit negative interaction tendencies, such as filtering similar items within 2 hours.

[0146] Four, Implicit negative interaction tendency In response to the implicit negative interaction tendency corresponding to the interactive operation, the first similarity between the cached recommended content and the recommended content corresponding to the interactive operation is determined, as well as the second similarity between the recommended content in the content pool and the recommended content corresponding to the interactive operation. Based on the first and second similarities, the cached recommended content and the recommended content in the content pool are partially filtered and rearranged to obtain the rearranged recommended content.

[0147] For implicit negative interaction tendencies, the negative interest filtering module will filter out similar items from the unexposed item list and content pool, then score the remaining items, obtain the reordering results, and send them to the client. The reordering results ensure that the subsequent filtering of similar items in implicit negative interests takes effect within a shorter period than that of explicit negative interaction tendencies, such as filtering similar items within 15 minutes.

[0148] It is worth noting that the scoring of items mentioned above is based on preset scoring rules, such as diversity rules and interest tag rules.

[0149] Step 1106: The server sends the rearranged recommended content back to the terminal.

[0150] After obtaining the rearranged recommended content, the rearranged recommended content is fed back to the target application on the terminal. The rearranged recommended content is used to replace the cached recommended content that is not displayed in the target application.

[0151] Step 1107: The terminal replaces the cached recommended content to be displayed with the rearranged recommended content.

[0152] Optionally, the stored cached recommended content can be cleared from the cache area, and the rearranged recommended content can be stored sequentially in the cache area based on the order of the rearranged recommended content.

[0153] In other words, the stored cached recommended content is deleted, and the rearranged recommended content is stored in the cache area and displayed after the recommended content in the current content browsing interface has been displayed.

[0154] In some embodiments, the client retrieves a list of rearranged results from the cloud. The list of rearranged results includes sequentially arranged rearranged recommended content. After retrieving the rearranged recommended content from the list of rearranged results, the client overwrites the cached recommended content for storage.

[0155] In some embodiments, after the rearranged recommended content is stored in the cache, the rearranged recommended content is implemented as the updated cached recommended content. When the content browsing interface continues to receive interactive operations, the updated cached recommended content will continue to be updated or not updated according to the interactive operation.

[0156] Step 1108: The terminal will rearrange the recommended content and connect it with the already displayed recommended content on the content browsing interface.

[0157] In some embodiments, the recommended content is rearranged based on its order in the cache, and then displayed sequentially on the content browsing interface.

[0158] Optionally, the rearranged recommended content is divided according to the order of arrangement to obtain at least two sequentially arranged content batches; the content batch at the top of the list is then linked with the already displayed recommended content and displayed on the content browsing interface.

[0159] In summary, the content recommendation method provided in this application matches reordering conditions based on the user's real-time interactive operations. When the reordering conditions are met, the recommended content is reordered based on the interactive operations to obtain reordered recommended content. The reordered recommended content replaces the cached but unloaded content for subsequent display, improving the real-time correlation between recommended content and interactive operations. It transforms the user's real-time operations into conditions for recalling recommended content, thereby improving the accuracy and effectiveness of content recommendation.

[0160] The method provided in this embodiment can effectively improve user retention in typical application scenarios, such as short video recommendation. Some test results show that for users using the reordering service, the cumulative average program usage time per user increases significantly by 0.81%, and the average number of launches per user increases by 0.37%.

[0161] The method provided in this embodiment, cloud-based collaborative recommendation, can leverage real-time user feedback on the client side to trigger cloud-based reordering logic, better capturing user interests. This increases the overall satisfaction of the recommendation system without significantly increasing the system's QPS. Specifically, cloud-based collaborative recommendation refers to a system where the client monitors user's real-time behavior and returns this feedback to the cloud, which then adjusts the recommendation results based on this behavior.

[0162] Figure 12 This is a structural block diagram of a content recommendation apparatus provided in an exemplary embodiment of this application, such as... Figure 12 As shown, the device includes:

[0163] The receiving module 1210 is used to receive interactive operations on the content browsing interface of the target application, the interactive operations being used to indicate the real-time interaction with the displayed recommended content, the target application also including cached recommended content to be displayed;

[0164] The acquisition module 1220 is used to acquire rearranged recommended content corresponding to the interaction operation in response to the interaction operation meeting the rearrangement conditions. The rearrangement conditions are conditions configured for interaction tendencies, and the rearranged recommended content is content rearranged based on the interaction tendencies corresponding to the interaction operation.

[0165] Update module 1230 is used to replace the cached recommended content to be displayed with the rearranged recommended content;

[0166] Display module 1240 is used to connect the rearranged recommended content with the already displayed recommended content and display it on the content browsing interface.

[0167] In an optional embodiment, such as Figure 13 As shown, the acquisition module 1220 includes:

[0168] The matching unit 1221 is used to match the interactive operation with operations in a preset operation set, wherein the preset operation set is an operation set set based on the interactive tendency.

[0169] The acquisition unit 1222 is used to acquire the rearranged recommended content corresponding to the interactive operation in response to the matching of the interactive operation with the operation in the preset operation set.

[0170] In an optional embodiment, the acquisition unit 1222 is further configured to acquire preset operation sets corresponding to at least two interaction tendencies, including positive interaction tendencies and negative interaction tendencies, wherein the positive interaction tendency is used to indicate an interest tendency towards the displayed recommended content, and the negative interaction tendency is used to indicate an inclination to ignore the displayed recommended content.

[0171] The matching unit 1221 is further configured to match the interactive operation with the operations in the preset operation set corresponding to the at least two interactive tendencies.

[0172] In an optional embodiment, the device further includes:

[0173] The sending module 1250 is used to send a reordering request to the server in response to the interaction operation meeting the reordering conditions. The reordering request includes the operation identifier of the interaction operation and the content identifier of the cached recommended content not displayed in the target application.

[0174] The receiving module 1210 is further configured to receive the rearranged recommended content fed back by the server, wherein the rearranged recommended content is the content after the server rearranges the content in the content pool and the cached recommended content based on the operation identifier of the interactive operation.

[0175] In an optional embodiment, the acquisition module 1220 is further configured to, in response to the interaction operation meeting the reordering conditions, reorder the cached recommended content to be displayed based on the interaction operation to obtain the reordered recommended content.

[0176] In an optional embodiment, the update module 1230 is further configured to clear the stored cached recommended content from the cache area; and to store the rearranged recommended content sequentially into the cache area based on the arrangement order of the rearranged recommended content.

[0177] In an optional embodiment, the display module 1240 is further configured to sequentially display the already displayed recommended content on the content browsing interface based on the arrangement order of the rearranged recommended content in the cache area.

[0178] In an optional embodiment, the display module 1240 is further configured to divide the rearranged recommended content according to the order of arrangement to obtain at least two sequentially arranged content batches; and to connect the first-ranked content batch with the already displayed recommended content for display on the content browsing interface.

[0179] Figure 14 This is a structural block diagram of a content recommendation apparatus provided in an exemplary embodiment of this application, such as... Figure 14 As shown, the device includes:

[0180] The receiving module 1410 is used to receive a reordering request sent by a terminal, wherein a target application is installed in the terminal, and the reordering request includes an interactive operation and cached recommended content not displayed in the target application, wherein the interactive operation is a real-time operation received by the terminal on the content browsing interface of the target application;

[0181] The determining module 1420 is used to determine the interaction tendency corresponding to the interactive operation, wherein the interaction tendency includes a positive interaction tendency and a negative interaction tendency. The positive interaction tendency is used to indicate the interest tendency towards the displayed recommended content, and the negative interaction tendency is used to indicate the tendency to ignore the displayed recommended content.

[0182] The sorting module 1430 is used to rearrange the cached recommended content and the recommended content in the content pool based on the interaction tendency to obtain rearranged recommended content;

[0183] The sending module 1440 is used to send the rearranged recommended content back to the terminal, and the rearranged recommended content is used to replace the cached recommended content for display.

[0184] In an optional embodiment, the determining module 1420 is further configured to obtain a preset operation set corresponding to at least two interaction tendencies; and match the interaction operation with the operations in the preset operation set corresponding to the at least two interaction tendencies to obtain the interaction tendency corresponding to the interaction operation.

[0185] In an optional embodiment, the arrangement module 1430 is further configured to, in response to the interaction tendency corresponding to the interaction operation being a positive interaction tendency, rearrange the cached recommended content and the recommended content in the content pool based on the first rearrangement rule corresponding to the positive interaction tendency, to obtain the rearranged recommended content;

[0186] The arrangement module 1430 is further configured to, in response to the interaction tendency corresponding to the interaction operation being a negative interaction tendency, rearrange the cached recommended content and the recommended content in the content pool based on the second rearrangement rule corresponding to the negative interaction tendency, to obtain the rearranged recommended content.

[0187] In an optional embodiment, the positive interaction tendency includes explicit positive interaction tendency and implicit positive interaction tendency;

[0188] The negative interaction tendency includes explicit negative interaction tendency and implicit negative interaction tendency.

[0189] In an optional embodiment, the sorting module 1430 is further configured to, in response to the interaction tendency corresponding to the interaction operation being a positive interaction tendency, determine a first similarity between the cached recommended content and the recommended content corresponding to the interaction operation, and a second similarity between the recommended content in the content pool and the recommended content corresponding to the interaction operation; and rearrange the cached recommended content and the recommended content in the content pool based on the first similarity and the second similarity to obtain the rearranged recommended content;

[0190] The arrangement module 1430 is further configured to, in response to the interaction tendency corresponding to the interaction operation being an implicit positive interaction tendency, determine a first similarity between the cached recommended content and the recommended content corresponding to the interaction operation, and a second similarity between the recommended content in the content pool and the recommended content corresponding to the interaction operation; after a preset number of recommended content items, arrange the cached recommended content and similar recommended content in the content pool based on the first similarity and the second similarity to obtain the rearranged recommended content.

[0191] In an optional embodiment, the sorting module 1430 is further configured to, in response to the interaction tendency corresponding to the interaction operation being a negative interaction tendency, determine a first similarity between the cached recommended content and the recommended content corresponding to the interaction operation, and a second similarity between the recommended content in the content pool and the recommended content corresponding to the interaction operation; and filter and rearrange the cached recommended content and the recommended content in the content pool based on the first similarity and the second similarity to obtain the rearranged recommended content;

[0192] The sorting module 1430 is further configured to, in response to the interaction tendency corresponding to the interaction operation being an implicit negative interaction tendency, determine a first similarity between the cached recommended content and the recommended content corresponding to the interaction operation, and a second similarity between the recommended content in the content pool and the recommended content corresponding to the interaction operation; and based on the first similarity and the second similarity, partially filter and rearrange the cached recommended content and the recommended content in the content pool to obtain the rearranged recommended content.

[0193] It should be noted that the content recommendation device provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the content recommendation device and the content recommendation method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0194] Figure 15 This illustration shows a schematic diagram of a server provided in an exemplary embodiment of this application. The server may be as follows: Figure 3 The terminal or server shown.

[0195] Specifically, server 1500 includes a central processing unit (CPU) 1501, a system memory 1504 including random access memory (RAM) 1502 and read-only memory (ROM) 1503, and a system bus 1505 connecting the system memory 1504 and the CPU 1501. Server 1500 also includes a mass storage device 1506 for storing the operating system 1513, application programs 1514, and other program modules 1515.

[0196] Mass storage device 1506 is connected to central processing unit 1501 via a mass storage controller (not shown) connected to system bus 1505. Mass storage device 1506 and its associated computer-readable media provide non-volatile storage for server 1500. That is, mass storage device 1506 may include computer-readable media (not shown) such as hard disk or compact disc read-only memory (CD-ROM) drive.

[0197] Without loss of generality, computer-readable media can include computer storage media and communication media. Computer storage media includes volatile and non-volatile, removable and non-removable media implemented using any method or technology for storing information such as computer-readable instructions, data structures, program modules, or other data. Computer storage media includes RAM, ROM, erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other solid-state storage technologies, CD-ROM, digital versatile disc (DVD) or other optical storage, magnetic tape cassettes, magnetic tape, disk storage, or other magnetic storage devices. Of course, those skilled in the art will recognize that computer storage media are not limited to the above-mentioned types. The system memory 1504 and mass storage device 1506 described above can be collectively referred to as memory.

[0198] According to various embodiments of this application, server 1500 can also be connected to a remote computer on a network, such as the Internet. That is, server 1500 can be connected to network 1512 via network interface unit 1511 connected to system bus 1505, or it can also use network interface unit 1511 to connect to other types of networks or remote computer systems (not shown).

[0199] The aforementioned memory also includes one or more programs, which are stored in the memory and configured to be executed by the CPU.

[0200] Embodiments of this application also provide a computer device that can be implemented as follows: Figure 2 The terminal or server shown. The computer device includes a processor and a memory, the memory storing at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, at least one program, code set, or instruction set is loaded and executed by the processor to implement the content recommendation method provided in the above-described method embodiments.

[0201] Embodiments of this application also provide a computer-readable storage medium storing at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, at least one program, code set, or instruction set is loaded and executed by a processor to implement the content recommendation method provided in the above-described method embodiments.

[0202] Embodiments of this application also provide a computer program product or computer program that includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform any of the content recommendation methods described in the above embodiments.

[0203] Optionally, the computer-readable storage medium may include: read-only memory (ROM), random access memory (RAM), solid-state drives (SSDs), or optical discs, etc. The random access memory may include resistive random access memory (ReRAM) and dynamic random access memory (DRAM). The sequence numbers of the embodiments in this application are merely descriptive and do not represent the superiority or inferiority of the embodiments.

[0204] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0205] The above description is merely an optional embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A content recommendation method, characterized in that, The method includes: Receive interactive actions on the content browsing interface of the target application, the interactive actions being used to indicate real-time interaction with the displayed recommended content, the target application also including cached recommended content to be displayed; Obtain a preset set of operations corresponding to at least two interaction tendencies, including positive interaction tendencies and negative interaction tendencies, wherein the positive interaction tendency is used to indicate an interest tendency towards the displayed recommended content, and the negative interaction tendency is used to indicate an inclination to ignore the displayed recommended content. Match the interactive operation with the preset operation set corresponding to the at least two interactive tendencies; In response to the interaction operation matching with the operation in the preset operation set, the rearranged recommended content corresponding to the interaction operation is obtained. Specifically, if the interaction tendency corresponding to the interaction operation is positive, the cached recommended content and the recommended content in the content pool are rearranged based on the first rearrangement rule corresponding to the positive interaction tendency to obtain the rearranged recommended content. If the interaction tendency corresponding to the interaction operation is negative, the cached recommended content and the recommended content in the content pool are rearranged based on the second rearrangement rule corresponding to the negative interaction tendency to obtain the rearranged recommended content. Replace the cached recommended content to be displayed with the rearranged recommended content; The rearranged recommended content is then linked to the already displayed recommended content and displayed on the content browsing interface.

2. The method according to claim 1, characterized in that, The method further includes: In response to the interaction meeting the reordering conditions, a reordering request is sent to the server. The reordering request includes the operation identifier of the interaction and the content identifier of the cached recommended content that is not displayed in the target application. The server receives the rearranged recommended content, which is the content rearranged by the server based on the operation identifier of the interaction operation, after rearranging the content in the content pool and the cached recommended content.

3. The method according to claim 1, characterized in that, The method further includes: In response to the interaction operation meeting the reordering conditions, the cached recommended content to be displayed is reordered based on the interaction operation to obtain the reordered recommended content.

4. The method according to claim 1, characterized in that, The step of replacing the cached recommended content to be displayed with the rearranged recommended content includes: Clear the cached recommendations that have been stored in the cache area; Based on the arrangement order of the rearranged recommended content, the rearranged recommended content is stored sequentially in the cache area.

5. The method according to claim 4, characterized in that, The step of connecting the rearranged recommended content with the already displayed recommended content and displaying it on the content browsing interface includes: Based on the rearranged recommended content's order in the cache area, the already displayed recommended content is sequentially displayed on the content browsing interface.

6. The method according to claim 5, characterized in that, The step of sequentially displaying the already displayed recommended content on the content browsing interface based on the rearranged order of the recommended content in the cache includes: The rearranged recommended content is divided according to the order of arrangement to obtain at least two batches of content arranged in sequence; The top-ranked content will be displayed in batches, linking the already shown recommended content to the content browsing interface.

7. A content recommendation method, characterized in that, The method includes: A reordering request is sent by a receiving terminal, the terminal having a target application installed thereon. The reordering request includes an interactive operation and cached recommended content not displayed in the target application. The interactive operation is a real-time operation received by the terminal on the content browsing interface of the target application. Determine the interaction tendency corresponding to the interactive operation, wherein the interaction tendency includes positive interaction tendency and negative interaction tendency, the positive interaction tendency is used to indicate the interest tendency on the displayed recommended content, and the negative interaction tendency is used to indicate the tendency to ignore the displayed recommended content; In response to a positive interaction tendency corresponding to the interactive operation, the cached recommended content and the recommended content in the content pool are rearranged based on the first rearrangement rule corresponding to the positive interaction tendency to obtain rearranged recommended content; in response to a negative interaction tendency corresponding to the interactive operation, the cached recommended content and the recommended content in the content pool are rearranged based on the second rearrangement rule corresponding to the negative interaction tendency to obtain rearranged recommended content. The rearranged recommended content is fed back to the terminal, and the rearranged recommended content is used to replace the cached recommended content for display.

8. The method according to claim 7, characterized in that, Determining the interaction tendency corresponding to the interactive operation includes: Obtain a set of preset operations corresponding to at least two interaction tendencies; The interactive operation is matched with the operations in the preset operation set corresponding to the at least two interactive tendencies to obtain the interactive tendency corresponding to the interactive operation.

9. The method according to claim 8, characterized in that, The positive interaction tendency includes explicit positive interaction tendency and implicit positive interaction tendency; The negative interaction tendency includes explicit negative interaction tendency and implicit negative interaction tendency.

10. The method according to claim 9, characterized in that, The interaction tendency corresponding to the interactive operation is a positive interaction tendency. Based on the first reordering rule corresponding to the positive interaction tendency, the cached recommended content and the recommended content in the content pool are reordered to obtain reordered recommended content, including: In response to the interaction tendency corresponding to the interactive operation being a positive interaction tendency, a first similarity is determined between the cached recommended content and the recommended content corresponding to the interactive operation, and a second similarity is determined between the recommended content in the content pool and the recommended content corresponding to the interactive operation; based on the first similarity and the second similarity, the cached recommended content and the recommended content in the content pool are rearranged to obtain the rearranged recommended content; In response to the interaction tendency corresponding to the interaction operation being an implicit positive interaction tendency, a first similarity is determined between the cached recommended content and the recommended content corresponding to the interaction operation, and a second similarity is determined between the recommended content in the content pool and the recommended content corresponding to the interaction operation; after a preset number of recommended content items are added, the cached recommended content and similar recommended content in the content pool are arranged based on the first similarity and the second similarity to obtain the rearranged recommended content.

11. The method according to claim 9, characterized in that, The interaction tendency corresponding to the interactive operation is a negative interaction tendency. Based on the second reordering rule corresponding to the negative interaction tendency, the cached recommended content and the recommended content in the content pool are reordered to obtain the reordered recommended content, including: In response to the interaction tendency corresponding to the interaction operation being a negative interaction tendency, a first similarity is determined between the cached recommended content and the recommended content corresponding to the interaction operation, and a second similarity is determined between the recommended content in the content pool and the recommended content corresponding to the interaction operation; based on the first similarity and the second similarity, the cached recommended content and the recommended content in the content pool are filtered and rearranged to obtain the rearranged recommended content; In response to the interaction tendency corresponding to the interaction operation being an implicit negative interaction tendency, a first similarity is determined between the cached recommended content and the recommended content corresponding to the interaction operation, and a second similarity is determined between the recommended content in the content pool and the recommended content corresponding to the interaction operation; based on the first similarity and the second similarity, the cached recommended content and the recommended content in the content pool are partially filtered and rearranged to obtain the rearranged recommended content.

12. A content recommendation device, characterized in that, The device includes: A receiving module is used to receive interactive operations on the content browsing interface of a target application, the interactive operations being used to indicate the real-time interaction with the displayed recommended content, the target application also including cached recommended content to be displayed; An acquisition module is configured to acquire preset operation sets corresponding to at least two interaction tendencies, including positive and negative interaction tendencies. The positive interaction tendency indicates an interest in displayed recommended content, and the negative interaction tendency indicates an inclination to ignore displayed recommended content. The module matches the interaction operations with operations in the preset operation sets corresponding to the at least two interaction tendencies. In response to the matching of the interaction operation with operations in the preset operation sets, the module acquires rearranged recommended content corresponding to the interaction operation. Specifically, in response to a positive interaction tendency, the cached recommended content and recommended content in the content pool are rearranged based on a first rearrangement rule corresponding to the positive interaction tendency to obtain the rearranged recommended content. Similarly, in response to a negative interaction tendency, the cached recommended content and recommended content in the content pool are rearranged based on a second rearrangement rule corresponding to the negative interaction tendency to obtain the rearranged recommended content. The update module is used to replace the cached recommended content to be displayed with the rearranged recommended content; The display module is used to connect the rearranged recommended content with the already displayed recommended content and display it on the content browsing interface.

13. A content recommendation device, characterized in that, The device includes: A receiving module is used to receive a reordering request sent by a terminal, wherein a target application is installed in the terminal, and the reordering request includes an interactive operation and cached recommended content not displayed in the target application, wherein the interactive operation is a real-time operation received by the terminal on the content browsing interface of the target application; The determination module is used to determine the interaction tendency corresponding to the interactive operation, wherein the interaction tendency includes positive interaction tendency and negative interaction tendency. The positive interaction tendency is used to indicate the interest tendency towards the displayed recommended content, and the negative interaction tendency is used to indicate the tendency to ignore the displayed recommended content. The sorting module is configured to, in response to a positive interaction tendency corresponding to the interactive operation, rearrange the cached recommended content and the recommended content in the content pool based on a first rearrangement rule corresponding to the positive interaction tendency to obtain rearranged recommended content; and in response to a negative interaction tendency corresponding to the interactive operation, rearrange the cached recommended content and the recommended content in the content pool based on a second rearrangement rule corresponding to the negative interaction tendency to obtain rearranged recommended content. The sending module is used to send the rearranged recommended content back to the terminal, and the rearranged recommended content is used to replace the cached recommended content for display.

14. A computer device, characterized in that, The computer device includes a processor and a memory, the memory storing at least one program, which is loaded and executed by the processor to implement the content recommendation method as described in any one of claims 1 to 11.

15. A computer-readable storage medium, characterized in that, The storage medium stores at least one program segment, which is loaded and executed by a processor to implement the content recommendation method as described in any one of claims 1 to 11.

16. A computer program product, characterized in that, It includes a computer program or instructions that, when executed by a processor, implement the content recommendation method as described in any one of claims 1 to 11.

Images