Server, display device and media asset recommendation method

Abstract

The invention relates to a server, a display device and a media asset recommendation method. The media asset recommendation method comprises the following steps: receiving a media asset acquisition request sent by a display device, wherein the media asset acquisition request comprises environment feature data; the environment feature data is used for representing a request environment corresponding to the media asset acquisition request; in response to the media asset acquisition request, sorting the recalled recommended media assets according to the environment feature data to obtain target recommended media assets; and sending the target recommended media assets to the display device, so that the display device displays the target recommended media assets. By adopting the method, the content related to the environment feature data can be preferentially recommended to the user from the recalled recommended media assets, the media asset recommendation requirement of the user in the current application scene is better met, and the use experience of the user is improved.

Description

Technical Field

[0001] The present application relates to the technical field of display devices, and in particular, to a server, a display device, and a media asset recommendation method. Background Art

[0002] Currently, with the rapid development of multimedia platforms, users are faced with a vast amount of media asset content. How to accurately provide users with the required media assets has become a key problem that urgently needs to be solved.

[0003] Traditional media recommendation methods often rely on statistical models. By statistically analyzing users' behaviors such as likes and collections, relevant media assets are recommended to users. In this way, it is difficult to capture users' real needs, and there are problems such as low relevance between the recommended media assets and users' needs and poor recommendation accuracy. Summary of the Invention

[0004] The present application provides a server, a display device, and a media asset recommendation method to solve the problems of low relevance between recommended media assets and users' needs and poor recommendation accuracy.

[0005] In a first aspect, some embodiments provide a server, including:

[0006] A communication device configured to communicate with a display device;

[0007] And at least one processor, connected to the communication device and configured to:

[0008] Receive a media asset acquisition request sent by the display device, where the media asset acquisition request includes environmental feature data; the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request;

[0009] In response to the media asset acquisition request, sort the recalled recommended media assets according to the environmental feature data to obtain target recommended media assets;

[0010] Send the target recommended media assets to the display device so that the display device displays the target recommended media assets.

[0011] In the above embodiments, by introducing environmental feature data, since the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request, a recommendation basis is provided for subsequent media asset recommendation. By responding to the media asset acquisition request, sorting the recalled recommended media assets according to the environmental feature data, and obtaining the target recommended media assets to be displayed on the display device, content related to the environmental feature data can be preferentially recommended to users from the recalled recommended media assets, which better meets the media asset recommendation needs of users in the current application scenario and is conducive to improving the user experience.

[0012] In some embodiments, when the processor executes sorting the recalled recommended media assets according to the environmental feature data to obtain the target recommended media assets, it is configured to: determine the application scenario category to which the environmental feature data belongs, and the media asset category of the recalled recommended media assets; determine the correlation degree between the media asset category of the recalled recommended media assets and the application scenario category; and sort the recalled recommended media assets according to the corresponding correlation degree of the recalled recommended media assets to obtain the target recommended media assets.

[0013] In the above embodiments, by introducing the application scenario category, it is possible to identify the application scenario where the user is currently located according to the environmental feature data, providing a basis for the subsequent precise matching of media assets. By determining the correlation degree between the media asset category of the recalled recommended media assets and the application scenario category, and sorting the recalled recommended media assets according to the corresponding correlation degree of the recalled recommended media assets, it is possible to focus on the media assets with high correlation, which is beneficial to improving the accuracy and relevance of media recommendations.

[0014] In some embodiments, when the processor executes determining the correlation degree between the media asset category of the recalled recommended media assets and the application scenario category, it is configured to: obtain a media asset scenario mapping relationship table; wherein, the media asset scenario mapping relationship table includes the correlation degrees between different media asset categories and the corresponding reference scenario categories; and find the corresponding correlation degree of the recalled recommended media assets from the media asset scenario mapping relationship table according to the media asset category and the application scenario category of the recalled recommended media assets.

[0015] In the above embodiments, by finding the corresponding correlation degree of the recalled recommended media assets from the media asset scenario mapping relationship table according to the media asset category and the application scenario category of the recalled recommended media assets, it is beneficial to improve the determination efficiency of the correlation degree corresponding to the recalled recommended media assets.

[0016] In some embodiments, the media asset acquisition request includes user data; when the processor executes sorting the recalled recommended media assets according to the corresponding correlation degree of the recalled recommended media assets to obtain the target recommended media assets, it is configured to: extract the media asset feature data corresponding to the recalled recommended media assets, and extract the user feature data corresponding to the user data; determine the similarity between the media asset feature data corresponding to the recalled recommended media assets and the user feature data; perform an initial sorting on the recalled recommended media assets according to the corresponding similarity of the recalled recommended media assets; and perform a re - sorting on the initially sorted recalled recommended media assets according to the corresponding correlation degree of the recalled recommended media assets to obtain the target recommended media assets.

[0017] In the above embodiments, the similarity between the media feature data corresponding to the recalled and recommended media assets and the user feature data is determined; based on the similarity corresponding to the recalled and recommended media assets, an initial ranking is performed on the recalled and recommended media assets, so that the recalled and recommended media assets are more in line with the user characteristics, realizing personalized recommendation for users. On this basis, by re-ranking the recalled and recommended media assets after the initial ranking according to the relevance corresponding to the recalled and recommended media assets, the recommendation of the recalled and recommended media assets is further optimized from the application scenario dimension, improving the recommendation accuracy and being beneficial to improving the user experience.

[0018] In some embodiments, the media asset acquisition request further includes user data; the processor is further configured to: extract the media feature data corresponding to the candidate recommended media assets, and extract the user feature data corresponding to the user data; select the recalled and recommended media assets from the candidate recommended media assets according to the user feature data and the media feature data corresponding to the candidate recommended media assets.

[0019] In the above embodiments, by selecting the recalled and recommended media assets from the candidate recommended media assets according to the user feature data and the media feature data corresponding to the candidate recommended media assets, the recalled and recommended media assets can be more in line with the user characteristics, realizing personalized recommendation for users.

[0020] In some embodiments, when the processor executes the selection of the recalled and recommended media assets from the candidate recommended media assets according to the user feature data and the media feature data corresponding to the candidate recommended media assets, it is configured to: enhance the attention of the feature data to be processed based on the environmental feature data to update the feature data to be processed; wherein, the feature data to be processed includes the user feature data or the media feature data corresponding to the candidate recommended media assets; select the recalled and recommended media assets from the candidate recommended media assets according to the similarity between the media feature data corresponding to the candidate recommended media assets and the user feature data.

[0021] In the above embodiments, by enhancing the attention of the feature data to be processed based on the environmental feature data, the user feature data or the media feature data corresponding to the candidate recommended media assets can be focused on the current application scenario, which is beneficial to improving the relevance between the recalled and recommended media assets and the current application scenario.

[0022] In some embodiments, when the processor executes the attention enhancement of the feature data to be processed, it is configured to: determine the semantic matching degree between the different dimension features of the feature data to be processed and the environmental feature data; determine the feature weights of the corresponding dimension features according to the semantic matching degrees corresponding to the different dimension features; determine the updated feature data to be processed according to the different dimension features and the corresponding feature weights.

[0023] In the above embodiments, the matching degree between different-dimensional features and environmental features is quantified through semantic matching degree, so that the feature weights can be dynamically adjusted according to the changes in the scenario. By determining the updated feature data to be processed based on different-dimensional features and corresponding feature weights, the attention enhancement of the feature data to be processed is realized, enabling the feature data to be processed to more accurately represent the current needs of the user.

[0024] In a second aspect, some embodiments provide a display device, including:

[0025] A display configured to display content;

[0026] At least one controller communicatively connected to the display and configured to:

[0027] Send a media asset acquisition request to the server, the media asset acquisition request including environmental feature data; the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request;

[0028] Receive a target recommended media asset feedback by the server based on the media asset acquisition request; the target recommended media asset is obtained by the server sorting the recalled recommended media assets according to the environmental feature data;

[0029] Control the display to display the target recommended media asset.

[0030] In the above embodiments, by introducing environmental feature data, since the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request, a recommendation basis is provided for subsequent media asset recommendation. By receiving the target recommended media asset feedback by the server based on the media asset acquisition request and controlling the display to display the target recommended media asset, the display of the target recommended media asset is realized. Since the target recommended media asset is obtained by the server sorting the recalled recommended media assets according to the environmental feature data, the target recommended media asset better meets the media asset recommendation needs of the user in the current application scenario, which is beneficial to improving the user experience.

[0031] In a third aspect, some embodiments provide a media asset recommendation method applied to a server, including:

[0032] Receive a media asset acquisition request sent by a display device, the media asset acquisition request including environmental feature data; the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request;

[0033] In response to the media asset acquisition request, sort the recalled recommended media assets according to the environmental feature data to obtain a target recommended media asset;

[0034] Send the target recommended media asset to the display device so that the display device displays the target recommended media asset.

[0035] In the above embodiments, by introducing environmental feature data, since the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request, a recommendation basis is provided for subsequent media asset recommendations. By responding to the media asset acquisition request and sorting the recalled recommended media assets according to the environmental feature data, the target recommended media assets are obtained and displayed on the display device, so that the content related to the environmental feature data can be preferentially recommended to the user from the recalled recommended media assets, which better meets the media asset recommendation requirements of the user in the current application scenario and is conducive to improving the user experience.

[0036] Fourthly, some embodiments provide another media asset recommendation method, which is applied to a display device and includes:

[0037] Sending a media asset acquisition request to a server, where the media asset acquisition request includes environmental feature data; the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request;

[0038] Receiving the target recommended media assets fed back by the server based on the media asset acquisition request; the target recommended media assets are obtained by the server sorting the recalled recommended media assets according to the environmental feature data;

[0039] Controlling the display to display the target recommended media assets.

[0040] In the above embodiments, by introducing environmental feature data, since the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request, a recommendation basis is provided for subsequent media asset recommendations. By receiving the target recommended media assets fed back by the server based on the media asset acquisition request and controlling the display to display the target recommended media assets, the display of the target recommended media assets is realized. Since the target recommended media assets are obtained by the server sorting the recalled recommended media assets according to the environmental feature data, the target recommended media assets better meet the media asset recommendation requirements of the user in the current application scenario and are conducive to improving the user experience.

[0041] Fifthly, some embodiments provide a media asset recommendation device, which is applied to a server and includes:

[0042] A first receiving module, configured to receive a media asset acquisition request sent by a display device, where the media asset acquisition request includes environmental feature data; the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request;

[0043] A processing module, configured to respond to the media asset acquisition request and sort the recalled recommended media assets according to the environmental feature data to obtain target recommended media assets;

[0044] A first sending module, configured to send the target recommended media assets to the display device so that the display device displays the target recommended media assets.

[0045] In the above embodiments, by introducing environmental feature data, since the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request, a recommendation basis is provided for subsequent media asset recommendations. By responding to the media asset acquisition request and sorting the recalled recommended media assets according to the environmental feature data, the target recommended media assets are obtained and displayed on the display device, so that content related to the environmental feature data can be preferentially recommended to the user from the recalled recommended media assets, which better meets the media asset recommendation requirements of the user in the current application scenario and is conducive to improving the user experience.

[0046] Sixthly, in some embodiments, another media asset recommendation device is further provided, which is applied to a display device and includes:

[0047] A second sending module, configured to send a media asset acquisition request to a server, where the media asset acquisition request includes environmental feature data; the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request;

[0048] A second receiving module, configured to receive target recommended media assets fed back by the server based on the media asset acquisition request; the target recommended media assets are obtained by the server sorting the recalled recommended media assets according to the environmental feature data;

[0049] A control module, configured to control the display to display the target recommended media assets.

[0050] In the above embodiments, by introducing environmental feature data, since the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request, a recommendation basis is provided for subsequent media asset recommendations. By receiving the target recommended media assets fed back by the server based on the media asset acquisition request and controlling the display to display the target recommended media assets, the display of the target recommended media assets is realized. Since the target recommended media assets are obtained by the server sorting the recalled recommended media assets according to the environmental feature data, the target recommended media assets better meet the media asset recommendation requirements of the user in the current application scenario and are conducive to improving the user experience.

[0051] Seventhly, in some embodiments, a computer-readable storage medium is further provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the methods provided in some embodiments of the third or fourth aspect are implemented.

[0052] Eighthly, in some embodiments, a computer program product is further provided, including a computer program. When the computer program is executed by a processor, the steps of the methods provided in some embodiments of the third or fourth aspect are implemented. Description of the Drawings

[0053] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings.

[0054] Figure 1 Schematic diagram of the operation scenario between the display device and the control device provided for some embodiments;

[0055] Figure 2 Schematic diagram of the hardware configuration of the display device provided for some embodiments;

[0056] Figure 3 Schematic diagram of the hardware configuration of the control device provided for some embodiments;

[0057] Figure 4 Schematic diagram of the software configuration of the display device provided for some embodiments;

[0058] Figure 5 Schematic diagram of the process of the media resource recommendation method provided for some embodiments;

[0059] Figure 6 Schematic diagram of the process of the sorting step of the recalled recommended media resources provided for some embodiments;

[0060] Figure 7 Schematic diagram of the process of the selection step of the recalled recommended media resources provided for some embodiments;

[0061] Figure 8 Schematic diagram of the process of the selection step of the recalled recommended media resources provided for other embodiments;

[0062] Figure 9 Schematic diagram of the two-tower recall model provided for some embodiments;

[0063] Figure 10 Processing flowchart of the input data of the two-tower recall model provided for some embodiments;

[0064] Figure 11 Schematic diagram of the screening order of magnitude of the target recommended media resource selection process provided for some embodiments;

[0065] Figure 12 Schematic diagram of the process of the media resource recommendation method provided for other embodiments;

[0066] Figure 13 Schematic diagram of the process of the media resource recommendation method provided for still other embodiments;

[0067] Figure 14Structural block diagram of a media asset recommendation device provided for some embodiments;

[0068] Figure 15 Structural block diagram of a media asset recommendation device provided for other embodiments;

[0069] Figure 16 Internal structure diagram of a computer device provided for some embodiments. Detailed implementation manners

[0070] Embodiments will be described in detail below, and examples thereof are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementation manners described in the following embodiments do not represent all implementation manners consistent with the present application. They are merely examples of systems and methods consistent with some aspects of the present application detailed in the claims.

[0071] It should be noted that the brief description of terms in the present application is only for facilitating the understanding of the following described implementation manners, rather than intending to limit the implementation manners of the present application. Unless otherwise specified, these terms should be understood in their ordinary and common meanings.

[0072] The terms "first", "second", "third", etc. in the specification, claims and the above-mentioned drawings of the present application are used to distinguish similar or homogeneous objects or entities, and do not necessarily mean to limit a specific order or sequence, unless otherwise noted. It should be understood that such terms can be interchanged under appropriate circumstances.

[0073] The terms "include" and "have" and any variations thereof are intended to cover but not exclude inclusion. For example, a product or device including a series of components does not necessarily have to be limited to all the clearly listed components, but may include other components not clearly listed or inherent to these products or devices.

[0074] The term "module" refers to any known or later developed hardware, software, firmware, artificial intelligence, fuzzy logic or a combination of hardware or / and software code that can perform functions related to the element.

[0075] In the embodiments of the present application, the display device 200 generally refers to a device having the capabilities of displaying images and processing data. For example, the display device 200 includes but is not limited to smart TVs, mobile terminals, computers, monitors, advertising screens, wearable devices, virtual reality devices, augmented reality devices, etc.

[0076] Figure 1 Schematic diagram of an operation scenario between a display device and a control device provided for some embodiments of the present application. As Figure 1As shown, the user can operate the display device 200 through touch operations, the mobile terminal 300, and the control device 100. For example, the control device 100 can be a remote control, a stylus, a handle, etc.

[0077] The mobile terminal 300 can be used as a control device to perform human-computer interaction between the user and the display device 200. The mobile terminal 300 can also be used as a communication device to establish a communication connection with the display device 200 for data interaction. In some embodiments, software applications can be installed on the mobile terminal 300 and the display device 200, and the connection communication can be achieved through network communication protocols to achieve the purpose of one-to-one control operations and data communication. It is also possible to transmit the audio and video content displayed on the mobile terminal 300 to the display device 200 to achieve the synchronous display function.

[0078] As Figure 1 Also shown in the figure, the display device 200 also communicates with the server 400 through various communication methods. The display device 200 is allowed to communicate and connect through a local area network (LAN), a wireless local area network (WLAN), and other networks.

[0079] The display device 200 can provide a broadcast receiving television function, and can also additionally provide an intelligent network television function with computer support functions, including but not limited to, network television, smart television, Internet Protocol Television (IPTV), etc.

[0080] In some embodiments, as Figure 1 shown, a media asset acquisition request can be sent to the display device 200 through the control device 100 and the mobile terminal 300. The media asset acquisition request includes environmental characteristic data; the environmental characteristic data is used to characterize the request environment corresponding to the media asset acquisition request. The display device 200 can send the media asset acquisition request to the server 400 through the communication device.

[0081] Figure 2 The hardware configuration block diagram of the display device 200 provided for some embodiments. Figure 1 is shown in the figure.

[0082] In some embodiments, the display device 200 can include at least one of a tuner demodulator 210, a communication device 220, a detector 230, a device interface 240, a controller 250, a display 260, an audio output device 270, a memory, a power supply, and a user input interface.

[0083] In some embodiments, the detector 230 is configured to collect signals of the external environment or for external interaction. For example, the detector 230 includes a light receiver, a sensor for collecting the intensity of ambient light; alternatively, the detector 230 includes an image collector, such as a camera, which can be used to collect external environmental scenes, user attributes or user interaction gestures. Or, the detector 230 includes a sound collector, such as a microphone, etc., for receiving external sounds.

[0084] In some embodiments, the display 260 includes a display function component for presenting a picture and a driving component for driving image display. The display 260 is configured to receive an image signal output from the controller 250 for display. For example, the display 260 can be used to display video content, image content, components of a menu manipulation interface, and a user manipulation UI interface, etc. The display 260 can be used to display target recommended media assets.

[0085] In some embodiments, the communication device 220 is a component for communicating with an external device or the server 400 according to various communication protocol types. The display device 200 can be provided with a plurality of communication devices 220 according to different supported communication methods. For example, when the display device 200 supports wireless network communication, the display device 200 can be provided with a communication device 220 including a WiFi function. When the display device 200 supports Bluetooth connection communication, the display device 200 needs to be provided with a communication device 220 including a Bluetooth function.

[0086] The communication device 220 can enable the display device 200 to communicate with an external device or the server 400 in a wireless or wired connection manner. Among them, the wired connection can connect the display device 200 with an external device through components such as a data cable and an interface. The wireless connection can connect the display device 200 with an external device through a wireless signal or a wireless network. The display device 200 can directly establish a connection relationship with an external device, or can indirectly establish a connection relationship through a gateway, a router, a connection device, etc.

[0087] In some embodiments, the controller 250 may include at least one of a central processing unit, a video processor, an audio processor, a graphics processor, and a power processor, and first interfaces to nth interfaces for input / output. The controller 250 controls the operation of the display device and responds to user operations through various software control programs stored in the memory. The controller 250 controls the overall operation of the display device 200.

[0088] In some embodiments, the controller 250 and the tuner demodulator 210 can be located in different split devices, that is, the tuner demodulator 210 can also be in an external device of the main device where the controller 250 is located, such as an external set-top box, etc.

[0089] In some embodiments, the user may input a user command through a Graphical User Interface (GUI) displayed on the display 260, and the user input interface receives the user input command through the GUI.

[0090] In some embodiments, the audio output device 270 may be the built-in speaker of the display device 200 or an external audio output device connected to the display device 200. Among them, for the external audio output device connected to the display device 200, the display device 200 may also be provided with an external audio output terminal, and the audio output device may be connected to the display device 200 through the external audio output terminal to output the sound of the display device 200.

[0091] In some embodiments, the user input interface 280 can be used to receive instructions from the user input.

[0092] Figure 3 Provided for some embodiments Figure 1 The hardware configuration block diagram of the control device in. As Figure 3 As shown, the control device 100 may include: a controller 110, a communication interface 130, a user input / output interface, a memory, and a power supply.

[0093] The control device 100 is configured to control the display device 200, and can receive the input operation instructions of the user, and convert the operation instructions into instructions recognizable and responsive by the display device 200, playing the role of an interaction intermediary between the user and the display device 200.

[0094] In some embodiments, the control device 100 may be an intelligent device. For example: The control device 100 can install various applications for controlling the display device 200 according to user needs.

[0095] In some embodiments, as Figure 1 As shown, after installing the application for controlling the display device 200, the mobile terminal 300 or other intelligent electronic devices can play a similar function to the control device 100.

[0096] The controller 110 includes a processor 112, a RAM 113, a ROM 114, a communication interface 130, and a communication bus. The controller 110 is used to control the operation and operation of the control device 100, as well as the communication and cooperation between internal components and the data processing functions between the external and internal.

[0097] Under the control of the controller 110, the communication interface 130 realizes the communication of control signals and data signals with the display device 200. The communication interface 130 may include at least one of a WiFi chip 131, a Bluetooth module 132, an NFC module 133, and other near-field communication modules.

[0098] User input / output interface 140, where the input interface includes at least one of a microphone 141, a touchpad 142, a sensor 143, a button 144, and other input interfaces.

[0099] In some embodiments, the control device 100 includes at least one of a communication interface 130 and an input / output interface 140. The communication interface 130 is configured in the control device 100, such as modules like WiFi, Bluetooth, NFC, etc., which can encode user input instructions through the WiFi protocol, or the Bluetooth protocol, or the NFC protocol and send them to the display device 200.

[0100] A memory 190, which is used to store various operating programs, data, and applications for driving and controlling the control device 100 under the control of a controller. The memory 190 can store various control signal instructions input by a user.

[0101] A power supply 180, which is used to provide operating power support for each component of the control device 100 under the control of a controller.

[0102] To perform user interaction, in some embodiments, the display device 200 may run an operating system. The operating system is a computer program for managing and controlling the hardware resources and software resources in the display device 200. The operating system can (control the display device) provide a user interface, allowing a user to interact with the display device 200 and supporting the running of various application programs.

[0103] It should be noted that the operating system can be a native operating system based on a specific operating platform, or a third-party operating system deeply customized based on a specific operating platform, or an independent operating system developed specifically for the display device.

[0104] The operating system can be divided into different modules or layers according to the functions implemented.

[0105] For example, as Figure 4 shown, in some embodiments, the system is divided into four layers, from top to bottom are the application layer (abbreviated as "application layer"), the application framework layer (abbreviated as "framework layer"), the system library layer, and the kernel layer.

[0106] In some embodiments, the application layer is used to provide services and interfaces for applications so that the display device 200 can run the applications and interact with users based on the applications. At least one application can run in the application layer. These applications can be window programs, system setting programs, clock programs, etc. that come with the operating system; they can also be applications developed by third-party developers. In specific implementations, the application packages in the application layer are not limited to the above examples.

[0107] The framework layer provides application programming interfaces (APIs) and programming frameworks for applications. The application framework layer includes some predefined functions. The application framework layer is equivalent to a processing center that determines the actions of the applications in the application layer. Through the API interface, applications can access the resources in the system and obtain the services of the system during execution.

[0108] As Figure 4 shown, in some embodiments, the application framework layer includes a view system, managers, content providers, etc. Among them, the view system can design and implement the interfaces and interactions of applications. The view system includes lists, grids, text boxes, buttons, etc. The managers include at least one of the following modules: The activity manager is used to interact with all the activities running in the system; the location manager is used to provide access to the system location service for system services or applications; the package manager is used to retrieve various information related to the application packages currently installed on the device; the notification manager is used to control the display and clearing of notification messages; the window manager is used to manage the icons, windows, toolbars, wallpapers, and desktop widgets on the user interface.

[0109] In some embodiments, the activity manager is used to manage the life cycles of various applications and the usual navigation back functions, such as controlling the exit, opening, and backward movement of applications. The window manager is used to manage all window programs, such as obtaining the size of the display screen, determining whether there is a status bar, locking the screen, taking screenshots, and controlling the changes of the display window. For example, shrinking the display window, jittering the display, distorting the display, etc.

[0110] In some embodiments, the system runtime library layer can provide support for the framework layer. When the framework layer is used, the operating system will run the instruction libraries included in the system runtime library layer, such as C / C++ instruction libraries, to implement the functions that the framework layer is intended to achieve.

[0111] In some embodiments, the kernel layer is a functional level between the hardware and software of the display device 200. The kernel layer can implement functions such as hardware abstraction, multitasking, and memory management. For example, as Figure 4 shown, hardware drivers can be configured in the kernel layer. The drivers included in the kernel layer can be at least one of the following drivers: audio driver, display driver, Bluetooth driver, camera driver, WIFI driver, USB driver, HDMI driver, sensor drivers (such as fingerprint sensors, temperature sensors, pressure sensors, etc.), and power drivers, etc.

[0112] It should be noted that the above examples are only a simple division of the functions of the operating system and do not limit the specific form of the operating system of the display device 200 in the embodiments. Depending on factors such as the functions of the display device and the type of the operating system, the number of levels and the specific level types included in the operating system can be in other forms.

[0113] Currently, with the rapid development of the multimedia platform, users are faced with a vast amount of media content. Traditional media recommendation methods often rely on statistical models, and by statistically analyzing users' behaviors such as likes and collections, relevant media are recommended to users. In this way, it is difficult to capture users' real needs, and there are problems such as low relevance between the recommended media and users' needs and poor recommendation accuracy.

[0114] In some alternative embodiments, referring to Figure 5 , a media recommendation method is provided, which is applied to the server 400 and includes:

[0115] S510. Receive a media acquisition request sent by the display device 200. The media acquisition request includes environmental feature data, and the environmental feature data is used to characterize the request environment corresponding to the media acquisition request.

[0116] Among them, the media acquisition request can be understood as a request sent by the display device 200 to the server 400 for acquiring media resources. The media resources can include at least one of texts, pictures, videos, and audios, etc.

[0117] Among them, the environmental feature data can be understood as data used to characterize the request environment corresponding to the media asset acquisition request. Exemplarily, the environmental feature data can include at least one of the request time, the location where the media asset acquisition request is sent, the location type of the location, and the weather condition of the location, etc. Among them, the location can include at least one of the location, the address, and the affiliated area, etc. The location type can include at least one of home, subway station, and coffee shop, etc. It should be noted that this embodiment does not make any limitation on the specific environment type and environment content of the request environment.

[0118] Optionally, the environmental feature data can further include the network status data of the display device 200. The network status data can include at least one of network latency, signal strength, and network bandwidth, etc.

[0119] Optionally, the environmental feature data can further include the current behavior data of the user. Among them, the current behavior data can include the current behavior state of the user. The behavior state can include at least one of the bedtime state, the fitness state, and the dining state, etc. This embodiment does not make any limitation on the specific behavior type of the current behavior data. Exemplarily, the current behavior data can be detected by the detector 230 of the display device 200 with the user's authorization.

[0120] S520. In response to the media asset acquisition request, sort the recalled recommended media assets according to the environmental feature data to obtain the target recommended media assets.

[0121] Among them, the recalled recommended media assets can be understood as the media resources preliminarily screened from the candidate recommended media assets. The target recommended media assets can be understood as the media resources after sorting the recalled recommended media assets. Among them, the number of the target recommended media assets can be at least one.

[0122] Exemplarily, the recalled recommended media assets can be sorted in descending order according to the matching degree or similarity between the recalled recommended media assets and the environmental feature data.

[0123] In an optional embodiment, the server 400 can obtain network hot data; according to the network hot data, preliminarily select the recalled recommended media assets from the candidate recommended media assets. Exemplarily, the server 400 can extract the media asset feature data corresponding to the candidate recommended media assets; extract the hot feature data corresponding to the network hot data; select the recalled recommended media assets from the candidate recommended media assets according to the hot feature data and the media asset feature data corresponding to the candidate recommended media assets.

[0124] Among them, the recall method of the recalled recommended media assets can be based on at least one of the traditional recall methods. This embodiment does not make any limitation on the specific recall method of the recalled recommended media assets.

[0125] Optionally, the network hot data may include at least one of hot topic data, hot media asset type data, etc.

[0126] In another optional embodiment, the media asset acquisition request may include user data. Correspondingly, the server 400 may extract user feature data corresponding to the user data; according to the user feature data, recall recommended media assets are selected from the candidate recommended media asset data. Exemplarily, the server 400 may extract media asset feature data corresponding to the candidate recommended media assets; according to the user feature data and the media asset feature data corresponding to the candidate recommended media assets, recall recommended media assets are selected from the candidate recommended media assets.

[0127] Optionally, the user data may include at least one of user historical behavior data, user attribute data, etc. The user historical behavior data may include at least one of historical like data, historical favorite data, historical view data, etc. The user attribute data may include at least one of user age, user gender, user occupation, etc.

[0128] In yet another optional embodiment, the server 400 may fuse the hot feature data and the user feature data to obtain fused feature data; according to the fused feature data and the media asset feature data corresponding to the candidate recommended media assets, recall recommended media assets are selected from the candidate recommended media assets.

[0129] In an optional embodiment, the server 400 may input the environmental feature data and the media asset feature data corresponding to the recall recommended media assets into a trained media asset ranking model to obtain the target recommended media assets. Among them, the media asset ranking model may be a traditional machine learning model, a neural network model, etc., and the model type of the media asset ranking model is not limited in this embodiment.

[0130] In another optional embodiment, the server 400 may determine the association degree between the recall recommended media assets and the environmental feature data; according to the association degree corresponding to the recall recommended media assets, the recall recommended media assets are sorted to obtain the target recommended media assets.

[0131] S530: Send the target recommended media assets to the display device so that the display device displays the target recommended media assets.

[0132] In an optional embodiment, the target recommended media assets and the media asset categories corresponding to the target recommended media assets may be sent to the display device so that the display device classifies and displays the target recommended media assets.

[0133] Optionally, the media asset categories may include at least one of movie type, short video type, music type, etc. The specific classification method of the media asset categories is not limited in this embodiment.

[0134] Exemplarily, take the target recommended media assets A, target recommended media asset B, target recommended media asset C, target recommended media asset D, etc. arranged in order as an example. Among them, target recommended media asset A and target recommended media asset C belong to the first media asset category, and target recommended media asset B and target recommended media asset D belong to the second media asset category. The display device can display target recommended media asset A and target recommended media asset C in order under the classification of the first media asset category, and display target recommended media asset B and target recommended media asset D in order under the classification of the second media asset category.

[0135] In the above embodiments, by introducing environmental feature data, since the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request, a recommendation basis is provided for subsequent media asset recommendations. By responding to the media asset acquisition request, sorting the recalled recommended media assets according to the environmental feature data, and obtaining the target recommended media assets to be displayed on the display device, it is possible to preferentially recommend content related to the environmental feature data to the user from the recalled recommended media assets, which better meets the media asset recommendation requirements of the user in the current application scenario and is conducive to improving the user experience.

[0136] Based on the technical solutions of the above embodiments, an optional embodiment is further provided. In this optional embodiment, the sorting step of the recalled recommended media assets is refined.

[0137] As Figure 6 shown in the flowchart of the sorting step of the recalled recommended media assets, it includes:

[0138] S610. Determine the application scenario category to which the environmental feature data belongs, and the media asset category of the recalled recommended media assets.

[0139] Among them, the application scenario category can be understood as the scenario category determined by the environmental feature data and used to characterize the application scenario where the user is located.

[0140] In an optional embodiment, different reference scenario categories can be preset in advance, and each reference scenario category can be obtained by describing at least one preset environmental feature. Correspondingly, the application scenario category to which the environmental feature data belongs can be determined from the reference scenario categories. Exemplarily, the application scenario category to which the environmental feature data belongs can be queried from the reference scenario categories.

[0141] For the sake of easy understanding, take the reference scenario categories including at least one of reference scenario category x1 and reference scenario category x2, etc. as an example, which should not be understood as a limitation on the specific classification method of the reference scenario categories.

[0142] Among them, the reference scenario category x1 can be described as "on weekends, during the time period from 18:00 to 24:00, the weather is rainy, and the location type is home". Exemplarily, the reference scenario category x2 can be described as "on weekdays, during the time period from 8:00 to 9:00, the weather is sunny, and the location type is subway station". It should be noted that the preset environmental features corresponding to the reference scenario category can be set by technicians according to needs or experience, or determined through a large number of experiments, and this embodiment does not make any limitations on this.

[0143] Among them, the media asset category can be understood as a classification label for classifying the recalled and recommended media assets according to a preset classification standard.

[0144] In an alternative embodiment, different reference media asset categories can be preset in advance, and each reference media asset category can be described by at least one preset media asset feature. Correspondingly, the media asset feature data of the recalled and recommended media assets can be extracted, and the media asset category to which the media asset feature data belongs can be determined from the reference media asset categories.

[0145] Optionally, the media asset category can include at least one of movie type, short video type, music type, etc. This embodiment does not make any limitations on the specific classification method of the media asset category.

[0146] S620. Determine the correlation degree between the media asset category of the recalled and recommended media assets and the application scenario category.

[0147] In an alternative embodiment, a media asset scenario mapping table can be obtained; among them, the media asset scenario mapping table includes the correlation degrees between different media asset categories and the corresponding reference scenario categories; according to the media asset category and the application scenario category of the recalled and recommended media assets, the corresponding correlation degree of the recalled and recommended media assets is found from the media asset scenario mapping table.

[0148] Optionally, the media asset scenario mapping table can be determined according to historical statistical data. The historical statistical data can include the playback volumes of users for each media asset category under different reference scenario categories.

[0149] Exemplarily, the media asset scenario mapping table can be an association matrix of the application scenario category and the media asset category. Among them, in the association matrix M, the matrix element m ij represents the correlation degree between the i-th reference scenario category and the j-th media asset category.

[0150] For example, for historical statistical data, in the application scenario of "weekend, evening, family", the playback volume of movie or variety show videos is often high. Therefore, the relevance of this scenario to movie or variety show media assets is also high. On the contrary, in the application scenario of "weekday, morning rush hour, subway station", the playback volume of movie or variety show videos is often low. Therefore, the relevance of this scenario to movie or variety show media assets is also low. Based on this, the relevance between different media asset categories and corresponding reference categories can be determined in advance according to historical statistical data, so that the relevance corresponding to the recalled and recommended media assets can be found from the media asset scenario mapping table according to the media asset category and application scenario category of the recalled and recommended media assets.

[0151] In another optional embodiment, the media asset category and application scenario category of the recalled and recommended media assets can be input into a pre-trained relevance recognition model, and the relevance corresponding to the recalled and recommended media assets is output. Among them, the relevance recognition model can be a traditional machine learning model or a neural network model, etc. The model type of the relevance recognition model is not limited in this embodiment.

[0152] S630. Sort the recalled and recommended media assets according to the relevance corresponding to the recalled and recommended media assets to obtain the target recommended media assets.

[0153] In an optional embodiment, the recalled and recommended media assets can be sorted according to the order of the relevance corresponding to the recalled and recommended media assets to obtain the target recommended media assets. For example, the recalled and recommended media assets can be sorted in descending order of relevance.

[0154] In another optional embodiment, the media asset acquisition request includes user data. Correspondingly, the media asset feature data corresponding to the recalled and recommended media assets can be extracted, and the user feature data corresponding to the user data can be extracted; the similarity between the media asset feature data corresponding to the recalled and recommended media assets and the user feature data is determined; the recalled and recommended media assets are initially sorted according to the similarity corresponding to the recalled and recommended media assets; and the recalled and recommended media assets after the initial sorting are re-sorted according to the relevance corresponding to the recalled and recommended media assets to obtain the target recommended media assets.

[0155] Exemplarily, the user data can include at least one of user historical behavior data and user attribute data, etc. The user historical behavior data can include at least one of historical like data, historical favorite data, and historical viewing data, etc. The user attribute data can include at least one of user age, user gender, and user occupation, etc.

[0156] Exemplarily, the recalled and recommended media assets can be initially sorted in descending order of the corresponding similarity. Among them, the media asset feature data and user feature data can be input into the similarity recognition model, and the similarity corresponding to the recalled and recommended media assets is output. The similarity recognition model can be a traditional machine learning module or a neural network model, and the specific model category of the similarity recognition model is not limited in this embodiment.

[0157] Exemplarily, the similarity corresponding to the recalled and recommended media assets can be adjusted according to the correlation degree corresponding to the recalled and recommended media assets to obtain the target similarity; the recalled and recommended media assets are re-sorted in descending order of the target similarity corresponding to the recalled and recommended media assets to obtain the target recommended media assets.

[0158] Exemplarily, the target similarity S of the recalled and recommended media assets can be determined according to the following formula new :

[0159] S new = S old * m sc

[0160] Among them, S old represents the similarity corresponding to the recalled and recommended media assets; S new represents the target similarity corresponding to the recalled and recommended media assets; m sc represents the correlation degree corresponding to the recalled and recommended media assets.

[0161] In the above embodiment, by introducing the application scenario category, the application scenario where the user is currently located can be identified according to the environmental feature data, providing a basis for the subsequent accurate matching of media assets. By determining the correlation degree between the media asset category of the recalled and recommended media assets and the application scenario category, and by sorting the recalled and recommended media assets according to the correlation degree corresponding to the recalled and recommended media assets, high-correlation media assets can be focused on, which is beneficial to improving the accuracy and relevance of media recommendations.

[0162] Based on the technical solutions of the above embodiments, an alternative embodiment is also provided. In this alternative embodiment, the step of selecting the recalled and recommended media assets is added. Among them, the media asset acquisition requests user data.

[0163] As Figure 7 shown in the flowchart of the steps for selecting the recalled and recommended media assets, it includes:

[0164] S710. Extract the media asset feature data corresponding to the candidate recommended media assets.

[0165] Among them, candidate recommended media assets can be understood as media resources obtained from the network or network platform for candidate recommendation. The media resources can include at least one of pictures, videos, and audios, etc. The media asset feature data can be understood as the key features extracted from the candidate recommended media assets.

[0166] In an optional embodiment, multi-modal feature data of the media asset can be extracted from the candidate recommended media asset; the multi-modal feature data is mapped to a low-dimensional vector space to obtain a media asset embedding vector I emb ; the media asset embedding vector I emb is used as the media asset feature data.

[0167] Among them, the multi-modal feature data can be understood as feature data of different modal types. Exemplarily, the multi-modal feature data can include at least one of text feature data, picture feature data, video feature data, and audio feature data, etc.

[0168] In another optional embodiment, multi-modal feature data of the media asset can be extracted from the candidate recommended media asset; the multi-modal feature data is mapped to a low-dimensional vector space to obtain a media asset embedding vector I emb ; the environmental feature data is converted into a feature vector form to obtain an environmental embedding vector S emb ; the media asset embedding vector I emb and the environmental embedding vector S emb are interacted to obtain a media asset feature vector I enhanced ; the media asset feature vector I enhanced is used as the media asset feature data. Exemplarily, the candidate recommended media asset can be input into an extraction model to output the multi-modal feature data of the candidate recommended media asset. It should be noted that the specific model type of the extraction model is not limited in this embodiment.

[0169] Optionally, the media asset feature vector I enhanced can be determined according to the following formula:

[0170] I enhanced = I emb * S emb

[0171] Among them, I enhanced represents the media asset feature vector; I emb represents the media asset embedding vector; S emb is the environmental embedding vector.

[0172] S720. Extract user feature data corresponding to the user data.

[0173] Among them, the user data can include at least one of the user historical behavior data and user attribute data mentioned above.

[0174] In an optional embodiment, user data, environmental feature data, and media asset feature data can be preprocessed. Exemplarily, data cleaning can be performed on the user data, and the user's historical behavior data in the user data can be organized into a sequence form; numerical encoding can be performed on the environmental feature data; and the media asset feature data can also be converted into a feature vector form to facilitate reducing the workload of subsequent steps.

[0175] Optionally, the user's historical behavior data can be mapped to a low-dimensional vector space through an embedding layer to obtain a historical behavior embedding vector H emb ; the user attribute data can be mapped to a low-dimensional vector space through an embedding layer to obtain an attribute embedding vector A emb ; the historical behavior embedding vector H emb and the attribute embedding vector A emb are concatenated and fused to obtain user feature data, that is, a user feature vector U emb . Among them, U emb = [H emb ; A emb .

[0176] Optionally, the user's historical behavior data can be mapped to a low-dimensional vector space through an embedding layer to obtain a historical behavior embedding vector H emb ; the user attribute data can be mapped to a low-dimensional vector space through an embedding layer to obtain an attribute embedding vector A emb ; the environmental feature data is converted into a feature vector form to obtain an environmental embedding vector S emb ; the historical behavior embedding vector H emb , the attribute embedding vector A emb and the environmental embedding vector S emb are concatenated and fused to obtain user feature data, that is, a user feature vector U fusion . Among them, U fusion = [H emb ; A emb ; S emb .

[0177] S730. Select recall recommended media assets from the candidate recommended media assets according to the user feature data and the media asset feature data corresponding to the candidate recommended media assets.

[0178] In an optional embodiment, the feature data to be processed can be enhanced in attention based on the environmental feature data to update the feature data to be processed; wherein, the feature data to be processed includes user feature data or media feature data corresponding to candidate recommended media assets; according to the similarity between the media feature data corresponding to the candidate recommended media assets and the user feature data, the recalled recommended media assets are selected from the candidate recommended media assets. By enhancing the attention of the feature data to be processed based on the environmental feature data, the user feature data or the media feature data corresponding to the candidate recommended media assets can focus on the current application scenario, which is beneficial to improving the relevance between the recalled recommended media assets and the current application scenario.

[0179] The following details the attention enhancement steps for the feature data to be processed.

[0180] Optionally, the semantic matching degree between different dimensional features of the feature data to be processed and the environmental feature data can be determined; according to the semantic matching degree corresponding to different dimensional features, the feature weights of the corresponding dimensional features are determined; according to different dimensional features and the corresponding feature weights, the updated feature data to be processed is determined. Among them, the semantic matching degree corresponding to the dimensional feature, that is, the attention score corresponding to the dimensional feature. In the above steps, by introducing the semantic matching degree and quantifying the matching degree between different dimensional features and the environmental features through the semantic matching degree, the feature weights can be dynamically adjusted following the change of the environmental feature data. By determining the updated feature data to be processed according to different dimensional features and the corresponding feature weights, the attention enhancement of the feature data to be processed is realized, enabling the feature data to be processed to more accurately represent the current needs of the user.

[0181] Exemplarily, the environmental query matrix corresponding to the environmental feature data can be determined, and the semantic matching degree between different dimensional features of the feature data to be processed and the environmental query matrix can be determined.

[0182] Exemplarily, the environmental query matrix corresponding to the environmental feature data can be determined according to a multi-layer perceptron. Among them, the environmental query matrix Q can be determined according to the following formula:

[0183] Q = f(S emb )

[0184] where Q represents the environmental query matrix; f represents a multi-layer perceptron (MLP, Multi-Layer Perceptron); S emb represents the environmental embedding vector corresponding to the environmental feature data.

[0185] Exemplarily, the semantic matching degree e corresponding to different dimensional features can be determined by the following formula i :

[0186]

[0187] Among them, X i represents the i-th dimensional feature of the feature data X to be processed; Q represents the environmental query matrix; e i represents the semantic matching degree corresponding to the i-th dimensional feature of the feature data X to be processed.

[0188] Exemplarily, the feature weight α corresponding to different dimensional features can be determined according to the following formula i :

[0189]

[0190] Among them, exp represents the natural exponential function; e i represents the semantic matching degree corresponding to the i-th dimensional feature of the feature data X to be processed; d represents the number of dimensions of the feature data to be processed; α i represents the feature weight corresponding to the i-th dimensional feature of the feature data X to be processed.

[0191] Exemplarily, the feature data v to be processed after attention enhancement can be determined according to the following formula:

[0192]

[0193] Among them, v represents the feature data v to be processed after attention enhancement; X i represents the i-th dimensional feature of the feature data X to be processed; α i represents the feature weight corresponding to the i-th dimensional feature of the feature data X to be processed; d represents the number of dimensions of the feature data to be processed.

[0194] It should be noted that the feature data X to be processed may include the user feature vector U fusion ; correspondingly, the user feature vector U after attention enhancement can be fusion represented as the first enhanced vector v U .

[0195] The feature data X to be processed may include the media asset feature vector I enhanced ; correspondingly, the media asset feature vector I after attention enhancement can be enhanced represented as the second enhanced vector v I .

[0196] The above content is an explanation of the attention enhancement step for the feature data to be processed. The following details the selection step of the recalled recommended media assets.

[0197] In some alternative embodiments, the recalled recommended media assets can be selected from the candidate recommended media assets according to the similarity between the media asset feature data and the user feature data.

[0198] In some other alternative embodiments, recall recommended media assets are selected from candidate recommended media assets according to the similarity between the first enhancement vector v U and media asset feature data. Alternatively, recall recommended media assets can be selected from candidate recommended media assets according to the similarity between user feature data and the second enhancement vector v I .

[0199] In still some other alternative embodiments, recall recommended media assets can be selected from candidate recommended media assets according to the similarity between the first enhancement vector v U and the second enhancement vector v I .

[0200] Optionally, the cosine similarity between the first enhancement vector v U and the second enhancement vector v I can be used as the similarity score of the candidate recommended media assets; recall recommended media assets are selected from the candidate recommended media assets according to the similarity scores of the candidate recommended media assets.

[0201] It can be understood that for each candidate recommended media asset, by determining the cosine similarity between the first enhancement vector v U and the second enhancement vector v I , the similarity score of the corresponding candidate recommended media asset can be obtained. The higher the similarity score of the candidate recommended media asset, the more it indicates that the candidate recommended media asset matches the user's true needs. Based on this, by sorting the candidate recommended media assets in the order of similarity scores, recall recommended media assets can be selected from the candidate recommended media assets.

[0202] Exemplarily, the cosine similarity between the first enhancement vector v U and the second enhancement vector v I can be determined according to the following formula:

[0203]

[0204] where, ‖v U ‖ represents the norm of the first enhancement vector v U ; ‖v I ‖ represents the norm of the second enhancement vector v I ; Sim(v U , v I ) represents the cosine similarity between the first enhancement vector v U and the second enhancement vector v I .

[0205] Exemplarily, the similarity scores of candidate recommended media assets can be used to sort each candidate recommended media asset in descending order of similarity score; from different candidate recommended media assets, the top K candidate recommended media assets with higher similarity scores are selected to obtain the recalled recommended media assets. Here, K represents the preset selection quantity. The preset selection quantity K can be set by technicians according to needs or experience, or determined through a large number of experiments, and the present application does not make any limitation on this.

[0206] In the above embodiments, by selecting the recalled recommended media assets from the candidate recommended media assets according to the user feature data and the media asset feature data corresponding to the candidate recommended media assets, the recalled recommended media assets can better conform to the user features, realizing personalized recommendation for the user. In addition, by enhancing the attention of the to-be-processed feature data based on the environmental feature data, the user feature data or the media asset feature data corresponding to the candidate recommended media assets can focus on the current application scenario, which is beneficial to improving the relevance between the recalled recommended media assets and the current application scenario.

[0207] Based on the technical solutions of the above embodiments, another optional embodiment is provided. In this optional embodiment, the selection steps of the recalled recommended media assets are described in detail.

[0208] As Figure 8 shown in the flowchart of the selection steps of the recalled recommended media assets, it includes:

[0209] S801. Convert the environmental feature data into an environmental embedding vector.

[0210] S802. Map the user historical behavior data and the user attribute data into a low-dimensional vector space respectively to obtain a historical behavior embedding vector and an attribute embedding vector.

[0211] S803. Perform feature fusion on the historical behavior embedding vector, the attribute embedding vector, and the environmental embedding vector to obtain a user feature vector.

[0212] S804. Extract multi-modal feature data from the candidate recommended media assets.

[0213] S805. Map the multi-modal feature data into a low-dimensional vector space to obtain a media asset embedding vector.

[0214] S806. Interact the media asset embedding vector and the environmental embedding vector to obtain a media asset feature vector.

[0215] S807. Based on the environmental embedding vector, enhance the attention of the user feature vector to obtain a first enhanced vector.

[0216] S808. Based on the environmental embedding vector, enhance the attention of the media asset feature vector to obtain a second enhanced vector.

[0217] S809. Determine the similarity between the first enhanced vector and the second enhanced vectors corresponding to different candidate recommended media assets.

[0218] S810. Select the recalled recommended media assets from different candidate recommended media assets according to the similarities corresponding to different candidate recommended media assets.

[0219] Based on the technical solutions of the above embodiments, a two - tower recall model is further provided for selecting recalled recommended media assets from candidate recommended media assets.

[0220] As Figure 9 shown is a schematic diagram of the two - tower recall model. The two - tower recall model includes a user tower 901 and an item tower 902. The following will separately describe the user tower 901 and the item tower 902.

[0221] Continue to refer to Figure 9 shown. In the user tower 901, the user historical behavior data and user attribute data can be respectively mapped into a low - dimensional vector space to obtain a historical behavior embedding vector and an attribute embedding vector; the historical behavior embedding vector, the attribute embedding vector, and the environment embedding vector are subjected to feature fusion to obtain a user feature vector; based on the environment embedding vector, the user feature vector is attentively enhanced to obtain a first enhanced vector.

[0222] Continue to refer to Figure 9 shown. In the item tower 902, multi - modal feature data can be extracted from candidate recommended media assets; the multi - modal feature data is mapped into a low - dimensional vector space to obtain a media asset embedding vector; the media asset embedding vector and the environment embedding vector are interacted to obtain a media asset feature vector; based on the environment embedding vector, the media asset feature vector is attentively enhanced to obtain a second enhanced vector.

[0223] By determining the similarity between the first enhanced vector and the second enhanced vectors corresponding to different candidate recommended media assets, the top K candidate recommended media assets with higher similarity scores can be selected from different candidate recommended media assets to obtain the recalled recommended media assets.

[0224] Refer to Figure 10The figure shows a processing flow chart of the input data of the dual - tower recall model. Optionally, the server 400 can obtain the environmental feature data and user data sent by the client of the display device 200. Exemplarily, it can receive a media asset acquisition request sent by the display device, and the media asset acquisition request includes environmental feature data and user data. Then, the environmental feature data and user data can be pre - processed, and then off - line calculation on the data platform can be performed. Among them, the off - line calculation process on the data platform includes two parts: feature engineering and data samples. For the feature engineering part: user features can be understood as user feature data extracted from user data; item features can be understood as media asset feature data extracted from candidate recommended media assets; environmental features are the environmental feature data; cross - features can be understood as cross - features obtained by the interaction between user features, item features, and environmental features. For the data sample part: mainly, the user further processes the data samples, for example, it can include sample splicing, anomaly filtering, sample sampling, and feature consistency processing, etc. At the same time, Figure 10 The selection process of the target recommended media asset is also shown. By obtaining candidate recommended media assets from the recommendation pool, performing rough ranking and fine ranking on the candidate recommended media assets, recall recommended media assets are obtained, and then the recall recommended media assets are re - ranked to obtain the target recommended media assets. Among them, the recommendation pool can include a network or a network platform.

[0225] Reference Figure 11 The figure shows a schematic diagram of the screening order of magnitude of the target recommended media asset selection process. It is possible to select thousands of candidate recommended media assets from a recommendation pool of millions of levels. By performing rough ranking and fine ranking on the candidate recommended media assets, hundreds of levels of recall recommended media assets are obtained. Finally, the recall recommended media assets of hundreds of levels are re - ranked to obtain the target recommended media assets of tens of levels.

[0226] In some alternative embodiments, referring to Figure 12 , another media asset recommendation method is provided, which is applied to the display device 200 and includes:

[0227] S1201. Send a media asset acquisition request to the server, and the media asset acquisition request includes environmental feature data; the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request.

[0228] S1202. Receive the target recommended media asset fed back by the server based on the media asset acquisition request; the target recommended media asset is obtained by the server sorting the recall recommended media assets according to the environmental feature data.

[0229] S1203. Control the display to display the target recommended media asset.

[0230] In the above embodiments, by introducing environmental feature data, since the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request, a basis for subsequent media asset recommendation is provided. By receiving the target recommended media assets fed back by the server based on the media asset acquisition request and controlling the display device to display the target recommended media assets, the display of the target recommended media assets is realized. Since the target recommended media assets are sorted by the server according to the environmental feature data from the recalled recommended media assets, the target recommended media assets better meet the media asset recommendation requirements of the user in the current application scenario, which is beneficial to improving the user experience.

[0231] Based on the technical solutions of the above embodiments, another optional embodiment is provided. In this optional embodiment, the media asset recommendation method is described in detail.

[0232] As Figure 13 shown in the flowchart of the media asset recommendation method, it includes:

[0233] S1301. The display device sends a media asset acquisition request to the server. The media asset acquisition request includes environmental feature data and user data; the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request.

[0234] S1302. The server extracts the media asset feature data corresponding to the candidate recommended media assets.

[0235] S1303. The server extracts the user feature data corresponding to the user data.

[0236] S1304. The server determines the semantic matching degree between different dimensional features of the to-be-processed feature data and the environmental feature data; the to-be-processed feature data includes user feature data or media asset feature data corresponding to the candidate recommended media assets.

[0237] S1305. The server determines the feature weights of the corresponding dimensional features according to the semantic matching degrees corresponding to different dimensional features.

[0238] S1306. The server determines the updated to-be-processed feature data according to different dimensional features and the corresponding feature weights.

[0239] S1307. The server selects recalled recommended media assets from the candidate recommended media assets according to the similarity between the media asset feature data corresponding to the candidate recommended media assets and the user feature data.

[0240] S1308. The server determines the application scenario category to which the environmental feature data belongs, and the media asset category of the recalled recommended media assets.

[0241] S1309. The server obtains a media asset scenario mapping relationship table; wherein, the media asset scenario mapping relationship table includes the association degrees between different media asset categories and the corresponding reference scenario categories.

[0242] S1310. The server looks up the correlation degree corresponding to the recalled recommended media asset from the media asset scenario mapping table according to the media asset category and application scenario category of the recalled recommended media asset.

[0243] S1311. The server sorts the recalled recommended media assets according to the correlation degree corresponding to the recalled recommended media assets to obtain the target recommended media assets.

[0244] S1312. The server sends the target recommended media assets to the display device.

[0245] S1313. The display device controls the display to display the target recommended media assets.

[0246] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are sequentially shown according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise clearly stated in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same moment, but can be executed at different moments. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0247] Based on the same inventive concept, an embodiment of the present application also provides a media asset recommendation device for implementing the above-mentioned media asset recommendation method. The implementation solution provided by this device to solve the problem is similar to the implementation solution described in the above method. Therefore, the specific limitations in one or more embodiments of the media asset recommendation device provided below can refer to the limitations on the media asset recommendation method in the above text, and will not be repeated here.

[0248] In an exemplary embodiment, as Figure 14 shown, a media asset recommendation device is provided, including: a first receiving module 1410, a processing module 1420, and a first sending module 1430, where:

[0249] The first receiving module 1410 is configured to receive a media asset acquisition request sent by the display device. The media asset acquisition request includes environmental feature data; the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request.

[0250] The processing module 1420 is configured to, in response to the media asset acquisition request, sort the recalled recommended media assets according to the environmental feature data to obtain the target recommended media assets.

[0251] A first sending module 1430, configured to send target recommended media assets to a display device, so that the display device displays the target recommended media assets.

[0252] In one embodiment, the processing module 1420 includes: a first determination unit, configured to determine the application scenario category to which the environmental feature data belongs, and the media asset category for recalling recommended media assets; a second determination unit, configured to determine the association degree between the media asset category for recalling recommended media assets and the application scenario category; and a processing unit, configured to sort the recalled recommended media assets according to the association degree corresponding to the recalled recommended media assets, so as to obtain target recommended media assets.

[0253] In one embodiment, the second determination unit includes: a first acquisition subunit, configured to acquire a media asset scenario mapping relationship table; wherein, the media asset scenario mapping relationship table includes the association degrees between different media asset categories and corresponding reference scenario categories; and a lookup subunit, configured to look up the association degree corresponding to the recalled recommended media assets from the media asset scenario mapping relationship table according to the media asset category and the application scenario category of the recalled recommended media assets.

[0254] In one embodiment, the processing unit includes: a first extraction subunit, configured to extract media asset feature data corresponding to the recalled recommended media assets; a second extraction subunit, configured to extract user feature data corresponding to user data; a first determination subunit, configured to determine the similarity between the media asset feature data corresponding to the recalled recommended media assets and the user feature data; a first processing subunit, configured to initially sort the recalled recommended media assets according to the similarity corresponding to the recalled recommended media assets; and a second processing subunit, configured to re-sort the initially sorted recalled recommended media assets according to the association degree corresponding to the recalled recommended media assets, so as to obtain target recommended media assets.

[0255] In one embodiment, it further includes: a first extraction module, configured to extract media asset feature data corresponding to candidate recommended media assets; a second extraction module, configured to extract user feature data corresponding to user data; and a selection module, configured to select recalled recommended media assets from the candidate recommended media assets according to the user feature data and the media asset feature data corresponding to the candidate recommended media assets.

[0256] In one embodiment, the selection module includes: an attention enhancement unit, configured to enhance the attention of the to-be-processed feature data based on the environmental feature data, so as to update the to-be-processed feature data; wherein, the to-be-processed feature data includes user feature data or media asset feature data corresponding to candidate recommended media assets; and a selection unit, configured to select recalled recommended media assets from the candidate recommended media assets according to the similarity between the media asset feature data corresponding to the candidate recommended media assets and the user feature data.

[0257] In one embodiment, the attention enhancement unit includes: a second determination subunit configured to determine the semantic matching degree between the different-dimensional features of the feature data to be processed and the environmental feature data; a third determination subunit configured to determine the feature weights of the corresponding dimensional features according to the semantic matching degrees corresponding to the different-dimensional features; and a fourth determination subunit configured to determine the updated feature data to be processed according to the different-dimensional features and the corresponding feature weights.

[0258] In an exemplary embodiment, as Figure 15 shown, a media asset recommendation device is further provided, including: a second sending module 1510, a second receiving module 1520, and a control module 1530, where:

[0259] The second sending module 1510 is configured to send a media asset acquisition request to the server, where the media asset acquisition request includes environmental feature data; the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request;

[0260] The second receiving module 1520 is configured to receive the target recommended media assets fed back by the server based on the media asset acquisition request; the target recommended media assets are obtained by the server sorting the recalled recommended media assets according to the environmental feature data;

[0261] The control module 1530 is configured to control the display to display the target recommended media assets.

[0262] Each module in the above media asset recommendation device can be implemented in whole or in part by software, hardware, and their combination. Each of the above modules can be embedded in or independent of a processor in a computer device in the form of hardware, or stored in a memory in the computer device in the form of software, so as to facilitate the processor to call and execute the operations corresponding to the above respective modules.

[0263] In an exemplary embodiment, a computer device is provided. The computer device can be a server, and its internal structure diagram can be as Figure 16As shown in the figure. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store candidate recommended media data. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a media recommendation method.

[0264] Those skilled in the art can understand that Figure 16 the structure shown in the figure is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have a different component layout.

[0265] In one embodiment, a computer device is further provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0266] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0267] In one embodiment, a computer program product is provided, including a computer program. When the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0268] 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, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0269] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in this application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in this application can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logics based on quantum computing, artificial intelligence (AI) processors, etc., without limitation.

[0270] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this application.

[0271] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the scope of the patent of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all fall within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A server, characterized in that, Including: A communication device configured to communicate with a display device; And at least one processor connected to the communication device and configured to: Receive a media asset acquisition request sent by the display device, where the media asset acquisition request includes environmental feature data; the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request; In response to the media asset acquisition request, sort the recalled recommended media assets according to the environmental feature data to obtain target recommended media assets; Send the target recommended media assets to the display device so that the display device displays the target recommended media assets.

2. The server according to claim 1, wherein When the processor executes sorting the recalled recommended media assets according to the environmental feature data to obtain target recommended media assets, it is configured to: Determine the application scenario category to which the environmental feature data belongs, and the media asset category of the recalled recommended media assets; Determine the correlation degree between the media asset category of the recalled recommended media assets and the application scenario category; Sort the recalled recommended media assets according to the correlation degree corresponding to the recalled recommended media assets to obtain the target recommended media assets.

3. The server according to claim 2, wherein When the processor executes determining the correlation degree between the media asset category of the recalled recommended media assets and the application scenario category, it is configured to: Obtain a media asset scenario mapping relationship table; where the media asset scenario mapping relationship table includes the correlation degrees between different media asset categories and corresponding reference scenario categories; According to the media asset category of the recalled recommended media assets and the application scenario category, look up the correlation degree corresponding to the recalled recommended media assets in the media asset scenario mapping relationship table.

4. The server according to claim 2, wherein The media asset acquisition request includes user data; when the processor executes sorting the recalled recommended media assets according to the correlation degree corresponding to the recalled recommended media assets to obtain the target recommended media assets, it is configured to: Extract the media asset feature data corresponding to the recalled recommended media assets, and Extract the user feature data corresponding to the user data; Determine the similarity between the media asset feature data corresponding to the recalled recommended media assets and the user feature data; Perform an initial sort on the recalled recommended media assets according to the similarity corresponding to the recalled recommended media assets; Re-sort the initially sorted recalled recommended media assets according to the correlation degree corresponding to the recalled recommended media assets to obtain the target recommended media assets.

5. The server according to any one of claims 1-4, characterized in that, The media asset acquisition request further includes user data; the processor is further configured to: Extract the media asset feature data corresponding to the candidate recommended media assets, and Extract the user feature data corresponding to the user data; Select the recalled recommended media assets from the candidate recommended media assets according to the user feature data and the media asset feature data corresponding to the candidate recommended media assets.

6. The server according to claim 5, wherein When the processor executes selecting the recalled recommended media assets from the candidate recommended media assets according to the user feature data and the media asset feature data corresponding to the candidate recommended media assets, it is configured to: Based on the environmental feature data, perform attention enhancement on the to-be-processed feature data to update the to-be-processed feature data; where the to-be-processed feature data includes the user feature data or the media asset feature data corresponding to the candidate recommended media assets; Select the recalled recommended media assets from the candidate recommended media assets according to the similarity between the media asset feature data corresponding to the candidate recommended media assets and the user feature data.

7. The server according to claim 6, wherein When the processor performs attention enhancement on the feature data to be processed, it is configured to: Determine the semantic matching degree between the feature of different dimensions of the feature data to be processed and the environmental feature data; Determine the feature weights of the corresponding dimension features according to the semantic matching degrees corresponding to the different dimension features; Determine the updated feature data to be processed according to the different dimension features and the corresponding feature weights.

8. A display device, characterized in that, Including: A display configured to perform content display; At least one controller communicatively connected to the display and configured to: Send a media asset acquisition request to the server, the media asset acquisition request including environmental feature data; the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request; Receive the target recommended media assets fed back by the server based on the media asset acquisition request; the target recommended media assets are obtained by the server sorting the recalled recommended media assets according to the environmental feature data; Control the display to display the target recommended media assets.

9. A media asset recommendation method, characterized in that The method is applied to a server, and the method includes: Receive a media asset acquisition request sent by a display device, the media asset acquisition request including environmental feature data; the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request; In response to the media asset acquisition request, sort the recalled recommended media assets according to the environmental feature data to obtain target recommended media assets; Send the target recommended media assets to the display device so that the display device displays the target recommended media assets.

10. A media asset recommendation method, characterized in that, The method is applied to a display device, and the method includes: Send a media asset acquisition request to the server, the media asset acquisition request including environmental feature data; the environmental feature data is used to characterize the request environment corresponding to the media asset acquisition request; Receive the target recommended media assets fed back by the server based on the media asset acquisition request; the target recommended media assets are sorted by the server from the recalled recommended media assets according to the environmental feature data; Control the display to display the target recommended media assets.

Images