Server, display device and strategy determination method

Abstract

The invention relates to a server, a display device and a policy determination method. Comprising the following steps: weighting and averaging a first index evaluation value associated with a first user identifier in a user identifier set and a first target weight to obtain a first comprehensive evaluation value, weighting and averaging a second index evaluation value associated with a second user identifier in the user identifier set and a second target weight to obtain a second comprehensive evaluation value, taking a difference value between the first comprehensive evaluation value and the second comprehensive evaluation value as an effect evaluation value of the target content pushing strategy; and obtaining the basic matching degree of the target content pushing strategy and the target user, the comprehensive effect evaluation value and the basic matching degree, selecting a target strategy identifier from the strategy identifier set, and determining the content pushing strategy corresponding to the target strategy identifier as the target content pushing strategy matched with the target user. The content pushing efficiency can be improved.

Description

Technical Field

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

[0002] At present, with the development of Internet technology, there are more and more scenarios for pushing content to users. For example, short videos or advertisements can be pushed to users in short video applications, and media or advertisements can be pushed to users in media playback applications.

[0003] In related technologies, to improve the accuracy of content pushed to users, it is common to filter suitable content push strategies for each user and then push content based on these strategies. This selection of appropriate content push strategies typically involves defining user stratification and evaluating these strategies using A / B experiments.

[0004] However, since AB experiments require AA fairness, it is usually necessary to find two groups of user groups that meet AA fairness before the AB experiment. AA fairness requires that the distribution of user characteristics of the two groups of user groups is consistent. Since finding two groups of user groups that meet AA fairness is time-consuming and labor-intensive, the cycle of screening content push strategies is long, which is not conducive to rapid content push and leads to low content push efficiency. Summary of the Invention

[0005] The present application provides a server, a display device, a policy determination method, and a media resource display method to solve the problem of low content push efficiency.

[0006] In the first aspect, some embodiments further provide a server comprising: a communication device configured to be communicatively connected to a display device; and at least one processor connected to the communication device and configured to: obtain a set of user identifiers, weight and average a first indicator evaluation value and a first target weight associated with a first user identifier in the user identifier set to obtain a first comprehensive evaluation value, wherein the first indicator evaluation value is an evaluation value of the user represented by the first user identifier under a preset indicator after the target content push strategy is set, and the preset indicator is used to reflect the user's interest in the target pushed content, and the target pushed content refers to the content pushed by the target content push strategy; weight and average a second indicator evaluation value and a second target weight associated with a second user identifier in the user identifier set to obtain a second comprehensive evaluation value, wherein the second indicator evaluation value is an evaluation value of the user represented by the second user identifier under a preset indicator after the target content push strategy is set. In the case of a strategy, the evaluation value under the preset indicators, the first target weight and the second target weight are used to balance the distribution of user features associated with the first user identifier and the distribution of user features associated with the second user identifier; the difference between the first comprehensive evaluation value and the second comprehensive evaluation value is used as the effect evaluation value of the target content push strategy, and the effect evaluation value is used to reflect the overall interest of the target user group in the target push content, and the target user group refers to the user group represented by the user identifier set; the basic matching degree between the target content push strategy and the target user is obtained, and the target matching degree between the target content push strategy and the target user is determined according to the effect evaluation value and the basic matching degree, and the target strategy identifier is selected from the strategy identifier set according to the size of the target matching degree, and the content push strategy corresponding to the target strategy identifier is determined as the target content push strategy that matches the target user.

[0007] In this embodiment, since the first target weight and the second target weight are used to balance the distribution of user features associated with the first user identifier and the distribution of user features associated with the second user identifier, the first indicator evaluation value is the evaluation value of the user represented by the first user identifier under the preset indicator after the target content push strategy is set, and the second indicator evaluation value is the evaluation value of the user represented by the second user identifier under the preset indicator when the target content push strategy is not set. Therefore, for the target content push strategy, the user represented by the first user identifier is the user in the experimental group, and the user represented by the second user identifier is the user in the control group. The first indicator evaluation value and the first target weight associated with the first user identifier in the user identifier set are weighted and averaged to obtain a first comprehensive evaluation value, and the second indicator evaluation value and the second target weight associated with the second user identifier in the user identifier set are weighted and averaged to obtain a second comprehensive evaluation value. Since the preset indicator is used to reflect the user's interest in the target push content, and the target push content refers to the content pushed by the target content push strategy, when the experimental group and the control group do not meet AA fairness, the first comprehensive evaluation value can reflect the user's interest in the target push content. The feedback of the experimental group when the experimental group and the control group meet AA fairness, and the second comprehensive evaluation value can reflect the feedback of the control group when the experimental group and the control group meet AA fairness. The difference between the first comprehensive evaluation value and the second comprehensive evaluation value is used as the effect evaluation value of the target content push strategy, which can make the effect evaluation value reflect the effect represented by the target content push strategy when the experimental group and the control group meet AA fairness, that is, the effect evaluation value can reflect the overall interest of the target user group in the target push content. The target user group refers to the user group represented by the user identifier set. Further obtain the basic matching degree between the target content push strategy and the target user, the comprehensive effect evaluation value and the basic matching degree, and obtain the target matching degree between the target content push strategy and the target user. According to the size of the target matching degree, the target policy identifier is selected from the policy identifier set, and the target policy identifier is matched with the content push strategy to be determined as the target content push strategy that matches the target user. In this way, the target content push strategy that matches the target user can be quickly determined without requiring the experimental group and the control group to meet AA fairness, which helps to improve the efficiency of content push.

[0008] In a first aspect, some embodiments provide a display device, comprising: a display and a controller; the controller is configured to: determine an interface identifier in response to an interface display operation, and send a data acquisition request to a server, wherein the data acquisition request carries the interface identifier and a user identifier of a target user; receive interface data returned by the server in response to the data acquisition request, wherein the interface data includes target media data, and the target media data is media data associated with a target content push strategy that matches the target user; control the display to display an interface corresponding to the interface identifier, and display the media data in the interface; wherein the step of determining the target content push strategy comprises: obtaining a user identifier set, weighting and averaging a first indicator evaluation value and a first target weight associated with a first user identifier in the user identifier set to obtain a first comprehensive evaluation value, wherein the first indicator evaluation value is an evaluation value of the user represented by the first user identifier under a preset indicator after the target content push strategy is set, and the preset indicator is used to reflect the user's interest in the target pushed content, and the target pushed content refers to the content pushed by the target content push strategy; weighting and averaging a second indicator evaluation value and a first target weight associated with a second user identifier in the user identifier set to obtain a first comprehensive evaluation value. The two target weights are weighted and averaged to obtain a second comprehensive evaluation value, wherein the second indicator evaluation value is the evaluation value of the user represented by the second user identifier under the preset indicator when the target content push strategy is not set, and the first target weight and the second target weight are used to balance the distribution of user characteristics associated with the first user identifier and the distribution of user characteristics associated with the second user identifier; the difference between the first comprehensive evaluation value and the second comprehensive evaluation value is used as the effect evaluation value of the target content push strategy, and the effect evaluation value is used to reflect the overall interest of the target user group in the target push content, and the target user group refers to the user group represented by the user identifier set; the basic matching degree between the target content push strategy and the target user is obtained; the target matching degree between the target content push strategy and the target user is obtained by combining the effect evaluation value and the basic matching degree, and a target policy identifier is selected from the policy identifier set according to the size of the target matching degree, and the content push strategy corresponding to the target policy identifier is determined as the target content push strategy matching the target user, and when requesting media resource data, the content data associated with the target content push strategy is returned, thereby improving the efficiency of content push.

[0009] In this embodiment, since the first target weight and the second target weight are used to balance the distribution of user features associated with the first user identifier and the distribution of user features associated with the second user identifier, the first indicator evaluation value is the evaluation value of the user represented by the first user identifier under the preset indicator after the target content push strategy is set, and the second indicator evaluation value is the evaluation value of the user represented by the second user identifier under the preset indicator when the target content push strategy is not set. Therefore, for the target content push strategy, the user represented by the first user identifier is the user in the experimental group, and the user represented by the second user identifier is the user in the control group. The first indicator evaluation value and the first target weight associated with the first user identifier in the user identifier set are weighted and averaged to obtain a first comprehensive evaluation value, and the second indicator evaluation value and the second target weight associated with the second user identifier in the user identifier set are weighted and averaged to obtain a second comprehensive evaluation value. Since the preset indicator is used to reflect the user's interest in the target push content, and the target push content refers to the content pushed by the target content push strategy, when the experimental group and the control group do not meet AA fairness, the first comprehensive evaluation value can reflect the user's interest in the target push content. The feedback of the experimental group when the experimental group and the control group meet AA fairness, and the second comprehensive evaluation value can reflect the feedback of the control group when the experimental group and the control group meet AA fairness. The difference between the first comprehensive evaluation value and the second comprehensive evaluation value is used as the effect evaluation value of the target content push strategy, which can make the effect evaluation value reflect the effect represented by the target content push strategy when the experimental group and the control group meet AA fairness, that is, the effect evaluation value can reflect the overall interest of the target user group in the target push content. The target user group refers to the user group represented by the user identifier set. Further obtain the basic matching degree between the target content push strategy and the target user, the comprehensive effect evaluation value and the basic matching degree, and obtain the target matching degree between the target content push strategy and the target user. According to the size of the target matching degree, the target policy identifier is selected from the policy identifier set, and the target policy identifier is matched with the content push strategy to be determined as the target content push strategy that matches the target user. In this way, the target content push strategy that matches the target user can be quickly determined without requiring the experimental group and the control group to meet AA fairness, which helps to improve the efficiency of content push.

[0010] In the third aspect, some embodiments also provide a policy determination method, which is applied to the server provided in the first aspect, and the server includes a storage module, a communication module, and a processor. The method includes: obtaining a user identification set, weighting and averaging the first indicator evaluation value and the first target weight associated with the first user identification in the user identification set to obtain a first comprehensive evaluation value, wherein the first indicator evaluation value is the evaluation value of the user represented by the first user identification under the preset indicator after the target content push policy is set, and the preset indicator is used to reflect the user's interest in the target push content, and the target push content refers to the content pushed by the target content push policy; weighting and averaging the second indicator evaluation value and the second target weight associated with the second user identification in the user identification set to obtain a second comprehensive evaluation value, wherein the second indicator evaluation value is the evaluation value of the user represented by the second user identification under the preset indicator when the target content push policy is not set, and the first target weight and the The second target weight is used to balance the distribution of user features associated with the first user identifier and the distribution of user features associated with the second user identifier; the difference between the first comprehensive evaluation value and the second comprehensive evaluation value is used as the effect evaluation value of the target content push strategy, and the effect evaluation value is used to reflect the overall interest of the target user group in the target push content, and the target user group refers to the user group represented by the user identifier set; the basic matching degree between the target content push strategy and the target user is obtained; the target matching degree between the target content push strategy and the target user is obtained by combining the effect evaluation value and the basic matching degree, and a target policy identifier is selected from the policy identifier set according to the size of the target matching degree, and the content push strategy corresponding to the target policy identifier is determined as the target content push strategy that matches the target user.

[0011] In this embodiment, since the first target weight and the second target weight are used to balance the distribution of user features associated with the first user identifier and the distribution of user features associated with the second user identifier, the first indicator evaluation value is the evaluation value of the user represented by the first user identifier under the preset indicator after the target content push strategy is set, and the second indicator evaluation value is the evaluation value of the user represented by the second user identifier under the preset indicator when the target content push strategy is not set. Therefore, for the target content push strategy, the user represented by the first user identifier is the user in the experimental group, and the user represented by the second user identifier is the user in the control group. The first indicator evaluation value and the first target weight associated with the first user identifier in the user identifier set are weighted and averaged to obtain a first comprehensive evaluation value, and the second indicator evaluation value and the second target weight associated with the second user identifier in the user identifier set are weighted and averaged to obtain a second comprehensive evaluation value. Since the preset indicator is used to reflect the user's interest in the target push content, and the target push content refers to the content pushed by the target content push strategy, when the experimental group and the control group do not meet AA fairness, the first comprehensive evaluation value can reflect the user's interest in the target push content. The feedback of the experimental group when the experimental group and the control group meet AA fairness, and the second comprehensive evaluation value can reflect the feedback of the control group when the experimental group and the control group meet AA fairness. The difference between the first comprehensive evaluation value and the second comprehensive evaluation value is used as the effect evaluation value of the target content push strategy, which can make the effect evaluation value reflect the effect represented by the target content push strategy when the experimental group and the control group meet AA fairness, that is, the effect evaluation value can reflect the overall interest of the target user group in the target push content. The target user group refers to the user group represented by the user identifier set. Further obtain the basic matching degree between the target content push strategy and the target user, the comprehensive effect evaluation value and the basic matching degree, and obtain the target matching degree between the target content push strategy and the target user. According to the size of the target matching degree, the target policy identifier is selected from the policy identifier set, and the target policy identifier is matched with the content push strategy to be determined as the target content push strategy that matches the target user. In this way, the target content push strategy that matches the target user can be quickly determined without requiring the experimental group and the control group to meet AA fairness, which helps to improve the efficiency of content push.

[0012] In a fourth aspect, some embodiments further provide a media asset display method, which is applied to the display device provided in the second aspect, the display device comprising: a display, a communication device, and a controller, the method comprising: in response to an interface display operation, determining an interface identifier, sending a data acquisition request to a server, the data acquisition request carrying the interface identifier and the user identifier of the target user; receiving interface data returned by the server in response to the data acquisition request, the interface data comprising target media asset data, the target media asset data being media asset data associated with a target content push strategy matching the target user; displaying an interface corresponding to the interface identifier, and displaying the media asset data in the interface; wherein the step of determining the target content push strategy comprises: obtaining a user identifier set, weighting and averaging a first indicator evaluation value and a first target weight associated with a first user identifier in the user identifier set to obtain a first comprehensive evaluation value, wherein the first indicator evaluation value is an evaluation value of the user represented by the first user identifier under a preset indicator after the target content push strategy is set, the preset indicator being used to reflect the user's interest in the target pushed content, and the target pushed content refers to the content pushed by the target content push strategy. ; Weighted and averaged the second indicator evaluation value and the second target weight associated with the second user identifier in the user identifier set to obtain a second comprehensive evaluation value, wherein the second indicator evaluation value is the evaluation value of the user represented by the second user identifier under the preset indicator when the target content push strategy is not set, and the first target weight and the second target weight are used to balance the distribution of user characteristics associated with the first user identifier and the distribution of user characteristics associated with the second user identifier; The difference between the first comprehensive evaluation value and the second comprehensive evaluation value is used as the effect evaluation value of the target content push strategy, and the effect evaluation value is used to reflect the overall interest of the target user group in the target push content, and the target user group refers to the user group represented by the user identifier set; Obtain the basic matching degree between the target content push strategy and the target user; Combining the effect evaluation value and the basic matching degree, obtain the target matching degree between the target content push strategy and the target user, select a target policy identifier from the policy identifier set according to the size of the target matching degree, and determine the content push strategy corresponding to the target policy identifier as the target content push strategy matching the target user.

[0013] In this embodiment, since the first target weight and the second target weight are used to balance the distribution of user characteristics associated with the first user identifier and the distribution of user characteristics associated with the second user identifier, the first indicator evaluation value is the evaluation value of the user represented by the first user identifier under the preset indicator after the target content push strategy is set, and the second indicator evaluation value is the evaluation value of the user represented by the second user identifier under the preset indicator when the target content push strategy is not set. Therefore, for the target content push strategy, the user represented by the first user identifier is the user in the experimental group, and the user represented by the second user identifier is the user in the control group. The first indicator evaluation value and the first target weight associated with the first user identifier in the user identifier set are weighted and averaged to obtain a first comprehensive evaluation value, and the second indicator evaluation value and the second target weight associated with the second user identifier in the user identifier set are weighted and averaged to obtain a second comprehensive evaluation value. Since the preset indicator is used to reflect the user's interest in the target pushed content, and the target pushed content refers to the content pushed by the target content push strategy, when the experimental group and the control group do not meet AA fairness, the first comprehensive evaluation value can reflect the situation where the experimental group and the control group meet AA fairness. The feedback of the experimental group under the condition that the second comprehensive evaluation value can reflect the feedback of the control group when the experimental group and the control group meet AA fairness. The difference between the first comprehensive evaluation value and the second comprehensive evaluation value is used as the effect evaluation value of the target content push strategy, which can make the effect evaluation value reflect the effect represented by the target content push strategy when the experimental group and the control group meet AA fairness, that is, the effect evaluation value can reflect the overall interest of the target user group in the target push content. The target user group refers to the user group represented by the user identifier set. The basic matching degree between the target content push strategy and the target user is further obtained, the comprehensive effect evaluation value and the basic matching degree are used to obtain the target matching degree between the target content push strategy and the target user. According to the size of the target matching degree, the target policy identifier is selected from the policy identifier set, and the target policy identifier is matched with the content push strategy to determine it as the target content push strategy that matches the target user. In this way, the target content push strategy that matches the target user can be quickly determined without requiring the experimental group and the control group to meet AA fairness, and the content data associated with the target content push strategy can be returned when requesting media resource data, thereby improving the efficiency of content push. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0015] Figure 1 A schematic diagram of an operation scenario between a display device and a control device provided in some embodiments of the present application;

[0016] Figure 2 A schematic diagram of the hardware configuration of a display device provided in some embodiments of the present application;

[0017] Figure 3 A schematic diagram of the hardware configuration of a control device provided in some embodiments of the present application;

[0018] Figure 4 A schematic diagram of software configuration of a display device provided in some embodiments of the present application;

[0019] Figure 5 A flowchart of a strategy determination method provided in some embodiments of the present application;

[0020] Figure 6 A schematic diagram of a process for determining a first target weight and a second target weight provided in some embodiments of the present application;

[0021] Figure 7 A schematic diagram of a process for determining a basic matching degree provided in some embodiments of the present application;

[0022] Figure 8 A schematic diagram of a process for determining the degree of feature matching in some embodiments of the present application;

[0023] Figure 9 This is a schematic diagram showing the interaction between a device and a server in some embodiments of the present application;

[0024] Figure 10 This is a timing diagram of the media resource display method in some embodiments of the present application. DETAILED DESCRIPTION

[0025] The following embodiments are described in detail, with examples illustrated in the accompanying drawings. When the following description refers to the drawings, identical numbers in different figures represent identical or similar elements unless otherwise indicated. The embodiments described in the following embodiments are not intended to represent all possible implementations consistent with the present application. They are merely examples of systems and methods consistent with certain aspects of the present application, as detailed in the claims.

[0026] It should be noted that the brief descriptions of terms in this application are only for the purpose of facilitating the understanding of the embodiments described below, and are not intended to limit the embodiments of this application. Unless otherwise specified, these terms should be understood according to their ordinary and usual meanings.

[0027] In the specification and claims of this application and the accompanying drawings, the terms "first," "second," "third," etc. are used to distinguish similar or similar objects or entities, and are not necessarily intended to limit a particular order or sequence, unless otherwise noted. It should be understood that the terms used in this manner are interchangeable under appropriate circumstances.

[0028] The terms "comprise," "include," and "have," and any variations thereof, are intended to cover but not exclude inclusion; for example, a product or device comprising a list of components is not necessarily limited to all the components expressly listed but may include other components not expressly listed or inherent to such product or device.

[0029] The term "module" refers to any known or later developed hardware, software, firmware, artificial intelligence, fuzzy logic, or combination of hardware and / or software code that is capable of performing the functionality associated with that element.

[0030] In the embodiments of the present application, the display device 200 generally refers to a device capable of displaying images and processing data. For example, the display device 200 includes but is not limited to a smart TV, a mobile terminal, a computer, a monitor, an advertising screen, a wearable device, a virtual reality device, an augmented reality device, etc.

[0031] Figure 1 This is a schematic diagram of an operation scenario between a display device and a control device provided in some embodiments of the present application. Figure 1 As shown in FIG, a user can operate the display device 200 through touch operation, the mobile terminal 300 and the control device 100. For example, the control device 100 can be a remote controller, a stylus pen, a handle, etc.

[0032] The mobile terminal 300 can function as a control device for performing human-computer interaction between a user and the display device 200. The mobile terminal 300 can also function as a communication device for establishing a communication connection with the display device 200 and exchanging data. In some embodiments, the mobile terminal 300 can install software applications with the display device 200, enabling connection and communication via a network communication protocol, enabling one-to-one control operations and data communication. Audio and video content displayed on the mobile terminal 300 can also be transmitted to the display device 200 for synchronized display.

[0033] like Figure 1 As shown in FIG, the display device 200 also communicates data with the server 400 through various communication methods. The display device 200 may be allowed to communicate via a local area network (LAN), a wireless local area network (WLAN), and other networks.

[0034] The display device 200 may provide a broadcast receiving television function, and may also additionally provide an intelligent network television function with a computer support function, including but not limited to network television, smart TV, Internet Protocol television (IPTV), etc.

[0035] Figure 2 Some embodiments of this application provide Figure 1 2 is a block diagram of the hardware configuration of the display device 200.

[0036] In some embodiments, the display device 200 may include at least one of a tuner 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.

[0037] In some embodiments, detector 230 is used to collect signals from the external environment or external interactions. For example, detector 230 may include a light receiver, such as a sensor for collecting ambient light intensity; or an image collector, such as a camera, for collecting external environmental scenes, user attributes, or user interaction gestures; or a sound collector, such as a microphone, for receiving external sounds.

[0038] In some embodiments, the display 260 includes a display component for presenting images and a driver component for driving image display. The display 260 is configured to receive image signals output from the controller 250 for display. For example, the display 260 can be used to display video content, image content, menu control interface components, and user control UI interfaces.

[0039] In some embodiments, the communication device 220 is a component used to communicate with an external device or server 400 according to various communication protocol types. The display device 200 can be provided with multiple communication devices 220 depending on the supported communication methods. For example, if the display device 200 supports wireless network communication, the display device 200 can be provided with a communication device 220 including WiFi functionality. If the display device 200 supports Bluetooth connection communication, the display device 200 needs to be provided with a communication device 220 including Bluetooth functionality.

[0040] The communication device 220 can establish a communication connection between the display device 200 and an external device or server 400 via a wireless or wired connection. A wired connection can connect the display device 200 to an external device via a data cable, an interface, or other components. A wireless connection can connect the display device 200 to an external device via a wireless signal or wireless network. The display device 200 can establish a connection with an external device directly or indirectly through a gateway, router, or connection device.

[0041] In some embodiments, the controller 250 may include at least one of a central processing unit (CPU), a video processor, an audio processor, a graphics processor, and a power processor, and first 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 a memory. The controller 250 controls the overall operation of the display device 200.

[0042] In some embodiments, the controller 250 and the tuner 210 may be located in different separate devices, that is, the tuner 210 may also be located in an external device of the main device where the controller 250 is located, such as an external set-top box.

[0043] 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 graphical user interface (GUI).

[0044] In some embodiments, the audio output device 270 may be a local speaker of the display device 200, or an external audio output device connected to the display device 200. For the external audio output device connected to the display device 200, the display device 200 may further be provided with an external audio output terminal, through which the audio output device may be connected to the display device 200 to output the sound of the display device 200.

[0045] In some embodiments, the user input interface 280 may be configured to receive instructions from a user.

[0046] Figure 3 Some embodiments of this application provide Figure 1 The hardware configuration diagram of the control device in the figure is as follows. 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.

[0047] The control device 100 is configured to control the display device 200 , and can receive user input operation instructions, and convert the operation instructions into instructions that the display device 200 can recognize and respond to, playing the role of an interactive intermediary between the user and the display device 200 .

[0048] In some embodiments, the control device 100 may be a smart device. For example, the control device 100 may be installed with various applications for controlling the display device 200 according to user needs.

[0049] In some embodiments, as Figure 1As shown, the mobile terminal 300 or other intelligent electronic devices can play a similar function as the control device 100 after installing the application for controlling the display device 200 .

[0050] The controller 110 includes a processor 112, RAM 113, 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 external and internal data processing functions.

[0051] Under the control of the controller 110, the communication interface 130 communicates 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, or other near field communication modules.

[0052] The user input / output interface 140 includes at least one of a microphone 141 , a touch panel 142 , a sensor 143 , a button 144 and other input interfaces.

[0053] In some embodiments, the control device 100 includes at least one of a communication interface 130 and an input / output interface 140. The control device 100 is configured with the communication interface 130, such as a WiFi, Bluetooth, or NFC module, to encode user input commands via the WiFi protocol, Bluetooth protocol, or NFC protocol and transmit them to the display device 200.

[0054] The memory 190 is used to store various operating programs, data and applications for driving and controlling the control device 100 under the control of the controller. The memory 190 can store various control signal instructions input by the user.

[0055] The power supply 180 is used to provide operating power support for each component of the control device 100 under the control of the controller.

[0056] To facilitate 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 and software resources in the display device 200. The operating system may provide a user interface (control the display device), allow the user to interact with the display device 200, and support the running of various application programs.

[0057] It should be noted that the operating system may 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 specially developed for the display device.

[0058] The operating system can be divided into different modules or layers according to the functions implemented.

[0059] For example, Figure 4 As shown, in some embodiments, the system is divided into four layers, from top to bottom, namely, the application layer (referred to as "application layer"), the application framework layer (referred to as "framework layer"), the system library layer and the kernel layer.

[0060] In some embodiments, the application layer provides services and interfaces for applications, enabling the display device 200 to run applications and interact with the user based on the applications. The application layer can host at least one application, which can include built-in window programs, system settings programs, clock programs, and the like, or applications developed by third-party developers. In specific implementations, the application packages in the application layer are not limited to the examples above.

[0061] The framework layer provides applications with an application programming interface (API) and programming framework. The application framework layer includes predefined functions. The application framework layer acts as a processing center, determining the actions taken by applications in the application layer. Through the API, applications can access system resources and services during execution.

[0062] like Figure 4 As shown, in the embodiment of the present application, the application framework layer includes a view system, managers, content providers, etc., wherein the view system can design and implement the interface and interaction of the application, and the view system includes lists, grids, text boxes, buttons, etc. The manager includes at least one of the following modules: an activity manager for interacting with all activities running in the system; a location manager for providing system services or applications with access to the system location service; a package manager for retrieving various information related to the application packages currently installed on the device; a notification manager for controlling the display and clearing of notification messages; and a window manager for managing icons, windows, toolbars, wallpapers, and desktop widgets on the user interface.

[0063] In some embodiments, the activity manager is used to manage the lifecycle of each application and common navigation back functions, such as controlling application exit, opening, and back. The window manager is used to manage all window programs, such as obtaining the display screen size, determining whether there is a status bar, locking the screen, taking screenshots, and controlling changes in display windows, such as shrinking, shaking, or distorting the display window.

[0064] In some embodiments, the system runtime layer can provide support for the framework layer. When the framework layer is used, the operating system will run the instruction library contained in the system runtime layer, such as the C / C++ instruction library, to implement the functions to be implemented by the framework layer.

[0065] In some embodiments, the kernel layer is a functional layer between the hardware and software of the display device 200. The kernel layer can implement functions such as hardware abstraction, multitasking, and memory management. Figure 4 As shown, the kernel layer can be configured with hardware drivers, and 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 driver (such as fingerprint sensor, temperature sensor, pressure sensor, etc.), and power driver, etc.

[0066] It should be noted that the above example is only a simple division of the operating system functions and does not constitute a limitation on the specific operating system form of the display device 200 in the embodiment of the present application. Depending on factors such as the function of the display device and the type of operating system, the number of levels and specific level types contained in the operating system may be expressed in other forms.

[0067] Based on this, in some embodiments, the present application provides a policy determination method, which is applied to a server, such as Figure 5 As shown, the method includes:

[0068] Step 502: Obtain a set of user identifiers, weight and average the first indicator evaluation value and the first target weight associated with the first user identifier in the user identifier set to obtain a first comprehensive evaluation value, wherein the first indicator evaluation value is the evaluation value of the user represented by the first user identifier under the preset indicator after the target content push strategy is set. The preset indicator is used to reflect the user's interest in the target push content, and the target push content refers to the content pushed by the target content push strategy.

[0069] The user identification set can be expressed as U = {u1,u2,…,u nThe user ID set represents a user group, and the user ID set includes the user ID of each user in the user group. The user group may be a user group on a media service platform, where each user in the user group has a registered user account on the media service platform. The media service platform may provide playback and search services. The server may be a backend server of the media service platform, and the display device may run a media service application corresponding to the media service platform, i.e., a client.

[0070] Content push strategy refers to the strategy for pushing content. There can be multiple content push strategies. Each content push strategy has a strategy identifier. The strategy identifier set can be used to store the strategy identifiers of each content push strategy. For example, if there are k content push strategies, the strategy identifier set can be S = {s1, s2, ..., s k The target content push strategy can be any content push strategy.

[0071] To verify the effectiveness of a content push policy, you can set the policy for a subset of users in the user group, while leaving the policy unset for the remaining users. For different content push policies, the number of users set for that policy can be adjusted based on actual needs. For example, if there are 200 users in a user group, to verify the effectiveness of content push policy 1, you can set content push policy 1 for 100 of them. To verify the effectiveness of content push policy 2, you can set content push policy 2 for 90 of them.

[0072] A content push policy is a strategy for pushing content. The pushed content can be, but is not limited to, advertising or media data. A content push policy can be associated with advertising data or media data. For example, a content push policy can include a media asset identifier, which is an identifier for the media asset. Media assets refer to content assets, and can be, but are not limited to, text, images, audio, or video produced by media organizations (such as newspapers, radio stations, television stations, websites, and news agencies). Media asset identifiers can be associated with media data, which can include images, text, or video related to the media asset. Users who have set a content push policy can receive the media data associated with that content push policy. For example, when a user triggers a display device to display an interface within a media service application using a remote control or voice, the display device can send a data retrieval request to the server. The data retrieval request is used to retrieve the data displayed in the interface, including the media data associated with the content push policy. The server can then return the media data associated with the content push policy to the display device, allowing the display device to display the media data.

[0073] Setting a content push strategy for a user can be understood as implementing a content push strategy for that user. Preset metrics are indicators used to evaluate the effectiveness of a content push strategy after it has been implemented. These metrics are related to the asset to which the pushed media asset data belongs. For example, preset metrics can include, but are not limited to, user click-through rate or user retention rate for the asset.

[0074] In some embodiments, the server can set a targeted content push policy for a first user in a user group. There are multiple first users, and users in the user group for whom no targeted content push policy has been set are referred to as second users. This can be understood as dividing the user group into an experimental group and a control group, with the users in the experimental group being the first users and the users in the control group being the second users. The first user identifier is the identifier of the first user, and the second user identifier is the identifier of the second user. An AB experiment is conducted using the experimental and control groups to verify the effectiveness of the targeted content push policy.

[0075] Before conducting an AB experiment, it's usually necessary to first find two user groups that meet the AA fairness requirement. Then, one of the two user groups, User Group A, serves as the control group, and the other, User Group B, serves as the experimental group. AA fairness requires that User Feature Distribution 1 for User Group A be consistent with the User Feature Distribution for User Group B. User Feature Distribution 2 for User Group A refers to the distribution formed by the user features of the users in User Group A. User Feature Distribution for User Group B refers to the distribution formed by the user features of the users in User Group B. For example, if User Group A includes users 1 to 100, and User Group B also includes users 101 to 200, then User Feature Distribution 1 is the distribution formed by the user features of each user in Users 1 to 100, and User Feature Distribution 2 is the distribution formed by the user features of each user in Users 101 to 200.

[0076] The user feature may include at least one attribute feature. For example, the user feature includes m attribute features. The user feature may be expressed as user feature X = {x1, x2, ..., x m}. Among them, x i is the i-th attribute feature, 1≤i≤m. Each attribute feature corresponds to an attribute dimension. Attribute dimensions may include, but are not limited to, age, gender, region, display device model such as a TV model, display device screen size such as a TV screen size, and membership tags. Membership tags can be divided into first member tags and second member tags, where the first member tag indicates that the user is a member and the second member tag indicates that the user is not a member. Attribute dimensions may also be statistical dimensions corresponding to the statistical data of the user's media playback behavior in various scenarios. The statistical data may include, but are not limited to, statistical information on the user's time preference for watching media, scene preference statistical information, actor preference information, category preference for watching content, and other statistical information.

[0077] In some embodiments, within a preset period of time, such as one week or one month, after setting the target content push strategy for the first user, the server can calculate the first indicator evaluation value of the first user and associate the first indicator evaluation value with the first user identifier, thereby obtaining the first indicator evaluation value associated with the first user identifier. Similarly, the server can calculate the first indicator evaluation value of the second user and associate the second indicator evaluation value with the second user identifier, thereby obtaining the second indicator evaluation value associated with the second user identifier.

[0078] Step 504, weight and average the second indicator evaluation value and the second target weight associated with the second user identifier in the user identifier set to obtain a second comprehensive evaluation value, wherein the second indicator evaluation value is the evaluation value of the user represented by the second user identifier under the preset indicator when the target content push strategy is not set, and the first target weight and the second target weight are used to balance the distribution of user characteristics associated with the first user identifier and the distribution of user characteristics associated with the second user identifier.

[0079] In some embodiments, the server may determine the weight of each user in the user group represented by the set of user identifiers, and associate the user's weight with the user's user identifier. The weight of the first user is referred to as the first target weight, and the weight of the second user is referred to as the second target weight. Different first users may each have their own first target weight, or the weights of the first users may be the same weight. Similarly, different second users may each have their own second target weight, or the weights of the second users may be the same weight.

[0080] In some embodiments, the server can generate a first feature distribution based on the user features associated with each first user identifier and the weight parameters corresponding to each first user identifier, and generate a second feature distribution based on the user features associated with each second user identifier and the weight parameters corresponding to each second user identifier, and adjust the weight parameters to minimize the difference between the first feature distribution and the second feature distribution, and determine the weight parameters corresponding to the first user identifier after the final adjustment as the first target weight, and determine the weight parameters corresponding to the second user identifier after the final adjustment as the second target weight. Then, the first user identifier is associated with the corresponding first target weight, and the second user identifier is associated with the corresponding second target weight. In this embodiment, the DCB (Differentiated Confounder Balancing) algorithm can be used to determine the first target weight and the second target weight. The DCB algorithm is a method used for causal effect assessment. In the DCB algorithm model, a weight is assigned to each sample or example, and the weights obtained by training generate a balanced data set in which the experimental group and the control group have similar distributions on the confounding factors.

[0081] As can be seen from this, the first target weight and the second target weight are used to balance the first feature distribution and the second feature distribution. The first feature distribution is the distribution of user features associated with each first user identifier, and the second feature distribution is the distribution of user features associated with each second user identifier. The first target weight and the second target weight can minimize the difference between the first feature distribution and the second feature distribution.

[0082] In step 506, the difference between the first comprehensive evaluation value and the second comprehensive evaluation value is used as the effectiveness evaluation value of the target content push strategy. The effectiveness evaluation value is used to reflect the overall interest of the target user group in the target push content. The target user group refers to the user group represented by the user identifier set.

[0083] Wherein, the target content push strategy is represented by s, the evaluation value under the preset indicator is represented by t, the first comprehensive evaluation value is represented by E[t|s=1], and the second comprehensive evaluation value is represented by E[t|s=0]. Then, the effect evaluation value of the target content push strategy ATE(s)=E[t|s=1]-E[t|s=0]. s=1 means that the evaluation value of the users in the experimental group, i.e., the first indicator evaluation value associated with the first user identifier, is used when calculating E[t|s=1]. s=0 means that the evaluation value of the users in the control group, i.e., the second indicator evaluation value associated with the second user identifier, is used when calculating E[t|s=0].

[0084] Step 508: Obtain the basic matching degree between the target content push strategy and the target user.

[0085] The target user may be any user, any user in the user group represented by the user identification set, or a user outside the user group represented by the user identification set, for example, a user newly registered in the media resource service platform.

[0086] In some embodiments, for each user in the user group, the server can calculate the basic degree of match between each content push strategy and each user. For example, the server can calculate the basic degree of match between each user and the target content push strategy. The calculated basic degree of match can then be associated with the user identifier and stored. Thus, the server can obtain the basic degree of match between the target content push strategy and the target user based on the target user's user identifier.

[0087] In some embodiments, the server may determine a basic matching degree between the target content push strategy and the target user based on attribute characteristics included in the user characteristics of the target user.

[0088] Step 510: Comprehensively consider the effect evaluation value and the basic matching degree to obtain the target matching degree between the target content push strategy and the target user, select the target policy identifier from the policy identifier set according to the size of the target matching degree, and determine the content push strategy corresponding to the target policy identifier as the target content push strategy that matches the target user.

[0089] Among them, the target matching degree is positively correlated with the effect evaluation value, and the target matching degree is positively correlated with the basic matching degree. The server can use the product of the effect evaluation value and the basic matching degree as the target matching degree. B(s) represents the basic matching degree between the target content push strategy and the target user, ρ(u,s) represents the target matching degree, and u represents the target user, then ρ(u,s) = ATE(s) * B(s). Since B(s) can reflect the matching degree between the target content push strategy and the target user, B(s) can be used to reflect the personalized strategy matching, and ATE(s) can reflect the push effect of the target content push strategy from a global perspective. Therefore, by combining the global and personalized strategies, the content push strategy can be matched for the target user, and a better content push strategy can be matched.

[0090] In some embodiments, each policy identifier in the policy identifier set represents a content push policy. The server can determine the target matching degree between the target user and each content push policy. Then, based on the target matching degree, the target policy identifier can be selected from the policy identifier set, wherein the target matching degree between the content push policy represented by the target policy identifier and the target user is the greatest. * represents the target policy identifier, then

[0091] In this embodiment, since the first target weight and the second target weight are used to balance the distribution of user features associated with the first user identifier and the distribution of user features associated with the second user identifier, the first indicator evaluation value is the evaluation value of the user represented by the first user identifier under the preset indicator after the target content push strategy is set, and the second indicator evaluation value is the evaluation value of the user represented by the second user identifier under the preset indicator when the target content push strategy is not set. Therefore, for the target content push strategy, the user represented by the first user identifier is the user in the experimental group, and the user represented by the second user identifier is the user in the control group. The first indicator evaluation value and the first target weight associated with the first user identifier in the user identifier set are weighted and averaged to obtain a first comprehensive evaluation value, and the second indicator evaluation value and the second target weight associated with the second user identifier in the user identifier set are weighted and averaged to obtain a second comprehensive evaluation value. Since the preset indicator is used to reflect the user's interest in the target push content, and the target push content refers to the content pushed by the target content push strategy, when the experimental group and the control group do not meet AA fairness, the first comprehensive evaluation value can reflect the user's interest in the target push content. The feedback of the experimental group when the experimental group and the control group meet AA fairness, and the second comprehensive evaluation value can reflect the feedback of the control group when the experimental group and the control group meet AA fairness. The difference between the first comprehensive evaluation value and the second comprehensive evaluation value is used as the effect evaluation value of the target content push strategy, which can make the effect evaluation value reflect the effect represented by the target content push strategy when the experimental group and the control group meet AA fairness, that is, the effect evaluation value can reflect the overall interest of the target user group in the target push content. The target user group refers to the user group represented by the user identifier set. Further obtain the basic matching degree between the target content push strategy and the target user, the comprehensive effect evaluation value and the basic matching degree, and obtain the target matching degree between the target content push strategy and the target user. According to the size of the target matching degree, the target policy identifier is selected from the policy identifier set, and the target policy identifier is matched with the content push strategy to be determined as the target content push strategy that matches the target user. In this way, the target content push strategy that matches the target user can be quickly determined without requiring the experimental group and the control group to meet AA fairness, which helps to improve the efficiency of content push.

[0092] In some embodiments, the present application provides a server comprising: a communication device configured to be communicatively connected to a display device; and at least one processor connected to the communication device and configured to: obtain a set of user identifiers, weight and average a first indicator evaluation value and a first target weight associated with a first user identifier in the user identifier set to obtain a first comprehensive evaluation value, wherein the first indicator evaluation value is an evaluation value of the user represented by the first user identifier under a preset indicator after a target content push strategy is set, and the preset indicator is used to reflect the user's interest in the target pushed content, and the target pushed content refers to the content pushed by the target content push strategy; weight and average a second indicator evaluation value and a second target weight associated with a second user identifier in the user identifier set to obtain a second comprehensive evaluation value, wherein the second indicator evaluation value is an evaluation value of the user represented by the second user identifier under a preset indicator after the target content push strategy is set. The evaluation value under the preset indicators when the target content push strategy is set, the first target weight and the second target weight are used to balance the distribution of user characteristics associated with the first user identifier and the distribution of user characteristics associated with the second user identifier; the difference between the first comprehensive evaluation value and the second comprehensive evaluation value is used as the effect evaluation value of the target content push strategy, and the effect evaluation value is used to reflect the overall interest of the target user group in the target push content, and the target user group refers to the user group represented by the user identifier set; the basic matching degree between the target content push strategy and the target user is obtained; the target matching degree between the target content push strategy and the target user is obtained by combining the effect evaluation value and the basic matching degree, and the target policy identifier is selected from the policy identifier set according to the size of the target matching degree, and the content push strategy corresponding to the target policy identifier is determined as the target content push strategy that matches the target user.

[0093] In this embodiment, since the first target weight and the second target weight are used to balance the distribution of user features associated with the first user identifier and the distribution of user features associated with the second user identifier, the first indicator evaluation value is the evaluation value of the user represented by the first user identifier under the preset indicator after the target content push strategy is set, and the second indicator evaluation value is the evaluation value of the user represented by the second user identifier under the preset indicator when the target content push strategy is not set. Therefore, for the target content push strategy, the user represented by the first user identifier is the user in the experimental group, and the user represented by the second user identifier is the user in the control group. The first indicator evaluation value and the first target weight associated with the first user identifier in the user identifier set are weighted and averaged to obtain a first comprehensive evaluation value, and the second indicator evaluation value and the second target weight associated with the second user identifier in the user identifier set are weighted and averaged to obtain a second comprehensive evaluation value. Since the preset indicator is used to reflect the user's interest in the target push content, and the target push content refers to the content pushed by the target content push strategy, when the experimental group and the control group do not meet AA fairness, the first comprehensive evaluation value can reflect the user's interest in the target push content. The feedback of the experimental group when the experimental group and the control group meet AA fairness, and the second comprehensive evaluation value can reflect the feedback of the control group when the experimental group and the control group meet AA fairness. The difference between the first comprehensive evaluation value and the second comprehensive evaluation value is used as the effect evaluation value of the target content push strategy, which can make the effect evaluation value reflect the effect represented by the target content push strategy when the experimental group and the control group meet AA fairness, that is, the effect evaluation value can reflect the overall interest of the target user group in the target push content. The target user group refers to the user group represented by the user identifier set. Further obtain the basic matching degree between the target content push strategy and the target user, the comprehensive effect evaluation value and the basic matching degree, and obtain the target matching degree between the target content push strategy and the target user. According to the size of the target matching degree, the target policy identifier is selected from the policy identifier set, and the target policy identifier is matched with the content push strategy to be determined as the target content push strategy that matches the target user. In this way, the target content push strategy that matches the target user can be quickly determined without requiring the experimental group and the control group to meet AA fairness, which helps to improve the efficiency of content push.

[0094] In some embodiments, the first target weight associated with each first user identifier is consistent, and the second target weight associated with each second user identifier is consistent, such as Figure 6As shown, the processor is further configured to: step 602, determine the mean of the user features associated with the first user identifier in the user identifier set to obtain a first average user feature, and determine the mean of the user features associated with the second user identifier in the user identifier set to obtain a second average user feature; step 604, determine a first product of the first average user feature and the first weight parameter, and determine a second product of the second average user feature and the second weight parameter; step 606, minimize the difference between the first product and the second product by adjusting the first weight parameter or the second weight parameter; step 608, use the first weight parameter in the case of minimizing the difference as the first target weight, and use the second weight parameter in the case of minimizing the difference as the second target weight.

[0095] Among them, the user features are in vector form, so the first product and the second product are both vectors, so the difference between the first product and the second product can be the distance between the vectors. The first average user feature can be expressed as , the second average user feature can be expressed as , the first weight parameter can be expressed as W t , the second weight parameter can be expressed as W c , then the first product is , the second product is , the difference is expressed as The function that determines the first target weight and the second target weight is expressed as f(U,X), then W=W t ∪W c .

[0096] In some embodiments, the first weight parameter can be a fixed value, the first target weight is equal to the first weight parameter, and only the second weight parameter is adjusted. The first weight parameter can be set as needed, for example, to 1, that is, W t =1.

[0097] In some embodiments, the second weight parameter can be a fixed value, the second target weight is equal to the second weight parameter, and only the first weight parameter is adjusted. The second weight parameter can be set as needed, for example, to 1, that is, W c =1.

[0098] In this embodiment, the difference between the first product and the second product is minimized by the first weight parameter or the second weight parameter, the first weight parameter when the difference is minimized is used as the first target weight, and the second weight parameter when the difference is minimized is used as the second target weight. This allows the difference to be minimized while minimizing the number of parameters to be adjusted, thereby improving the effect of minimizing the difference and enhancing the efficiency of obtaining the first target weight and the second target weight.

[0099] In some embodiments, when the processor executes weighted averaging of the first indicator evaluation value and the first target weight associated with the first user identifier in the user identifier set to obtain a first comprehensive evaluation value, it is configured to: multiply the first indicator evaluation value associated with the first user identifier by the first target weight associated with the first user identifier to obtain a first updated evaluation value corresponding to the first user identifier; and average the first updated evaluation values corresponding to the first user identifier in the user identifier set to obtain a first comprehensive evaluation value.

[0100] Similarly, when the processor executes the weighted averaging of the second indicator evaluation value and the second target weight associated with the second user identifier in the user identifier set to obtain the second comprehensive evaluation value, it is configured to: multiply the second indicator evaluation value associated with the second user identifier by the second target weight associated with the second user identifier to obtain the second updated evaluation value corresponding to the second user identifier; and average the second updated evaluation values corresponding to the second user identifier in the user identifier set to obtain the second comprehensive evaluation value.

[0101] When the processor executes the operation of using the difference between the first comprehensive evaluation value and the second comprehensive evaluation value as the effect evaluation value of the target content push strategy, the processor is configured to:

[0102] The effectiveness evaluation value of the target content push strategy is determined according to the following formula:

[0103]

[0104] Among them, s represents the target content push strategy, ATE(s) represents the effect evaluation value, represents the first comprehensive evaluation value, represents the second comprehensive evaluation value, w i is the target weight associated with the i-th user ID in the user ID set, t i is the indicator evaluation value associated with the i-th user ID in the user ID set. If the i-th user ID is the first user ID, then w i is the first target weight and t i is the first indicator evaluation value. If the i-th user identifier is the second user identifier, then w i is the second target weight and t i is the second indicator evaluation value, f i (s,1) represents the first user type label corresponding to the i-th user ID. If the i-th user ID is the first user ID, then f i The value of (s,1) is 1. If the i-th user ID is the second user ID, then f i The value of (s,1) is 0, f i (s,0) represents the second user type label corresponding to the i-th user ID. If the i-th user ID is the first user ID, then fi The value of (s,0) is 0. If the i-th user ID is the second user ID, then f i The value of (s,0) is 1, and n is the number of users in the target user group. n1 is also the number of users in the experimental group, that is, the number of first users. n2 is also the number of users in the control group, that is, the number of second users. The i-th user ID represents the i-th user, so t i is the index evaluation value (first index evaluation value or second index evaluation value) of the i-th user. i ∈W is the weight (first target weight or second target weight) of the i-th user in the user group represented by the user identity set, and W includes the weight of each user in the user group. i (s,1) is used to determine the user in the test group, that is, the first user. If the user is the first user, then f i (s,1)=1, otherwise, f i (s,1)=0. f i (s,0) is used to determine the user in the control group is the second user. If the user is the second user, then f i (s,0)=1, otherwise, f i (s,0)=0.

[0105] Among them, the principle of calculating ATE can be understood as an offline causal effect evaluation method, which can accurately evaluate the effectiveness of strategy delivery under non-AB experimental conditions.

[0106] In this embodiment, the first indicator evaluation value associated with the first user identifier and the first target weight are multiplied to obtain a first updated evaluation value, and the first updated evaluation values corresponding to each first user identifier are averaged to obtain a first comprehensive evaluation value. In this way, the first comprehensive evaluation value can accurately reflect the feedback of the experimental group without requiring the experimental group and the control group to meet AA fairness.

[0107] In some embodiments, the user features include attribute features of the target attribute dimension, such as Figure 7As shown, when the processor executes to obtain the basic matching degree between the target content push strategy and the target user, it is configured as follows: Step 702, obtain multiple feature category tags corresponding to the target attribute dimension, the feature category tags are used to distinguish the attribute features of different categories of the target attribute dimension; Step 704, determine the third user identifier corresponding to the first feature category tag from the user identifier set according to the first feature category tag, wherein the first feature category tag is any one of the multiple feature category tags, and the attribute features of the target attribute dimension in the user features associated with the third user identifier match the first feature category tag; Step 706, determine the feature matching degree between the target content push strategy and the first feature category tag based on the indicator evaluation value associated with the third user identifier; Step 708, determine the second feature category tag from the multiple feature category tags that matches the attribute features of the target attribute dimension in the user features of the target user; Step 710, determine the basic matching degree between the target content push strategy and the target user according to the feature matching degree between the target content push strategy and the second feature category tag, wherein the basic matching degree is positively correlated with the feature matching degree.

[0108] The feature matching degree reflects the interest level of the user represented by the third user identifier in the target pushed content. Specifically, it reflects the interest level of the user with the attribute features represented by the first feature category label in the target pushed content. User features can include attribute features from multiple attribute dimensions, including but not limited to age, gender, region, display device model (e.g., TV model), display device screen size (e.g., TV screen size), and membership tags. The target attribute dimension is any attribute dimension. Feature category labels are used to distinguish between different categories of attribute features within the attribute dimension. For example, if the attribute dimension is gender, and the gender attribute features are divided into "male" and "female," then age has two feature category labels: one for "male" and the other for "female." Dividing feature category labels can be understood as bucketing: one feature category label represents one bucket, and different feature category labels represent different buckets. Multiple buckets can be created for a single attribute dimension.

[0109] The third user identifier refers to a user identifier whose attribute characteristics of the target attribute dimension in the user characteristics associated with the user identifier set match the first feature category label. Taking the target attribute dimension as "display device model" as an example, if the first feature category label represents a display device of model A, and if the attribute characteristics of the target attribute dimension in the user characteristics associated with the user identifier are characteristics representing model A, then the user identifier is the third user identifier.

[0110] The first feature category label is any feature category label corresponding to the target attribute dimension. Therefore, using the same method, the feature matching degree between each feature category label corresponding to the target attribute dimension and the target content push strategy can be determined. Taking gender as the target attribute dimension as an example, the feature matching degree between the feature category label representing "male" and the target content push strategy, as well as the feature matching degree between the feature category label representing "female" and the target content push strategy can be determined.

[0111] For the same user, the attribute features of the target attribute dimension in the user features can only match one feature category label corresponding to the target attribute dimension. The second feature category label is the feature category label that matches the attribute features of the target attribute dimension in the user features of the target user.

[0112] In some embodiments, the server may use the feature matching degree between the target content push strategy and the second feature category tag as the basic matching degree between the target content push strategy and the target user.

[0113] In this embodiment, since the feature category label is used to distinguish the attribute characteristics of different categories of the target attribute dimension, thereby determining the feature matching degree between the target content push strategy and the feature category label, the impact degree of the target content push strategy on users with different attribute characteristics can be clarified. Since the second feature category label matches the attribute characteristics of the target attribute dimension in the user characteristics of the target user, the basic matching degree of the target content push strategy and the target user is determined based on the feature matching degree between the target content push strategy and the second feature category label. The basic matching degree can be used to accurately evaluate the impact of the target content push strategy on the target user locally.

[0114] In some embodiments, when the processor executes to obtain multiple feature category labels corresponding to the target attribute dimension, it is configured as follows: if the attribute feature of the target attribute dimension is a discrete numerical value, then the different attribute features of the target attribute dimension are encoded respectively, and the encoded values of the different attribute features are used as the feature category labels corresponding to the target attribute dimension; if the attribute feature of the target attribute dimension is a continuous numerical value, then the user identifiers in the user identifier set are sorted according to the size of the attribute feature of the target attribute dimension to obtain a user identifier sequence, a reference user identifier is determined from the user identifier sequence, and the feature category label corresponding to the target attribute dimension is determined based on the attribute feature of the target attribute dimension in the user features associated with the reference user identifier, wherein the proportion of the user identifiers in the subsequence from the first user identifier to the reference user identifier in the user identifier sequence is equal to the preset proportion.

[0115] Attribute characteristics of the attribute dimension can be categorized as discrete or continuous. For example, gender is categorized as male and female, and gender attributes are categorized as attributes representing males and attributes representing females. Therefore, gender is discrete, while age is also discrete. For discrete attribute characteristics, you can encode the attribute and then compare the encoded value with the feature category label. If the comparison is consistent, the comfort characteristic is determined to match the feature category label.

[0116] The preset proportion can be set as needed, and there can be multiple preset proportions, and the preset proportion can be but not limited to 10% or 20%, etc. The subsequence includes the first user identifier and the reference user identifier. The server can determine the feature type label based on the attribute characteristics of the target attribute dimension in the user feature associated with the reference user identifier. For example, the attribute dimension is age, and there are 2 preset proportions, namely 50% and 80%. If the age in the user feature associated with the reference user identifier at 50% is 25 years old, then the value range of 0-25 years old can be used as a feature category label. If the age in the user feature associated with the reference user identifier at 80% is 45 years old, then the value range of 25 to 45 years old can be used as another feature category label, and the value range of 45 to 100 years old can be used as another feature category label, thereby obtaining multiple feature category labels.

[0117] Encoding can be, but is not limited to, one-hot encoding. For discrete attribute features, bucketing can be performed using one-hot encoding, generating multiple feature category labels. For example, if gender is divided into male and female, one-hot encoding can be used to represent "male" and "female" with 0 and 1, resulting in two feature category labels of 0 and 1, respectively.

[0118] For continuous attribute features, we can use decile coding to divide them into 10 buckets, that is, generate 10 feature category labels, which can be calculated using B x′ =bin(x′)={b 1,x′ ,b 2,x′ ,…}. Represents the set of feature category labels generated for the target attribute dimension x′.

[0119] In some embodiments, the server may sort the user identifiers in the user identifier set according to the size of the attribute features of the target attribute dimension to obtain a user identifier sequence. The larger the attribute features of the target attribute dimension, the higher the user identifier is ranked in the user identifier sequence. Alternatively, the smaller the attribute features of the target attribute dimension, the higher the user identifier is ranked in the user identifier sequence. The server may divide the user identifier sequence into multiple sub-user identifier sequences. For example, deciles encoding may be used to divide the user identifier sequence into multiple sub-user identifier sequences. The ratio of the number of user identifiers in each sub-user identifier sequence to the total number of users is 10%, where the total number of users refers to the number of user identifiers in the user identifier set.

[0120] For example, if a user ID sequence includes user IDs of 200 users and the target attribute dimension is age, the 200 user IDs can be sorted in ascending order to obtain a user ID sequence. Then, the first 0-10% constitute a sub-user ID sequence, the first 10%-20% constitute a sub-user ID sequence, the first 30%-40% constitute a sub-user ID sequence, the first 40%-50% constitute a sub-user ID sequence, the first 50%-60% constitute a sub-user ID sequence, the first 60%-70% constitute a sub-user ID sequence, the first 70%-80% constitute a sub-user ID sequence, the first 80%-90% constitute a sub-user ID sequence, and the first 90%-100% constitute a sub-user ID sequence.

[0121] In some embodiments, the server may use the attribute feature of the target attribute dimension in the user features associated with the last user ID in the sub-user ID sequence as a reference attribute feature, use the value range formed by two adjacent reference attribute features as a feature category label, and use the value range formed by the smallest attribute feature of the target attribute dimension and the smallest reference attribute feature as a feature category label. For example, the age value in the user features associated with the last user ID in the sub-user ID sequence that is composed of the first 0-10% is used as the reference attribute feature.

[0122] In this embodiment, for discrete attribute features, feature category labels are determined by encoding, and for continuous attribute features, feature category labels are determined by using attribute features, so that feature category labels can distinguish different types of attribute features.

[0123] In some embodiments, as Figure 8As shown, when the processor executes the indicator evaluation value associated with the third user identifier to determine the degree of feature matching between the target content push strategy and the first feature category label, it is configured as follows: Step 802, taking the proportion of the indicator evaluation value associated with the third user identifier in the total indicator evaluation value as the evaluation value proportion, and the total indicator evaluation value is the sum of the indicator evaluation values associated with the user identifiers in the user identifier set; Step 804, taking the proportion of the third user identifier in the user identifier set as the user number proportion; Step 806, taking the ratio of the evaluation value proportion to the user number proportion as the degree of feature matching between the target content push strategy and the first feature category label.

[0124] Among them, the total index evaluation value can be expressed as n is the number of users in the user group represented by the user identification set, and is also the number of user identifications in the user identification set.

[0125] In some embodiments, the user identification set includes multiple third user identifications. The server can sum the indicator evaluation values associated with each third user identification to obtain the target indicator evaluation value, and use the ratio of the target indicator evaluation value to the total indicator evaluation value as the evaluation value ratio. k,x′ For example, the evaluation value ratio can be expressed as Among them, f(u i ,x,b k,x′ ) takes a value of 0 or 1, representing the i-th user u i Whether the attribute feature x of the target attribute dimension x′ is the first feature category label b corresponding to the target attribute dimension x′ k,x′ Match, if it matches, then f(u i ,x,b k,x′ ) is 1. If it does not match, then f(u i ,x,b k,x′ ) is 0. Thus, Represents the sum of the indicator evaluation values associated with each third user identifier.

[0126] In some embodiments, the server can count the number of third user identifiers in the user identifier set to obtain the target number, count the number of user identifiers in the user identifier set to obtain the total number of users, and use the ratio of the target number to the total number of users as the user number ratio. The user number ratio can be expressed as Among them, |U| represents the total number of users, that is, |U|=n, represents the number of targets, p(b k,x′) represents the proportion of users. Therefore, since the third user identifier is the user identifier whose attribute feature of the target attribute dimension in the associated user feature matches the first feature category label, in fact, is the number of third user identifiers in the user identifier set, that is, the target number.

[0127] In some embodiments, the feature matching degree between the target content push strategy and the first feature category label can be expressed as:

[0128]

[0129] Among them, ROI(b k,x′ ) represents the degree of feature matching between the target content push strategy and the kth feature category label corresponding to the target attribute dimension x′. ROI (Return on Investment) can be understood as the return on investment of the content push strategy. A higher ROI value indicates greater user interest, thus reflecting a higher rate of return on revenue from users with matching feature category labels.

[0130] In this embodiment, since the evaluation value ratio can reflect the overall feedback of users with attribute characteristics matching the first feature category label, the larger the evaluation value ratio, the more interested the users with attribute characteristics matching the first feature category label are in the content pushed by the target content push strategy. For example, if the evaluation value is click-through rate, the larger the click-through rate ratio, the more interested they are. The user number ratio can reflect the ratio of users with attribute characteristics matching the first feature category label. Therefore, by combining the evaluation value ratio and the user number ratio, the degree of feature matching between the target content push strategy and the first feature category label can be accurately determined.

[0131] In some embodiments, the user characteristics include attribute characteristics of multiple attribute dimensions, there are multiple second feature category labels, and different second feature category labels are selected from feature category labels corresponding to different attribute dimensions; when the processor executes the feature matching degree between the target content push strategy and the second feature category label to determine the basic matching degree between the target content push strategy and the target user, it is configured to: comprehensively calculate the feature matching degree between the multiple second feature category labels and the target content push strategy, and obtain the basic matching degree between the target content push strategy and the target user.

[0132] Among them, for each attribute dimension, the server can select a second feature category label that matches the attribute feature of the attribute dimension in the user feature of the target user from multiple feature category labels corresponding to the attribute dimension. For example, the attribute dimension is gender, the gender of the target user is male, and the attribute dimension "gender" corresponds to feature category label 1 and feature category label 2. Feature category label 1 represents the attribute feature "male", and feature category label 2 represents the attribute feature "female". Therefore, the second feature category label is feature category label 1. For another example, the attribute dimension is age, and the attribute dimension "age" corresponds to feature category label a, feature category label b, and feature category label c. Feature category label a represents the age range of 1 to 18, feature category label b represents the age range of 19 to 48, and feature category label c represents the age range of 49 to 100. If the target user is 15 years old, the second feature category label is feature category label a, so that multiple second feature category labels can be determined.

[0133] In this embodiment, the feature matching degrees corresponding to multiple second feature category labels are comprehensively calculated to obtain the basic matching degree between the target content push strategy and the target user, thereby evaluating the basic matching degree between the target content push strategy and the target user from a multi-dimensional perspective, avoiding the instability problem of single-dimensional evaluation.

[0134] In some embodiments, the server may calculate the average of the feature matching degrees between the determined multiple second feature category tags and the target content push strategy, and use the calculated average value as the basic matching degree between the target content push strategy and the target user. For example, the basic matching degree B(s) can be expressed as When the processor executes the comprehensive effect evaluation value and the basic matching degree to obtain the target matching degree between the target content push strategy and the target user, it is configured as follows:

[0135] The target matching degree between the target content push strategy and the target user is determined according to the following formula:

[0136]

[0137] Among them, ρ(u,s) represents the target matching degree, represent The degree of feature matching with the target content push strategy. X′ represents the set of attribute dimensions, and |X′| represents the number of attribute dimensions. j Represents the j-th attribute dimension. represents the second feature category label selected from multiple feature category labels corresponding to the j-th attribute dimension, u represents the target user, Represents the basic matching degree between the target content push strategy and the target user. If there are m attribute dimensions, then 1≤j≤m.

[0138] In this embodiment, the feature matching degrees corresponding to multiple second feature category labels are averaged to obtain the basic matching degree between the target content push strategy and the target user, thereby evaluating the basic matching degree between the target content push strategy and the target user from a multi-dimensional perspective, avoiding the instability problem of single-dimensional evaluation.

[0139] In some embodiments, the present application provides a media asset display method, which is applied to a display device, and the method includes: responding to an interface display operation, determining an interface identifier, and sending a data acquisition request to a server, wherein the data acquisition request carries the interface identifier and the user identifier of the target user; receiving interface data returned by the server in response to the data acquisition request, wherein the interface data includes target media asset data, and the target media asset data is media asset data associated with a target content push strategy that matches the target user; displaying an interface corresponding to the interface identifier, and displaying the media asset data in the interface; wherein the target content push strategy is determined by the above-mentioned policy determination method.

[0140] Among them, the server responds to the data acquisition request and obtains the user identifier of the target user from the data acquisition request. The user identifier of the target user is called the target user identifier. Then, the server can determine the target content push strategy that matches the target user identifier and the media asset data associated with the target content push strategy. The media asset data associated with the target content push strategy is the media asset data planned to be pushed by the target content push strategy. The server can generate a request response result corresponding to the data acquisition request and return the request response result to the display device. The request response result carries the media asset data associated with the target content push strategy. Figure 9 As shown, the display device sends a data acquisition request to the server, and the server responds to the data acquisition request and returns a request response result to the display device.

[0141] In this embodiment, since the first target weight and the second target weight are used to balance the distribution of user features associated with the first user identifier and the distribution of user features associated with the second user identifier, the first indicator evaluation value is the evaluation value of the user represented by the first user identifier under the preset indicator after the target content push strategy is set, and the second indicator evaluation value is the evaluation value of the user represented by the second user identifier under the preset indicator when the target content push strategy is not set. Therefore, for the target content push strategy, the user represented by the first user identifier is the user in the experimental group, and the user represented by the second user identifier is the user in the control group. The first indicator evaluation value and the first target weight associated with the first user identifier in the user identifier set are weighted and averaged to obtain a first comprehensive evaluation value, and the second indicator evaluation value and the second target weight associated with the second user identifier in the user identifier set are weighted and averaged to obtain a second comprehensive evaluation value. This can make it possible for the first comprehensive evaluation value to reflect the situation where the experimental group and the control group do not meet AA fairness. When the experimental group and the control group meet AA fairness, the feedback of the experimental group, and the second comprehensive evaluation value can reflect the feedback of the control group when the experimental group and the control group meet AA fairness. The difference between the first comprehensive evaluation value and the second comprehensive evaluation value is used as the effect evaluation value of the target content push strategy, so that the effect evaluation value can reflect the effect represented by the target content push strategy when the experimental group and the control group meet AA fairness. Further, the basic matching degree between the target content push strategy and the target user is obtained, the comprehensive effect evaluation value and the basic matching degree are obtained, and the target matching degree between the target content push strategy and the target user is obtained. According to the size of the target matching degree, the target policy identifier is selected from the policy identifier set, and the target policy identifier is matched with the content push strategy to determine it as the target content push strategy that matches the target user. In this way, the target content push strategy that matches the target user can be quickly determined without requiring the experimental group and the control group to meet AA fairness, which helps to improve the efficiency of content push.

[0142] The company's existing operational strategies rely heavily on the capabilities of the AB platform, hindering the progress of agile strategy deployment. These include: With a large number of operational strategies deployed, existing AB experimentation capabilities are time-consuming and labor-intensive, hindering agile strategy deployment evaluation. AB experiments require AA fairness, which means that users in the experimental group's traffic will not quickly experience the improved user experience brought by the latest marketing strategies. Without AB experiments, it is difficult to conduct an unbiased estimate of the effectiveness of strategy deployment under agile strategy deployment conditions. User behavior is diverse, and different users have different feedback on product marketing strategies. The company's current strategy deployment lacks the ability to personalize its delivery. The company's operational strategies are diverse, and each strategy operates independently, lacking the ability to optimize the overall strategy.

[0143] The strategy determination method provided in this application does not require AA fairness, thus accurately evaluating the effectiveness of strategy delivery under non-AB experiment conditions, which can compensate for the slow efficiency and loss of delivery effectiveness of existing AB platforms. The input-output ratio of strategy delivery is evaluated from a multi-dimensional perspective, avoiding the instability of single-dimensional evaluation. Based on the user's individual behavioral attributes, the causal effect of each strategy and the input-output ratio from a multi-dimensional perspective are combined to locally match the optimal strategy for the user. By effectively combining strategy effectiveness and ROI evaluation, the benefits of the entire operational strategy are maximized. The strategy determination method provided in this application uses causal inference to perform unbiased estimation and evaluation of the effectiveness of strategy delivery under non-AA fair comparisons, significantly reducing the dependence of strategy delivery on the AB platform and achieving the effect of rapid strategy iteration. Through multi-dimensional analysis of the input-output ratio of strategy delivery, user attribute sets with high strategy effectiveness are identified. Based on the user's individual behavior, operational strategies are matched based on the input-output ratio of the strategy's causal effect, achieving the effect of personalized marketing strategy delivery, thereby achieving the effect of maximizing the overall strategy benefits.

[0144] In some embodiments, the present application provides a display device, including: a display and a controller; the controller is configured to: determine an interface identifier in response to an interface display operation, and send a data acquisition request to a server, wherein the data acquisition request carries the interface identifier and the user identifier of the target user; receive interface data returned by the server in response to the data acquisition request, wherein the interface data includes target media data, and the target media data is media data associated with a target content push strategy that matches the target user; control the display to display an interface corresponding to the interface identifier, and display the media data in the interface; wherein the target content push strategy is determined by the above-mentioned policy determination method.

[0145] In this embodiment, since the first target weight and the second target weight are used to balance the distribution of user features associated with the first user identifier and the distribution of user features associated with the second user identifier, the first indicator evaluation value is the evaluation value of the user represented by the first user identifier under the preset indicator after the target content push strategy is set, and the second indicator evaluation value is the evaluation value of the user represented by the second user identifier under the preset indicator when the target content push strategy is not set. Therefore, for the target content push strategy, the user represented by the first user identifier is the user in the experimental group, and the user represented by the second user identifier is the user in the control group. The first indicator evaluation value and the first target weight associated with the first user identifier in the user identifier set are weighted and averaged to obtain a first comprehensive evaluation value, and the second indicator evaluation value and the second target weight associated with the second user identifier in the user identifier set are weighted and averaged to obtain a second comprehensive evaluation value. This can make it possible for the first comprehensive evaluation value to reflect the situation where the experimental group and the control group do not meet AA fairness. When the experimental group and the control group meet AA fairness, the feedback of the experimental group, and the second comprehensive evaluation value can reflect the feedback of the control group when the experimental group and the control group meet AA fairness. The difference between the first comprehensive evaluation value and the second comprehensive evaluation value is used as the effect evaluation value of the target content push strategy, so that the effect evaluation value can reflect the effect represented by the target content push strategy when the experimental group and the control group meet AA fairness. Further, the basic matching degree between the target content push strategy and the target user is obtained, the comprehensive effect evaluation value and the basic matching degree are obtained, and the target matching degree between the target content push strategy and the target user is obtained. According to the size of the target matching degree, the target policy identifier is selected from the policy identifier set, and the target policy identifier is matched with the content push strategy to determine it as the target content push strategy that matches the target user. In this way, the target content push strategy that matches the target user can be quickly determined without requiring the experimental group and the control group to meet AA fairness, which helps to improve the efficiency of content push.

[0146] In some embodiments, as Figure 10 As shown, a timing diagram corresponding to a media resource display method is provided. Specifically,

[0147] 1. Weight and average the first indicator evaluation value and the first target weight associated with the first user identifier in the user identifier set to obtain a first comprehensive evaluation value; weight and average the second indicator evaluation value and the second target weight associated with the second user identifier in the user identifier set to obtain a second comprehensive evaluation value; and use the difference between the first comprehensive evaluation value and the second comprehensive evaluation value as the effectiveness evaluation value of the target content push strategy;

[0148] The first indicator evaluation value is the evaluation value of the user represented by the first user identifier under the preset indicator after the target content push policy is set. The second indicator evaluation value is the evaluation value of the user represented by the second user identifier under the preset indicator without the target content push policy being set. The first target weight and the second target weight are used to balance the distribution of user features associated with the first user identifier with the distribution of user features associated with the second user identifier.

[0149] 2. Obtain multiple feature category labels corresponding to the target attribute dimension, determine the third user identifier corresponding to the first feature category label from the user identifier set based on the first feature category label, determine the feature matching degree between the target content push strategy and the first feature category label based on the indicator evaluation value associated with the third user identifier, determine the second feature category label that matches the attribute characteristics of the target attribute dimension in the user's user characteristics from the multiple feature category labels, and determine the basic matching degree between the target content push strategy and the user based on the feature matching degree between the target content push strategy and the second feature category label.

[0150] The user is any user in the user group represented by the set of user identifiers. The feature category label is used to distinguish different categories of attribute features in the target attribute dimension. The first feature category label is any one of multiple feature category labels, and the attribute features in the target attribute dimension in the user features associated with the third user identifier match the first feature category label. The basic matching degree is positively correlated with the feature matching degree.

[0151] 3. Comprehensively evaluate the effect and the basic matching degree to obtain the target matching degree between the target content push strategy and the user. Select the target strategy identifier from the strategy identifier set based on the target matching degree, and determine the content push strategy corresponding to the target strategy identifier as the target content push strategy that matches the user.

[0152] 4. Receive a data acquisition request sent by the display device, wherein the data acquisition request carries the interface identifier and the user identifier of the target user.

[0153] 5. If the target content push strategy that matches the target user has been determined, the media data associated with the target content push strategy is obtained, and the request response result carrying the media data is returned to the display device; if the target content push strategy that matches the target user has not yet been determined, the process of determining the target content push strategy that matches the target user can be executed, and then the media data associated with the target content push strategy is obtained, and the request response result carrying the media data is returned to the display device.

[0154] 6. The controller in the display device obtains the request response result and controls the display to display the media data.

[0155] It should be understood that, although the steps in the flowcharts of the above embodiments are shown in sequence as indicated by the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be performed in other orders. Moreover, at least a portion of the steps in the flowcharts of the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily to be performed in sequence, but can be performed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0156] Based on the same inventive concept, the present application also provides a policy determination device for implementing the aforementioned policy determination method. The solution provided by this device is similar to the solution described in the aforementioned method. For specific limitations, please refer to the limitations of the policy determination method above and will not be repeated here.

[0157] In some embodiments, a policy determination device is provided, which is used to: obtain a user identification set, weight and average a first indicator evaluation value associated with a first user identification in the user identification set and a first target weight of the first user identification to obtain a first comprehensive evaluation value, wherein the first indicator evaluation value is an evaluation value of the user represented by the first user identification under a preset indicator after the target content push strategy has been executed, and the preset indicator is used to reflect the user's interest in the target pushed content, and the target pushed content refers to the content pushed by the target content push strategy; weight and average a second indicator evaluation value associated with a second user identification in the user identification set and a second target weight of the second user identification to obtain a second comprehensive evaluation value, wherein the second indicator evaluation value is an evaluation value of the user represented by the second user identification under a preset indicator after the target content push strategy has not been executed. In this case, the evaluation value under the preset indicators, the first target weight and the second target weight are used to balance the distribution of user features associated with the first user identifier and the distribution of user features associated with the second user identifier; the difference between the first comprehensive evaluation value and the second comprehensive evaluation value is used as the effect evaluation value of the target content push strategy, and the effect evaluation value is used to reflect the overall interest of the target user group in the target push content, and the target user group refers to the user group represented by the user identifier set; the basic matching degree between the target content push strategy and the target user is obtained; the target matching degree between the target content push strategy and the target user is obtained by combining the effect evaluation value and the basic matching degree, and the target policy identifier is selected from the policy identifier set according to the size of the target matching degree, and the content push strategy corresponding to the target policy identifier is determined as the target content push strategy that matches the target user.

[0158] In some embodiments, a media asset display device is provided, which is used to: determine an interface identifier in response to an interface display operation, and send a data acquisition request to a server, wherein the data acquisition request carries the interface identifier and the user identifier of the target user; receive interface data returned by the server in response to the data acquisition request, wherein the interface data includes target media asset data, and the target media asset data is media asset data associated with a target content push strategy that matches the target user; display an interface corresponding to the interface identifier, and display the media asset data in the interface.

[0159] In some embodiments, the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above-mentioned policy determination method when the computer program is executed by a processor.

[0160] In some embodiments, the present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned policy determination method when executed by a processor.

[0161] In some embodiments, the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above-mentioned media resource display method when the computer program is executed by a processor.

[0162] In some embodiments, the present application further provides a computer program product, including a computer program, which implements the steps of the above-mentioned media asset display method when executed by a processor.

[0163] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and 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-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile memory and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), programmable logic devices (PLDs), data processing logic devices based on quantum computing, artificial intelligence (AI) processors, and the like.

[0164] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, 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, they should be considered to be within the scope of this application.

[0165] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A server, characterized in that: include: a communication device configured to be communicatively connected with the display device; and at least one processor, connected to the communication device and configured to: Obtaining a set of user identifiers, weighting and averaging a first indicator evaluation value and a first target weight associated with a first user identifier in the set of user identifiers to obtain a first comprehensive evaluation value, wherein the first indicator evaluation value is an evaluation value of the user represented by the first user identifier under a preset indicator after a target content push policy is set, the preset indicator being used to reflect the user's interest in target push content, the target push content being content pushed by the target content push policy; weighting and averaging the second indicator evaluation value and the second target weight associated with the second user identifier in the set of user identifiers to obtain a second comprehensive evaluation value, wherein the second indicator evaluation value is an evaluation value of the user represented by the second user identifier under the preset indicator when the target content push policy is not set, and the first target weight and the second target weight are used to balance the distribution of user characteristics associated with the first user identifier and the distribution of user characteristics associated with the second user identifier; The difference between the first comprehensive evaluation value and the second comprehensive evaluation value is used as an effectiveness evaluation value of the target content push strategy, where the effectiveness evaluation value is used to reflect the overall interest of the target user group in the target pushed content, where the target user group refers to the user group represented by the set of user identifiers; Obtaining a basic matching degree between the target content push strategy and the target user; The target matching degree between the target content push strategy and the target user is obtained by comprehensively considering the effect evaluation value and the basic matching degree. A target policy identifier is selected from the policy identifier set according to the size of the target matching degree, and the content push strategy corresponding to the target policy identifier is determined as the target content push strategy that matches the target user.

2. The server according to claim 1, wherein: The first target weight associated with each first user identifier is consistent, and the second target weight associated with each second user identifier is consistent, and the processor is further configured to: Determining the average of user features associated with a first user identifier in the set of user identifiers to obtain a first average user feature, and determining the average of user features associated with a second user identifier in the set of user identifiers to obtain a second average user feature; Determining a first product of the first average user feature and a first weight parameter, and determining a second product of the second average user feature and a second weight parameter; Minimizing the difference between the first product and the second product by adjusting the first weight parameter or the second weight parameter; The first weight parameter in the case of minimizing the difference is used as the first target weight, and the second weight parameter in the case of minimizing the difference is used as the second target weight.

3. The server according to claim 1, wherein: When the processor executes the step of using the difference between the first comprehensive evaluation value and the second comprehensive evaluation value as the effect evaluation value of the target content push strategy, the processor is configured to: The effect evaluation value of the target content push strategy is determined according to the following formula: Among them, s represents the target content push strategy, ATE(s) represents the effect evaluation value, represents the first comprehensive evaluation value, represents the second comprehensive evaluation value, w i is the target weight associated with the i-th user ID in the user ID set, t i is the indicator evaluation value associated with the i-th user ID in the user ID set. If the i-th user ID is the first user ID, then w i is the first target weight and t i is the evaluation value of the first indicator. If the i-th user identifier is the second user identifier, then w i is the second target weight and t i is the second indicator evaluation value, f i (s,1) represents the first user type label corresponding to the i-th user ID. If the i-th user ID is the first user ID, then f i The value of (s,1) is 1. If the i-th user ID is the second user ID, then f i The value of (s,1) is 0, f i (s,0) represents the second user type label corresponding to the i-th user ID. If the i-th user ID is the first user ID, then f i The value of (s,0) is 0. If the i-th user ID is the second user ID, then f i The value of (s,0) is 1, and n is the number of users in the target user group.

4. The server according to any one of claims 1 to 3, characterized in that: The user characteristics include attribute characteristics of a target attribute dimension. When the processor executes the step of obtaining a basic matching degree between the target content push strategy and the target user, the processor is configured to: Acquire multiple feature category labels corresponding to the target attribute dimension, where the feature category labels are used to distinguish different categories of attribute features of the target attribute dimension; Determining, from the user identifier set according to the first feature category label, a third user identifier corresponding to the first feature category label, wherein the first feature category label is any one of the multiple feature category labels, and the attribute feature of the target attribute dimension in the user feature associated with the third user identifier matches the first feature category label; Determining a degree of feature matching between the target content push strategy and the first feature category label based on the indicator evaluation value associated with the third user identifier, wherein the degree of feature matching is used to reflect the degree of interest of the user represented by the third user identifier in the target pushed content; Determining, from the plurality of feature category labels, a second feature category label that matches the attribute feature of the target attribute dimension in the user feature of the target user; According to the feature matching degree between the target content push strategy and the second feature category label, a basic matching degree between the target content push strategy and the target user is determined, wherein the basic matching degree is positively correlated with the feature matching degree.

5. The server according to claim 4, wherein: When the processor executes the step of obtaining a plurality of feature category labels corresponding to the target attribute dimension, the processor is configured to: If the attribute feature of the target attribute dimension is a discrete value, different attribute features of the target attribute dimension are encoded respectively, and the encoded values of different attribute features are used as feature category labels corresponding to the target attribute dimension; If the attribute feature of the target attribute dimension is a continuous numerical value, the user identifiers in the user identifier set are sorted according to the size of the attribute feature of the target attribute dimension to obtain a user identifier sequence, a reference user identifier is determined from the user identifier sequence, and a feature category label corresponding to the target attribute dimension is determined based on the attribute feature of the target attribute dimension in the user feature associated with the reference user identifier, wherein the proportion of the user identifiers in the subsequence from the first user identifier to the reference user identifier in the user identifier sequence in the user identifier sequence is equal to a preset proportion.

6. The server according to claim 4, wherein: When the processor executes the indicator evaluation value associated with the third user identifier to determine the degree of feature matching between the target content push strategy and the first feature category label, the processor is configured to: The proportion of the index evaluation value associated with the third user identifier in the total index evaluation value is used as the evaluation value proportion, and the total index evaluation value is the sum of the index evaluation values associated with the user identifiers in the user identifier set; Taking the proportion of the third user identifier in the set of user identifiers as the proportion of the number of users; The ratio of the evaluation value proportion to the user quantity proportion is used as the feature matching degree between the target content push strategy and the first feature category label.

7. The server according to claim 4, wherein: The user features include attribute features of multiple attribute dimensions, and there are multiple second feature category labels, and different second feature category labels are selected from feature category labels corresponding to different attribute dimensions; When the processor determines the basic matching degree between the target content push strategy and the target user based on the feature matching degree between the target content push strategy and the second feature category label, the processor is configured to: The feature matching degrees between the plurality of second feature category tags and the target content push strategy are comprehensively calculated to obtain a basic matching degree between the target content push strategy and the target user.

8. The server according to claim 7, wherein: When the processor executes the step of synthesizing the effect evaluation value and the basic matching degree to obtain the target matching degree between the target content push strategy and the target user, the processor is configured to: The target matching degree between the target content push strategy and the target user is determined according to the following formula: Among them, ρ(u,s) represents the target matching degree, u represents the target user, |X′| represents the number of attribute dimensions, X′ represents the set of attribute dimensions, and x j ′ represents the jth attribute dimension, Represents the second feature category label selected from multiple feature category labels corresponding to the j-th attribute dimension, represent The degree of feature matching with the target content push strategy, Represents the basic matching degree between the target content push strategy and the target user.

9. A display device, characterized in that: include: Display and controller; The controller is configured to: In response to the interface display operation, determining the interface identifier, and sending a data acquisition request to the server, wherein the data acquisition request carries the interface identifier and the user identifier of the target user; Receiving interface data returned by the server in response to the data acquisition request, wherein the interface data includes target media asset data, and the target media asset data is media asset data associated with a target content push strategy that matches the target user; Controlling the display to display an interface corresponding to the interface identifier, and displaying the media asset data in the interface; Wherein, the target content push strategy is determined by the server described in any one of claims 1 to 7 above.

10. A strategy determination method, characterized in that: Applied to a server, the method includes: Obtaining a set of user identifiers, weighting and averaging a first indicator evaluation value and a first target weight associated with a first user identifier in the set of user identifiers to obtain a first comprehensive evaluation value, wherein the first indicator evaluation value is an evaluation value of the user represented by the first user identifier under a preset indicator after the target content push policy is set, and the preset indicator is used to reflect the user's interest in the target pushed content, and the target pushed content refers to the content pushed by the target content push policy; weighting and averaging the second indicator evaluation value and the second target weight associated with the second user identifier in the set of user identifiers to obtain a second comprehensive evaluation value, wherein the second indicator evaluation value is an evaluation value of the user represented by the second user identifier under the preset indicator when the target content push policy is not set, and the first target weight and the second target weight are used to balance the distribution of user characteristics associated with the first user identifier and the distribution of user characteristics associated with the second user identifier; The difference between the first comprehensive evaluation value and the second comprehensive evaluation value is used as an effectiveness evaluation value of the target content push strategy, where the effectiveness evaluation value is used to reflect the overall interest of the target user group in the target pushed content, where the target user group refers to the user group represented by the set of user identifiers; Obtaining a basic matching degree between the target content push strategy and the target user; The target matching degree between the target content push strategy and the target user is obtained by comprehensively considering the effect evaluation value and the basic matching degree. A target policy identifier is selected from the policy identifier set according to the size of the target matching degree, and the content push strategy corresponding to the target policy identifier is determined as the target content push strategy that matches the target user.

Images