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

Abstract

The invention relates to a content recommendation method and device, equipment, a storage medium and a program product. The method comprises the following steps: after a content recommendation event is detected, receiving a wireless signal sent by a remote controller through an antenna array; wherein the wireless signal carries attitude data of the remote controller; determining the spatial orientation of the remote controller relative to the media playing equipment according to the phase difference of the wireless signals received by the antennas in the antenna array; according to the attitude data, orientation information of the remote controller is determined; determining a target spatial position representation of a user using the remote controller according to the spatial orientation, the orientation information and the signal intensity of the wireless signal received by the antenna array; and according to the target spatial position representation, content recommendation is performed to the user. By adopting the method, accurate recommendation of the content can be realized.

Description

Technical Field

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

[0002] Smart TVs and Internet Protocol Television (IPTV) are often used by multiple family members at home, and people of different ages and genders have very different preferences for TV content. During the cold start phase of the content recommendation system, the lack of effective user profiles and behavioral data leads to inaccurate recommendations, impacting user experience.

[0003] Traditional page recommendation strategies rely on users actively logging in, filling in information, or accumulating historical long-term behaviors to make recommendations based on user information. The recommendation process is fixed, and the recommendation is based on a single dimension. There is a lack of targeting when making page recommendations for different users, resulting in low recommendation accuracy and efficiency. Summary of the Invention

[0004] Based on this, it is necessary to provide a content recommendation method, device, communication equipment, storage medium and program product to address the above technical problems, which can accurately and efficiently recommend content.

[0005] In a first aspect, the present application provides a content recommendation method, comprising:

[0006] After detecting a content recommendation event, receiving a wireless signal sent by a remote controller through an antenna array; wherein the wireless signal carries the gesture data of the remote controller;

[0007] determining a spatial orientation of the remote controller relative to the media playback device based on a phase difference of the wireless signal received by each antenna in the antenna array; and

[0008] determining orientation information of the remote controller according to the posture data;

[0009] determining a target spatial position representation of a user using the remote controller based on the spatial orientation, the orientation information, and a signal strength of a wireless signal received by the antenna array;

[0010] Recommending content to the user based on the target spatial position representation.

[0011] In one embodiment, determining a target spatial position representation of a user using the remote control based on the spatial orientation, the orientation information, and the signal strength of the wireless signal received by the antenna array includes:

[0012] Normalizing the signal intensity to obtain a normalized intensity;

[0013] concatenating the spatial orientation, the orientation information, and the normalized intensity to obtain an initial feature representation;

[0014] The initial feature representation is standardized using standardized parameters to obtain a target spatial position representation of the user using the remote control; wherein the standardized parameters include a mean value and a standard deviation; the mean value and the standard deviation are determined based on the sample spatial position representation.

[0015] In one embodiment, the concatenating the spatial orientation, the orientation information, and the normalized intensity to obtain an initial feature representation includes:

[0016] performing weighted processing on the spatial orientation and the orientation information respectively to obtain weighted spatial orientation and weighted orientation information;

[0017] The weighted spatial orientation, the weighted orientation information, and the normalized intensity are concatenated to obtain an initial feature representation.

[0018] In one embodiment, recommending content to the user based on the target spatial position representation includes:

[0019] selecting a target location cluster from different candidate location clusters based on the distances between the target spatial location representation and different candidate location clusters; wherein the different candidate location clusters are obtained by clustering the sample spatial location representations, and the distance between the target location cluster and the target spatial location representation is the shortest;

[0020] Recommending content to the user based on the first content tag corresponding to the target location cluster.

[0021] In one embodiment, recommending content to the user based on the first content tag corresponding to the target location cluster includes:

[0022] determining a second content tag based on the user's historical behavior data;

[0023] Obtaining an extended content tag for the user;

[0024] For each content tag, determining a recommendation score for the content tag based on a probability that the content tag originates from the target location cluster, a probability that the content tag originates from the historical behavior data, and a probability that the content tag originates from preset data;

[0025] A target content tag is selected from each content tag according to the recommendation score corresponding to each content tag, and content is recommended to the user based on the target content tag; wherein each content tag includes the extended content tag, the first content tag, and the second content tag.

[0026] In one embodiment, determining the recommendation score of the content tag based on the probability that the content tag originates from the target location cluster, the probability that the content tag originates from the historical behavior data, and the probability that the content tag originates from preset data includes:

[0027] Using a first weight, weighting the probability that the content tag originates from the target location cluster to obtain a first weighted probability;

[0028] Using a second weight, weighting the probability that the content tag is derived from the historical behavior data to obtain a second weighted probability;

[0029] Using a third weight, weighting the probability that the content tag is derived from preset data to obtain a third weighted probability;

[0030] A recommendation score for the content tag is determined according to the sum of the first weighted probability, the second weighted probability, and the third weighted probability.

[0031] In a second aspect, the present application further provides a content recommendation device, comprising:

[0032] A signal sending module, configured to receive a wireless signal sent by a remote controller via an antenna array after detecting a content recommendation event; wherein the wireless signal carries gesture data of the remote controller;

[0033] an information determination module, configured to determine a spatial orientation of the remote control relative to the media playback device based on a phase difference of the wireless signal received by each antenna in the antenna array; and to determine orientation information of the remote control based on the posture data;

[0034] a position determination module, configured to determine a target spatial position representation of a user using the remote controller based on the spatial orientation, the orientation information, and the signal strength of the wireless signal received by the antenna array;

[0035] A content recommendation module is used to recommend content to the user based on the target spatial position representation.

[0036] In a third aspect, the present application further provides a communication device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0037] After detecting a content recommendation event, receiving a wireless signal sent by a remote controller through an antenna array; wherein the wireless signal carries the gesture data of the remote controller;

[0038] determining a spatial orientation of the remote controller relative to the media playback device based on a phase difference of the wireless signal received by each antenna in the antenna array; and

[0039] determining orientation information of the remote controller according to the posture data;

[0040] determining a target spatial position representation of a user using the remote controller based on the spatial orientation, the orientation information, and a signal strength of a wireless signal received by the antenna array;

[0041] Recommending content to the user based on the target spatial position representation.

[0042] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0043] After detecting a content recommendation event, receiving a wireless signal sent by a remote controller through an antenna array; wherein the wireless signal carries the gesture data of the remote controller;

[0044] determining a spatial orientation of the remote controller relative to the media playback device based on a phase difference of the wireless signal received by each antenna in the antenna array; and

[0045] determining orientation information of the remote controller according to the posture data;

[0046] determining a target spatial position representation of a user using the remote controller based on the spatial orientation, the orientation information, and a signal strength of a wireless signal received by the antenna array;

[0047] Recommending content to the user based on the target spatial position representation.

[0048] In a fifth aspect, the present application further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the following steps:

[0049] After detecting a content recommendation event, receiving a wireless signal sent by a remote controller through an antenna array; wherein the wireless signal carries the gesture data of the remote controller;

[0050] determining a spatial orientation of the remote controller relative to the media playback device based on a phase difference of the wireless signal received by each antenna in the antenna array; and

[0051] determining orientation information of the remote controller according to the posture data;

[0052] determining a target spatial position representation of a user using the remote controller based on the spatial orientation, the orientation information, and a signal strength of a wireless signal received by the antenna array;

[0053] Recommending content to the user based on the target spatial position representation.

[0054] The above-mentioned content recommendation method, apparatus, device, storage medium and program product, after detecting a content recommendation event, receives a wireless signal carrying the posture data of the remote control sent by the remote control through the antenna array, and determines the spatial orientation of the remote control relative to the media playback device based on the phase difference of the wireless signal received by each antenna in the antenna array, and determines the orientation information of the remote control based on the posture data, thereby ensuring the accuracy of the remote control orientation when the user uses the remote control; further, based on the spatial orientation, orientation information and the signal strength of the wireless signal received by the antenna array, determines the target spatial position representation of the user using the remote control, and recommends content to the user based on the target spatial position representation, which is equivalent to binding the user's habit of using the remote control with the user's preferences, thereby ensuring the pertinence and accuracy of the content recommended to the user, and improving the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. 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 paying any creative work.

[0056] Figure 1 A diagram illustrating an application environment of a content recommendation method according to an embodiment;

[0057] Figure 2 1 is a flow chart of a content recommendation method according to an embodiment;

[0058] Figure 3 A schematic diagram of a process for determining a target spatial position representation in one embodiment;

[0059] Figure 4 A schematic diagram of a process for determining an initial feature representation in one embodiment;

[0060] Figure 5 A schematic diagram of a process for recommending content to a user in one embodiment;

[0061] Figure 6A An application scenario diagram provided in one embodiment;

[0062] Figure 6B A schematic diagram of candidate location clusters provided for one embodiment;

[0063] Figure 7 A schematic diagram of a process for recommending content to a user in another embodiment;

[0064] Figure 8 is a flowchart of a content recommendation method in another embodiment;

[0065] Figure 9 is a structural block diagram of a content recommendation device in one embodiment;

[0066] Figure 10 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0067] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0068] The content recommendation method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the media playback device 101 includes an antenna array for receiving wireless signals sent by the remote control 102 through the Bluetooth module. The media playback device 101 can be a smart TV or a TV set-top box. The remote control 102 is a handheld remote control for controlling the media playback device 101. Optionally, after the media playback device 101 detects a content recommendation event, it receives a wireless signal carrying the posture data of the remote control sent by the remote control 102 through the antenna array, and determines the spatial orientation of the remote control relative to the media playback device based on the phase difference of the wireless signal received by each antenna in the antenna array, and determines the orientation information of the remote control based on the posture data; further, the media playback device 101 determines the target spatial position representation of the user using the remote control based on the spatial orientation, orientation information and signal strength of the wireless signal received by the antenna array; and recommends content to the user based on the target spatial position representation.

[0069] In an exemplary embodiment, Figure 2 As shown, a content recommendation method is provided, which is described by taking the method applied to a media playback device 101 as an example, and specifically includes the following steps:

[0070] S201: After detecting a content recommendation event, a wireless signal sent by a remote controller is received via an antenna array.

[0071] A content recommendation event is an event that requires recommending content to a user. The wireless signal carries the remote control's posture data. A remote control is a device that controls a media player. Posture data refers to digital information about the remote control's physical characteristics, such as its position, orientation, and motion state in space. This posture data can be measured using a measurement device within the remote control, such as a magnetometer, accelerometer, or gyroscope. This posture data includes, but is not limited to, angular velocity, acceleration, and geomagnetic intensity.

[0072] In the embodiment of the present application, the triggering method of the content recommendation event may be to recognize the power-on signal of the control device (such as a remote control) of the media playback device, the signal of opening the content recommendation page of the media playback device for the first time, the signal of refreshing the content recommendation page, etc. For example, the content recommendation event may be a user clicking the power button of the remote control, entering the hot column, or scrolling the page to update, etc.

[0073] After the media playback device detects the content recommendation time, the measuring device in the remote control measures the angular velocity, acceleration and geomagnetic intensity of the remote control; further, the inertial measurement unit in the remote control can measure the Euler angle rotation matrix and direction vector.

[0074] S202 : determining the spatial orientation of the remote control relative to the media player device based on the phase difference of the wireless signals received by each antenna in the antenna array, and determining the orientation information of the remote control based on the posture data.

[0075] Among them, the spatial orientation of the remote control relative to the media playback device represents the direction and position relationship of the remote control relative to the media playback device in three-dimensional space; the orientation information of the remote control represents the direction in which the remote control points in three-dimensional space, which can be specifically represented by a direction vector.

[0076] It should be noted that when the same signal reaches different antennas in the antenna array, phase differences will occur due to differences in path length. Therefore, after receiving the wireless signal sent by the remote control, the antenna array calculates the azimuth and elevation of the wireless signal based on the phase differences between the wireless signal reaching each antenna, and obtains the received wireless signal strength. Based on the azimuth and elevation, the spatial orientation of the remote control relative to the media player is calculated, which can be expressed as (x1, y1, z1). Specifically,

[0077]

[0078] Where θ is the azimuth angle, is the pitch angle.

[0079] Furthermore, the inertial measurement unit (IMU) can be used to calculate the remote control's Euler rotation matrix based on the attitude data. The Euler rotation matrix is ​​a set of parameters used to describe the orientation of a rigid body in three-dimensional space. The object's attitude is defined by rotations around specific coordinate axes. The direction vector is then calculated from the Euler rotation matrix to obtain the remote control's orientation information, which can be expressed as (x², y², z²).

[0080] S203 : Determine a target spatial position representation of the user using the remote controller according to the spatial pointing and orientation information and the signal strength of the wireless signal received by the antenna array.

[0081] The wireless signal strength represents the strength of the wireless signal, typically expressed in dBm. Wireless signal strength generally decreases with distance and is significantly affected by obstructions, multipath, and interference. The target spatial position represents the user's specific location relative to the media player device in three-dimensional space and their remote control usage habits.

[0082] Optionally, a position prediction model can be pre-trained, where the input of the position prediction model is the spatial pointing and orientation information and the signal strength of the wireless signal received by the antenna array, and the output is a representation of the target spatial position of the user using the remote control. Furthermore, the spatial pointing and orientation information and the signal strength of the wireless signal received by the antenna array can be input into the position prediction model, so that the position prediction model analyzes and calculates the spatial pointing and orientation information and the signal strength of the wireless signal received by the antenna array based on the model parameters, and outputs a representation of the target spatial position of the user using the remote control.

[0083] S204: Recommend content to the user based on the target spatial position representation.

[0084] Optionally, the target spatial position representation can reflect the remote control user's usage habits to a certain extent. Furthermore, the media player's database can pre-store learned user preference characteristics, i.e., the content preferences corresponding to different user IDs. Furthermore, the target spatial position representation can, to a certain extent, locate the remote control user ID. In this case, the user's preferred content can be located based on the user ID.

[0085] Furthermore, based on the user's preferred content, content that matches the preferred content, such as videos, programs, posters, etc., is selected from the database and then formatted and displayed on the page.

[0086] In the above-mentioned content recommendation method, after a content recommendation event is detected, a wireless signal carrying the posture data of the remote control sent by the remote control is received through the antenna array, and the spatial orientation of the remote control relative to the media playback device is determined based on the phase difference of the wireless signals received by each antenna in the antenna array, and the orientation information of the remote control is determined based on the posture data, thereby ensuring the accuracy of the orientation of the remote control when the user uses the remote control; further, based on the spatial orientation, orientation information and signal strength of the wireless signal received by the antenna array, the target spatial position representation of the user using the remote control is determined, and based on the target spatial position representation, content is recommended to the user, which is equivalent to binding the user's habit of using the remote control with the user's preferences, thereby ensuring the pertinence and accuracy of the content recommended to the user, and improving the user experience.

[0087] Optionally, in an exemplary embodiment, as Figure 3 As shown, a method for determining the target spatial position representation of a user using a remote controller is provided to refine S103 in the above embodiment, specifically comprising the following steps:

[0088] S301 , normalizing the signal intensity to obtain normalized intensity.

[0089] Optionally, the maximum and minimum values ​​of the signal intensity may be determined first, and the signal intensity may be normalized according to the maximum and minimum values ​​using a linear normalization or logarithmic normalization method, so that the signal intensity is mapped to a uniform interval to obtain a normalized intensity.

[0090] S302: Concatenate the spatial orientation, direction information, and normalized intensity to obtain an initial feature representation.

[0091] In the embodiment of the present application, both spatial orientation and orientation information can be expressed in the form of coordinates. On this basis, the spatial orientation, orientation information, and normalized intensity can be spliced ​​according to the prediction rules and order to obtain a feature vector as the initial feature representation. For example, if the spatial orientation is represented as (x1, y1, z1) and the orientation information is represented as (x2, y2, z2), the initial feature representation can be:

[0092] Feature=[x1,y1,z1,x2,y2,z2,n rssi ] (4)

[0093] Among them, n rssi is the normalized intensity.

[0094] S303: Using standardized parameters, the initial feature representation is standardized to obtain a target spatial position representation of the user using the remote control.

[0095] Among them, the standardized parameters include mean and standard deviation; mean and standard deviation are determined according to the sample space position representation.

[0096] It is understandable that the features in the initial feature representation may be in different feature dimensions, so the initial feature representation needs to be standardized to eliminate the influence of different feature dimensions and value ranges so that the data conforms to the standard normal distribution.

[0097] Optionally, the mean and standard deviation of the initial feature representation can be calculated. The standard deviation reflects the degree of data discreteness of the initial feature representation. Furthermore, the initial feature representation is linearly transformed using the mean and standard deviation to obtain the target spatial position representation after the transformation.

[0098] In this embodiment, the initial feature representation is obtained by splicing the spatial pointing, orientation information and normalized intensity, which ensures that the initial feature representation can fully reflect the position and orientation of the remote control. The initial feature representation is then standardized to ensure that the dimensions of the eigenvalues ​​in the obtained target spatial position representation are the same, which facilitates subsequent calculations.

[0099] Optionally, to enhance the initial feature representation, in one embodiment, as Figure 4 As shown, a method for determining an initial feature representation is provided to refine S202 in the above embodiment, specifically comprising the following steps:

[0100] S401 , performing weighted processing on the spatial orientation and orientation information respectively to obtain weighted spatial orientation and weighted orientation information.

[0101] Optionally, the weights of the spatial pointing and orientation information parameters can be determined in advance based on their importance, and then the spatial pointing and orientation information can be weighted according to their respective weights to obtain weighted spatial pointing and weighted orientation information. For example, if the spatial pointing is represented by (x1, y1, z1) and the weight of the spatial pointing is a, then the weighted spatial pointing is represented by (a*x1, a*x1y1, a*x1z1); the orientation information is represented by (x2, y2, z2) and the weight of the orientation information is b, then the weighted orientation information is represented by (bx2, by2, bz2).

[0102] S402: Concatenate the weighted spatial orientation, weighted direction information, and normalized intensity to obtain an initial feature representation.

[0103] Optionally, the weighted spatial orientation, weighted orientation information and normalized intensity may be directly concatenated, or the weighted spatial orientation, weighted orientation information and normalized intensity may be concatenated in a preset order to obtain an initial feature representation.

[0104] In this embodiment, by weighting the spatial orientation and orientation information respectively and concatenating the weighted spatial orientation, weighted orientation information and normalized intensity, the features of the obtained initial feature representation are enhanced and the accuracy of the obtained initial feature representation is ensured.

[0105] Optionally, in one embodiment, Figure 5 As shown, a method for recommending content to a user is provided, which specifically includes the following steps:

[0106] S501 : Selecting a target position cluster from different candidate position clusters according to the distances between the target spatial position representation and different candidate position clusters.

[0107] A location cluster is a cluster of the spatial locations of multiple users. Spatial locations within the same location cluster share common characteristics or represent the same user. Different candidate location clusters are obtained by clustering sample spatial location representations, with the target location cluster and the target spatial location representation being the shortest distance apart. It's understandable that different users have different usage habits and heights, leading to significant differences in remote control pointing and distance. Therefore, by clustering sample spatial location representations, each candidate location cluster can, to some extent, represent a user's usage habits.

[0108] It should be noted that a deep embedding clustering model can be used to cluster the sample space position representation. First, sample data needs to be collected, and the sample data includes the sample space position representation and the content label corresponding to the sample space position representation. The user's historical behavior data and the recommended content corresponding to the user's historical behavior data can be collected from the user's historical operation records on the media playback device. Furthermore, based on the method provided in the above embodiment, a sample space position representation is constructed according to the historical behavior data, and the recommended content corresponding to the sample space position representation is used as a label. At this point, the sample space position representation can be used to train the pre-trained deep embedding clustering model to obtain a trained deep embedding clustering model.

[0109] In addition, in order to ensure the accuracy of the recommended content used as a label, the user's operation data based on the recommended content can also be obtained to filter the recommended content.

[0110] For example, Figure 6AAs shown, for a certain media playback device (such as IPTV (Internet Protocol Television, interactive network television) / smart TV), users who frequently use the media playback device include user 1, user 2 and user 3. User 1 (the hostess) is accustomed to operating the remote control on sofa 1, holding the remote control at a height of 1 meter, and the front is perpendicular to the TV Bluetooth receiver; user 2 (the host) is accustomed to operating the remote control on sofa 2, holding the remote control at a height of 1.5 meters, and tilted 45 degrees to the right side of the TV Bluetooth receiver; user 3 (the child) is accustomed to operating the remote control on sofa 2, holding the remote control at a height of 0.75 meters, and tilted 25 degrees to the right side of the TV Bluetooth receiver. Clustering the sample spatial position representations corresponding to the media player device yields three candidate position clusters, namely, user 1 = [0.0, 1.0, 0.0, 0.0, 1.0, 0.0, 1.0] (directly in front, at the same height, 1m away), user 2 = [0.59, 0.59, 0.42, 0.6, 0.8, 0.0, 0.0] (45° to the right, 0.5m high, 1.3m away), and user 3 = [0.41, 0.89, -0.20, 0.3, 0.95, 0.0, 0.33] (25° to the right, 0.25m low, 1.2m away). It is assumed that the Bluetooth receiver of the media player device is at the coordinate origin (0, 0, 0). Figure 6B The figure shows a schematic diagram of three candidate location clusters obtained by clustering.

[0111] On this basis, the distance between the target spatial position representation and each candidate position cluster can be calculated. For example, the Euclidean distance between the target spatial position and each candidate position cluster can be calculated, and the candidate position cluster with the shortest distance to the target spatial position representation is used as the target position cluster.

[0112] S502 : Recommend content to the user based on the first content tag corresponding to the target location cluster.

[0113] The content tags represent the content recommended to the user, for example, content tags include but are not limited to cartoons, early childhood education, children's songs, sports, shopping, etc. It should be noted that each target location cluster corresponds to at least one content tag.

[0114] Optionally, based on the first content tag corresponding to the target location cluster, recommended content, such as videos and pictures, matching the first content tag can be selected from a database of the media player. Furthermore, the selected recommended content is arranged and displayed on a display screen of the media player to recommend content to the user.

[0115] In this embodiment, by introducing candidate location clusters and selecting target location clusters based on the distances between the target spatial location representation and different candidate location clusters, it is ensured that the target location cluster is closest to the characteristics of the target spatial location representation, thereby ensuring that the determined first content label is closer to user needs, and ensuring the targeted and efficient content recommendation for users.

[0116] Optionally, in one embodiment, Figure 7 As shown, a method for recommending content to a user is provided to refine S502 in the above embodiment, specifically comprising the following steps:

[0117] S701: Determine a second content tag based on the user's historical behavior data.

[0118] The user's historical behavior data represents the user's interaction behavior with the media playback device in the past, and the second content tag represents the content that the user prefers to watch.

[0119] Optionally, feature extraction can be performed on the historical behavior data, and a neural network model can be used to extract the page content that the user frequently browses or prefers to browse based on the extracted features. Further, a second content tag can be generated based on the extracted page content that the user frequently browses or prefers to browse.

[0120] S702: Obtain an extended content tag for the user.

[0121] Among them, extended content tags are tags for other content that can be recommended to users. This content can be content that is predicted to be preferred by the user, or content that is currently popular or has a high click-through rate. In embodiments of the present application, extended content tags can be predicted based on preset data. The preset data can include historical behavior data of other users similar to the user corresponding to the media playback device, and data with high click-through rates or popularity in the current time period. For example, if a topic is currently popular and has a high click-through rate, the user's behavior data for videos or images related to that topic can be used as the preset data.

[0122] Optionally, based on the preference prediction model, preset data may be input into the preference prediction model, so that the preference prediction model processes the preset data based on preset model parameters to predict content tags that the user may prefer as extended content tags.

[0123] S703 , for each content tag, determine a recommendation score for the content tag based on the probability that the content tag originates from the target location cluster, the probability that the content tag originates from historical behavior data, and the probability that the content tag originates from preset data.

[0124] In order to recommend content in a more targeted and comprehensive manner, for each content tag, it is necessary to calculate the probability that the content tag comes from different data and calculate the recommendation score for each content tag.

[0125] For a target location cluster, if there is at least one content tag corresponding to the target location cluster, if the content tag is not the content tag corresponding to the target location cluster, the probability that the content tag originated from the target location cluster is set to 0. If the content tag is the content tag corresponding to the target location cluster, the frequency of occurrence of the content tag and its related tags in the target location cluster is calculated to calculate the probability that the content tag originated from the target location cluster. Accordingly, the same probability calculation method can be used for historical behavior data and preset data.

[0126] Furthermore, the recommendation score of the content tag may be the sum of the probability that the content tag originates from the target location cluster, the probability that the content tag originates from historical behavior data, and the probability that the content tag originates from preset data.

[0127] In an optional embodiment, in order to ensure the accuracy of the determined recommendation score, the probability that the content tag originates from the target location cluster, the probability that the content tag originates from historical behavior data, and the probability that the content tag originates from preset data can be weighted and summed to calculate the recommendation score of the content tag. For example, a first weight is used to weight the probability that the content tag originates from the target location cluster to obtain a first weighted probability; a second weight is used to weight the probability that the content tag originates from historical behavior data to obtain a second weighted probability; a third weight is used to weight the probability that the content tag originates from preset data to obtain a third weighted probability; and the recommendation score of the content tag is determined based on the sum of the first weighted probability, the second weighted probability, and the third weighted probability. The first weight, the second weight, and the third weight can be set according to the importance of the target location cluster, the historical behavior data, and the preset data.

[0128] S704 , selecting a target content tag from each content tag according to the recommendation score corresponding to each content tag, and recommending content to the user based on the target content tag.

[0129] Each content tag includes an extended content tag, a first content tag, and a second content tag.

[0130] Optionally, a score threshold can be preset. For example, if the recommendation score ranges from 0 to 100, the score can be preset to 70. Furthermore, the recommendation score corresponding to each content tag is compared with the score threshold, and the content tag with a recommendation score greater than the score threshold is used as the target content tag.

[0131] After selecting the target content tag, the target content tag can be input into the page recommendation engine, so that the page recommendation engine selects corresponding videos, pictures, channels, etc. from the database according to the target content tag for typeset and generates a recommendation page.

[0132] In this embodiment, by introducing the second content tag and the extended content tag, the content tags recommended to users are comprehensively expanded. Furthermore, by calculating the recommendation score for each content tag and selecting the target content tag based on the recommendation score, it is ensured that the content corresponding to the selected target content tag is more in line with the user's preferences and habits.

[0133] Figure 8 FIG. 1 is a flow chart of a content recommendation method in another embodiment. Based on the above embodiment, this embodiment provides an optional example of a content recommendation method. Figure X The specific implementation process is as follows:

[0134] S801: After a content recommendation event is detected, a wireless signal carrying gesture data of the remote controller is received via an antenna array.

[0135] S802: Determine the spatial orientation of the remote control relative to the media playback device based on the phase difference of the wireless signal received by each antenna in the antenna array.

[0136] S803: Determine the orientation information of the remote controller according to the posture data.

[0137] S804: normalize the signal intensity to obtain normalized intensity.

[0138] S805: Concatenate the spatial orientation, direction information, and normalized intensity to obtain an initial feature representation.

[0139] Optionally, the spatial pointing and orientation information are weighted respectively to obtain weighted spatial pointing and weighted orientation information; the weighted spatial pointing, weighted orientation information and normalized intensity are concatenated to obtain an initial feature representation.

[0140] S806: Using standardized parameters, the initial feature representation is standardized to obtain a target spatial position representation of the user using the remote control.

[0141] Among them, the standardized parameters include mean and standard deviation; mean and standard deviation are determined according to the sample space position representation.

[0142] S807 : Select a target position cluster from different candidate position clusters according to the distances between the target spatial position representation and different candidate position clusters.

[0143] Among them, different candidate position clusters are obtained by clustering the sample spatial position representation, and the distance between the target position cluster and the target spatial position representation is the shortest.

[0144] S808: Determine a second content tag based on the user's historical behavior data.

[0145] S809: Obtain an extended content tag for the user.

[0146] S810 , determining a recommendation score for each content tag based on the probability that each content tag originates from the target location cluster, the probability that each content tag originates from historical behavior data, and the probability that the nursery rhyme content tag originates from preset data.

[0147] Optionally, a first weight is used to weight the probability that the content tag originates from the target location cluster to obtain a first weighted probability; a second weight is used to weight the probability that the content tag originates from historical behavior data to obtain a second weighted probability; a third weight is used to weight the probability that the content tag originates from preset data to obtain a third weighted probability; and a recommendation score for the content tag is determined based on the sum of the first weighted probability, the second weighted probability, and the third weighted probability.

[0148] S811 , selecting a target content tag from each content tag according to the recommendation score corresponding to each content tag, and recommending content to the user based on the target content tag.

[0149] Each content tag includes an extended content tag, a first content tag, and a second content tag.

[0150] The specific process of the above S801-S811 can be found in the description of the above method embodiment. The implementation principle and technical effects are similar and will not be repeated here.

[0151] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence 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 executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0152] Based on the same inventive concept, embodiments of the present application also provide a content recommendation device for implementing the aforementioned content recommendation method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more content recommendation device embodiments provided below can be found in the above-mentioned limitations of the content recommendation method and will not be further elaborated here.

[0153] In an exemplary embodiment, Figure 9 As shown, a content recommendation device 900 is provided, comprising: a signal sending module 910, an information determination module 920, a position determination module 930 and a content recommendation module 940, wherein:

[0154] The signal sending module 910 is used to receive a wireless signal sent by the remote control through the antenna array after detecting a content recommendation event; wherein the wireless signal carries the gesture data of the remote control.

[0155] The information determination module 920 is configured to determine the spatial orientation of the remote control relative to the media player device based on the phase difference of the wireless signals received by each antenna in the antenna array; and to determine the orientation information of the remote control based on the posture data.

[0156] The position determination module 930 is configured to determine a target spatial position representation of a user using a remote controller based on the spatial orientation and direction information and the signal strength of the wireless signal received by the antenna array.

[0157] The content recommendation module 940 is used to recommend content to the user based on the target spatial position representation.

[0158] The above-mentioned content recommendation device receives the wireless signal carrying the posture data of the remote control sent by the remote control through the antenna array, and determines the spatial orientation of the remote control relative to the media playback device based on the phase difference of the wireless signals received by each antenna in the antenna array, and determines the orientation information of the remote control based on the posture data, thereby ensuring the accuracy of the remote control orientation when the user uses the remote control; further, based on the spatial orientation, orientation information and the signal strength of the wireless signal received by the antenna array, the target spatial position representation of the user using the remote control is determined, and content is recommended to the user based on the target spatial position representation, which is equivalent to binding the user's habit of using the remote control with the user's preferences, thereby ensuring the pertinence and accuracy of the content recommended to the user, and improving the user experience.

[0159] In one embodiment, the location determination module 930 includes:

[0160] The normalization unit is used to perform normalization processing on the signal intensity to obtain a normalized intensity.

[0161] The feature splicing unit is used to splice the spatial orientation, orientation information and normalized intensity to obtain the initial feature representation.

[0162] The standardization unit is used to standardize the initial feature representation using standardization parameters to obtain a target spatial position representation of the user using the remote control; wherein the standardization parameters include a mean value and a standard deviation; the mean value and the standard deviation are determined based on the sample spatial position representation.

[0163] In one embodiment, the feature splicing unit is specifically configured to:

[0164] The spatial orientation and orientation information are weighted respectively to obtain weighted spatial orientation and weighted orientation information; the weighted spatial orientation, weighted orientation information and normalized intensity are concatenated to obtain the initial feature representation.

[0165] In one embodiment, the content recommendation module 940 includes:

[0166] The selection unit is used to select a target position cluster from different candidate position clusters based on the distance between the target spatial position representation and different candidate position clusters; wherein the different candidate position clusters are obtained by clustering the sample spatial position representation, and the distance between the target position cluster and the target spatial position representation is the shortest.

[0167] The content recommendation unit is configured to recommend content to the user based on the first content tag corresponding to the target location cluster.

[0168] In one embodiment, the content recommendation unit includes:

[0169] The tag determination subunit is configured to determine a second content tag based on the user's historical behavior data.

[0170] The tag acquisition subunit is used to obtain the extended content tag for the user.

[0171] The score determination subunit is used to determine the recommendation score of each content tag based on the probability that the content tag comes from the target location cluster, the probability that the content tag comes from historical behavior data, and the probability that the content tag comes from preset data.

[0172] The content recommendation subunit is used to select a target content tag from each content tag based on the recommendation score corresponding to each content tag, and recommend content to the user based on the target content tag; wherein each content tag includes an extended content tag, a first content tag, and a second content tag.

[0173] In one embodiment, the score determination subunit is specifically configured to:

[0174] A first weight is used to weight the probability that the content tag originates from the target location cluster to obtain a first weighted probability; a second weight is used to weight the probability that the content tag originates from the historical behavior data to obtain a second weighted probability; a third weight is used to weight the probability that the content tag originates from the preset data to obtain a third weighted probability; and a recommendation score for the content tag is determined based on the sum of the first weighted probability, the second weighted probability, and the third weighted probability.

[0175] Each module in the aforementioned content recommendation device may be implemented in whole or in part through software, hardware, or a combination thereof. Each module may be embedded in or independent of a processor in a computer device in the form of hardware, or may be stored in a memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0176] In one embodiment, a communication device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 10 As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a content recommendation method is implemented.

[0177] Those skilled in the art will understand that Figure 10 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0178] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0179] After detecting a content recommendation event, the wireless signal sent by the remote control is received through the antenna array; wherein the wireless signal carries the gesture data of the remote control;

[0180] determining a spatial orientation of the remote control relative to the media playback device based on a phase difference of wireless signals received by each antenna in the antenna array; and

[0181] Determine the orientation of the remote control based on the posture data;

[0182] Determining a target spatial position representation of a user using a remote control based on the spatial pointing and orientation information and the signal strength of the wireless signal received by the antenna array;

[0183] Recommend content to users based on the target spatial location representation.

[0184] In one embodiment, when the processor executes the computer program to determine the target spatial position representation of the user using the remote control based on the spatial pointing and orientation information and the signal strength of the wireless signal received by the antenna array, the processor further implements the following steps:

[0185] The signal strength is normalized to obtain normalized intensity; the spatial pointing, orientation information and normalized intensity are concatenated to obtain an initial feature representation; the initial feature representation is normalized using standardized parameters to obtain a target spatial position representation of a user using a remote control; wherein the standardized parameters include a mean value and a standard deviation; the mean value and the standard deviation are determined based on the sample spatial position representation.

[0186] In one embodiment, when the processor executes the computer program to combine the spatial orientation, the orientation information, and the normalized intensity to obtain the initial feature representation, the processor further implements the following steps:

[0187] The spatial orientation and orientation information are weighted respectively to obtain weighted spatial orientation and weighted orientation information; the weighted spatial orientation, weighted orientation information and normalized intensity are concatenated to obtain the initial feature representation.

[0188] In one embodiment, when the processor executes the computer program to recommend content to the user based on the target spatial position representation, the processor further implements the following steps:

[0189] A target location cluster is selected from different candidate location clusters based on the distance between the target spatial location representation and different candidate location clusters; wherein the different candidate location clusters are obtained by clustering the sample spatial location representations, and the distance between the target location cluster and the target spatial location representation is the shortest; content is recommended to the user based on the first content tag corresponding to the target location cluster.

[0190] In one embodiment, when the processor executes the computer program to recommend content to the user based on the first content tag corresponding to the target location cluster, the processor further implements the following steps:

[0191] Determine a second content tag based on the user's historical behavior data; obtain extended content tags for the user; determine a recommendation score for each content tag based on the probability that the content tag originates from the target location cluster, the probability that the content tag originates from the historical behavior data, and the probability that the content tag originates from the preset data; select a target content tag from each content tag based on the recommendation score corresponding to each content tag, and recommend content to the user based on the target content tag; wherein each content tag includes an extended content tag, a first content tag, and a second content tag.

[0192] In one embodiment, when the processor executes the computer program to determine the recommendation score of the content tag based on the probability that the content tag originates from the target location cluster, the probability that the content tag originates from historical behavior data, and the probability that the content tag originates from preset data, the processor further implements the following steps:

[0193] A first weight is used to weight the probability that the content tag originates from the target location cluster to obtain a first weighted probability; a second weight is used to weight the probability that the content tag originates from the historical behavior data to obtain a second weighted probability; a third weight is used to weight the probability that the content tag originates from the preset data to obtain a third weighted probability; and a recommendation score for the content tag is determined based on the sum of the first weighted probability, the second weighted probability, and the third weighted probability.

[0194] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0195] After detecting a content recommendation event, the wireless signal sent by the remote control is received through the antenna array; wherein the wireless signal carries the gesture data of the remote control;

[0196] determining a spatial orientation of the remote control relative to the media playback device based on a phase difference of wireless signals received by each antenna in the antenna array; and

[0197] Determine the orientation of the remote control based on the posture data;

[0198] Determining a target spatial position representation of a user using a remote control based on the spatial pointing and orientation information and the signal strength of the wireless signal received by the antenna array;

[0199] Recommend content to users based on the target spatial location representation.

[0200] In one embodiment, when the processor executes the computer program to determine the target spatial position representation of the user using the remote control based on the spatial pointing and orientation information and the signal strength of the wireless signal received by the antenna array, the processor further implements the following steps:

[0201] The signal strength is normalized to obtain normalized intensity; the spatial pointing, orientation information and normalized intensity are concatenated to obtain an initial feature representation; the initial feature representation is normalized using standardized parameters to obtain a target spatial position representation of a user using a remote control; wherein the standardized parameters include a mean value and a standard deviation; the mean value and the standard deviation are determined based on the sample spatial position representation.

[0202] In one embodiment, when the processor executes the computer program to combine the spatial orientation, the orientation information, and the normalized intensity to obtain the initial feature representation, the processor further implements the following steps:

[0203] The spatial orientation and orientation information are weighted respectively to obtain weighted spatial orientation and weighted orientation information; the weighted spatial orientation, weighted orientation information and normalized intensity are concatenated to obtain the initial feature representation.

[0204] In one embodiment, when the processor executes the computer program to recommend content to the user based on the target spatial position representation, the processor further implements the following steps:

[0205] A target location cluster is selected from different candidate location clusters based on the distance between the target spatial location representation and different candidate location clusters; wherein the different candidate location clusters are obtained by clustering the sample spatial location representations, and the distance between the target location cluster and the target spatial location representation is the shortest; content is recommended to the user based on the first content tag corresponding to the target location cluster.

[0206] In one embodiment, when the processor executes the computer program to recommend content to the user based on the first content tag corresponding to the target location cluster, the processor further implements the following steps:

[0207] Determine a second content tag based on the user's historical behavior data; obtain extended content tags for the user; determine a recommendation score for each content tag based on the probability that the content tag originates from the target location cluster, the probability that the content tag originates from the historical behavior data, and the probability that the content tag originates from the preset data; select a target content tag from each content tag based on the recommendation score corresponding to each content tag, and recommend content to the user based on the target content tag; wherein each content tag includes an extended content tag, a first content tag, and a second content tag.

[0208] In one embodiment, when the processor executes the computer program to determine the recommendation score of the content tag based on the probability that the content tag originates from the target location cluster, the probability that the content tag originates from historical behavior data, and the probability that the content tag originates from preset data, the processor further implements the following steps:

[0209] A first weight is used to weight the probability that the content tag originates from the target location cluster to obtain a first weighted probability; a second weight is used to weight the probability that the content tag originates from the historical behavior data to obtain a second weighted probability; a third weight is used to weight the probability that the content tag originates from the preset data to obtain a third weighted probability; and a recommendation score for the content tag is determined based on the sum of the first weighted probability, the second weighted probability, and the third weighted probability.

[0210] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0211] After detecting a content recommendation event, the wireless signal sent by the remote control is received through the antenna array; wherein the wireless signal carries the gesture data of the remote control;

[0212] determining a spatial orientation of the remote control relative to the media playback device based on a phase difference of wireless signals received by each antenna in the antenna array; and

[0213] Determine the orientation of the remote control based on the posture data;

[0214] Determining a target spatial position representation of a user using a remote control based on the spatial pointing and orientation information and the signal strength of the wireless signal received by the antenna array;

[0215] Recommend content to users based on the target spatial location representation.

[0216] In one embodiment, when the processor executes the computer program to determine the target spatial position representation of the user using the remote control based on the spatial pointing and orientation information and the signal strength of the wireless signal received by the antenna array, the processor further implements the following steps:

[0217] The signal strength is normalized to obtain normalized intensity; the spatial pointing, orientation information and normalized intensity are concatenated to obtain an initial feature representation; the initial feature representation is normalized using standardized parameters to obtain a target spatial position representation of a user using a remote control; wherein the standardized parameters include a mean value and a standard deviation; the mean value and the standard deviation are determined based on the sample spatial position representation.

[0218] In one embodiment, when the processor executes the computer program to combine the spatial orientation, the orientation information, and the normalized intensity to obtain the initial feature representation, the processor further implements the following steps:

[0219] The spatial orientation and orientation information are weighted respectively to obtain weighted spatial orientation and weighted orientation information; the weighted spatial orientation, weighted orientation information and normalized intensity are concatenated to obtain the initial feature representation.

[0220] In one embodiment, when the processor executes the computer program to recommend content to the user based on the target spatial position representation, the processor further implements the following steps:

[0221] A target location cluster is selected from different candidate location clusters based on the distance between the target spatial location representation and different candidate location clusters; wherein the different candidate location clusters are obtained by clustering the sample spatial location representations, and the distance between the target location cluster and the target spatial location representation is the shortest; content is recommended to the user based on the first content tag corresponding to the target location cluster.

[0222] In one embodiment, when the processor executes the computer program to recommend content to the user based on the first content tag corresponding to the target location cluster, the processor further implements the following steps:

[0223] Determine a second content tag based on the user's historical behavior data; obtain extended content tags for the user; determine a recommendation score for each content tag based on the probability that the content tag originates from the target location cluster, the probability that the content tag originates from the historical behavior data, and the probability that the content tag originates from the preset data; select a target content tag from each content tag based on the recommendation score corresponding to each content tag, and recommend content to the user based on the target content tag; wherein each content tag includes an extended content tag, a first content tag, and a second content tag.

[0224] In one embodiment, when the processor executes the computer program to determine the recommendation score of the content tag based on the probability that the content tag originates from the target location cluster, the probability that the content tag originates from historical behavior data, and the probability that the content tag originates from preset data, the processor further implements the following steps:

[0225] A first weight is used to weight the probability that the content tag originates from the target location cluster to obtain a first weighted probability; a second weight is used to weight the probability that the content tag originates from the historical behavior data to obtain a second weighted probability; a third weight is used to weight the probability that the content tag originates from the preset data to obtain a third weighted probability; and a recommendation score for the content tag is determined based on the sum of the first weighted probability, the second weighted probability, and the third weighted probability.

[0226] It should be noted that the data involved in this application (including but not limited to posture data for remote controls, etc.) are all information and data fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0227] 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 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 be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database 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 processor involved in the various embodiments provided herein may be, but are not limited to, a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, and the like.

[0228] The technical features of the above embodiments can be combined arbitrarily. 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 specification.

[0229] 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 content recommendation method, characterized in that: Applied to a media playback device, the method includes: After detecting a content recommendation event, receiving a wireless signal sent by a remote controller through an antenna array; wherein the wireless signal carries the gesture data of the remote controller; determining a spatial orientation of the remote controller relative to the media playback device based on a phase difference of the wireless signal received by each antenna in the antenna array; and determining orientation information of the remote controller according to the posture data; determining a target spatial position representation of a user using the remote controller based on the spatial orientation, the orientation information, and a signal strength of a wireless signal received by the antenna array; Recommending content to the user based on the target spatial position representation.

2. The method according to claim 1, characterized in that The determining, based on the spatial orientation, the orientation information, and the signal strength of the wireless signal received by the antenna array, a target spatial position representation of a user using the remote controller includes: Normalizing the signal intensity to obtain a normalized intensity; concatenating the spatial orientation, the orientation information, and the normalized intensity to obtain an initial feature representation; The initial feature representation is standardized using standardized parameters to obtain a target spatial position representation of the user using the remote control; wherein the standardized parameters include a mean value and a standard deviation; the mean value and the standard deviation are determined based on the sample spatial position representation.

3. The method according to claim 2, characterized in that The step of concatenating the spatial orientation, the orientation information, and the normalized intensity to obtain an initial feature representation includes: performing weighted processing on the spatial orientation and the orientation information respectively to obtain weighted spatial orientation and weighted orientation information; The weighted spatial orientation, the weighted orientation information, and the normalized intensity are concatenated to obtain an initial feature representation.

4. The method according to claim 1, wherein The recommending content to the user according to the target spatial position representation includes: selecting a target location cluster from different candidate location clusters based on the distances between the target spatial location representation and different candidate location clusters; wherein the different candidate location clusters are obtained by clustering the sample spatial location representations, and the distance between the target location cluster and the target spatial location representation is the shortest; Recommending content to the user based on the first content tag corresponding to the target location cluster.

5. The method according to claim 4, characterized in that The recommending content to the user based on the first content tag corresponding to the target location cluster includes: determining a second content tag based on the user's historical behavior data; Obtaining an extended content tag for the user; For each content tag, determining a recommendation score for the content tag based on a probability that the content tag originates from the target location cluster, a probability that the content tag originates from the historical behavior data, and a probability that the content tag originates from preset data; A target content tag is selected from each content tag according to the recommendation score corresponding to each content tag, and content is recommended to the user based on the target content tag; wherein each content tag includes the extended content tag, the first content tag, and the second content tag.

6. The method according to claim 5, characterized in that Determining the recommendation score of the content tag based on the probability that the content tag originates from the target location cluster, the probability that the content tag originates from the historical behavior data, and the probability that the content tag originates from preset data includes: Using a first weight, weighting the probability that the content tag originates from the target location cluster to obtain a first weighted probability; Using a second weight, weighting the probability that the content tag is derived from the historical behavior data to obtain a second weighted probability; Using a third weight, weighting the probability that the content tag is derived from preset data to obtain a third weighted probability; A recommendation score for the content tag is determined according to the sum of the first weighted probability, the second weighted probability, and the third weighted probability.

7. A content recommendation device, characterized in that: Configured in a media playback device, the apparatus includes: A signal sending module, configured to receive a wireless signal sent by a remote controller via an antenna array after detecting a content recommendation event; wherein the wireless signal carries gesture data of the remote controller; an information determination module, configured to determine a spatial orientation of the remote control relative to the media playback device based on a phase difference of the wireless signal received by each antenna in the antenna array; and to determine orientation information of the remote control based on the posture data; a position determination module, configured to determine a target spatial position representation of a user using the remote controller based on the spatial orientation, the orientation information, and the signal strength of the wireless signal received by the antenna array; A content recommendation module is used to recommend content to the user based on the target spatial position representation.

8. A communication device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images