Media data recommendation method and apparatus

Abstract

The application relates to the field of media data processing, and provides a media data recommendation method and device. The media data recommendation method comprises the following steps: acquiring parameter values of multiple parameter categories of each media data information in multiple media data information, each media data information corresponding to at least one media data; determining a first ranking of each media data information in the multiple parameter categories according to the size order of the parameter values; determining a first priority of the media data information according to the first ranking of each media data information in the multiple parameter categories; and determining recommended media data in multiple media data corresponding to the multiple media data information according to the first priority of the multiple media data information. Based on the technical method, the recommended media data can be more in line with user preferences.

Description

Technical Field

[0001] This application relates to the field of media data processing, and more specifically, to a media data recommendation method and apparatus. Background Technology

[0002] As terminal devices become increasingly feature-rich, people can use the photo, audio, and video recording functions of smartphones and other smart terminals to collect media data, use the editing functions of smart terminals to modify and edit media data, and display, play, or share media data with others through smart terminals.

[0003] When a user opens the media library, the smart terminal can determine the recommended media data that the user likes based on the amount of interaction the user has with each media data in the library, as well as the aesthetic appeal of each media data, and display it to the user. This reduces the time required for the user to filter through the large amount of data in the media library and improves the user experience.

[0004] In determining recommended media data, the parameter values ​​of multiple parameter types for each media data point need to be weighted and summed to obtain a recommendation score. The recommended media data is then selected for inclusion in the media library based on this score. However, in some cases, for certain parameter types, the parameter values ​​of multiple media data points differ significantly. The parameter value of one media data point may be at least an order of magnitude larger than the parameter values ​​of most other media data points. In such cases, the parameter values ​​of most other media data points have almost no impact on the recommendation score, meaning their recommendation score is likely lower than that of the individual media data point, leading to that individual media data point being selected as the recommended media data. This large parameter value of the individual media data point may be due to user error, work / life needs, or other factors, and does not necessarily reflect the user's actual preferences. In other words, the recommended media data may not perfectly match user preferences, resulting in a poor user experience. Summary of the Invention

[0005] This application provides a media data recommendation method and apparatus, which can make the recommended media data more in line with user preferences and improve user experience.

[0006] In a first aspect, a media data recommendation method is provided, comprising: obtaining parameter values ​​of multiple parameter categories for each media data information in a plurality of media data information, wherein each media data information corresponds to at least one media data; determining a first ranking of each media data information in the plurality of parameter categories according to the order of the parameter values; determining a first priority of each media data information in the plurality of parameter categories according to the first ranking of each media data information in the plurality of parameter categories; and determining recommended media data among the plurality of media data corresponding to the plurality of media data information according to the first priority of the plurality of media data information.

[0007] In the embodiments of this application, recommended media data is determined based on the ranking of the parameter values ​​of each media data in multiple parameter types, in descending order of magnitude. The difference between the rankings of two adjacent media data in a certain parameter type is only 1. Even if the interaction volume of some media data in a certain parameter type is much greater than that of other media data, the difference in ranking is small. Compared with the method of adding the parameter values ​​of multiple parameter types of media data, the solution of this application reduces the impact of individual parameter values ​​that differ significantly in value from other parameter values ​​on the determination of recommended media data, making the recommended media data more in line with user preferences.

[0008] In one possible implementation, different parameter categories are different interaction categories, and the parameter value of each parameter category for each media data information is the amount of interaction between the user and at least one media data corresponding to the media data information in that parameter category.

[0009] Due to user errors or the demands of daily life and work, in some interaction categories, the amount of interaction with a particular media data may be significantly greater than that with other media data. Multiple media data points are sorted according to the amount of interaction across multiple interaction categories. Based on the ranking of media data across these categories, recommended media data is determined from among the various media data points, making the recommended media data more aligned with user preferences.

[0010] In one possible implementation, different media data information represents different semantic categories; determining recommended media data from multiple media data corresponding to the multiple media data information based on the first priority of the multiple media data information includes: for each media data, weighted summing of the first priorities of at least one semantic category contained in the media data to obtain a semantic score for the media data; and determining the recommended media data based on the semantic scores of the multiple media data.

[0011] Different media data fall into different semantic categories, which allows the semantic scores of media data to accurately reflect users' semantic preferences, thus enabling the recommended media data to better match users' preferences.

[0012] In one possible implementation, the method further includes: obtaining recommended topics, where different topics correspond to different weight combinations, and each weight combination includes weights corresponding to each semantic category; for each media data, weighted summing of the first priorities of at least one semantic category contained in the media data to obtain a semantic score for the media data includes: for each media data, weighted summing of the first priorities of at least one semantic category contained in the media data according to the weight combination corresponding to the recommended topic to obtain a semantic score for the media data.

[0013] The first priority of different media data information represents the user's preference for different semantic categories. Multiple topics are associated with multiple weight combinations. When calculating the semantic score of media data, the preference for the semantic categories included in the media data is weighted and accumulated according to the weight combination corresponding to the recommended topic. This makes the determined recommended media data more consistent with the recommended topic and the media recommendation method more flexible.

[0014] In one possible implementation, different media data information corresponds to different media data; the method further includes: obtaining a second priority for each media data among the multiple media data corresponding to the multiple media data information; determining recommended media data among the multiple media data corresponding to the multiple media data information according to the first priority of the multiple media data information includes: determining a second ranking of the multiple media data according to the order of the second priority; determining a third ranking of the multiple media data according to the order of the first priority; and determining the recommended media data according to the second ranking and the third ranking of each media data.

[0015] Different media data are distinct. User interactions with media data, and the frequency of those interactions, implicitly reveal user preferences for that specific media data. The first priority of media data can be understood as the degree of user preference determined by comprehensively considering the interaction volume across different interaction categories. The influence of other factors on user preferences for media data can be represented as the second priority. Considering these other factors, recommending media data is determined based on its ranking in the first and second priorities. This reduces the impact of individual values ​​with significant differences from other priorities on the determination of recommended media data, further improving the degree to which recommended media data aligns with user preferences.

[0016] In one possible implementation, the multiple parameter categories for each media data information include at least two of the following: an interaction score, a semantic score, and an aesthetic score. The interaction score is determined based on the amount of interaction between the user and the media data. The semantic score is used to represent the degree of the user's semantic preference for the media data, and the aesthetic score is used to represent the aesthetic appeal of the media data.

[0017] The amount of user interaction with media data can reflect user preferences to some extent. Aesthetic appeal and user preference for semantic categories also influence user liking for media data. Recommended media data is determined based on a subset of the interaction score, semantic score, and aesthetic score, making the recommended media data more aligned with user preferences.

[0018] Furthermore, among the interaction scores, semantic scores, or aesthetic scores of multiple media data, there may be scores that differ significantly from the majority of other scores. Sort the multiple media data according to the score size of each different type, and determine the recommended media data based on the ranking of the media data in multiple score types. This can make the recommended media data more in line with user preferences.

[0019] Secondly, a media data recommendation apparatus is provided, including units for performing the method of the first aspect and any possible implementation thereof. The apparatus may be a terminal device or a chip within a terminal device. The apparatus may include an acquisition unit and a processing unit.

[0020] When the device is a terminal device, the processing unit may be a processor, and the acquisition unit may be a communication interface; the terminal device may also include a memory for storing computer program code, which, when the processor executes the computer program code stored in the memory, causes the terminal device to perform the method of the first aspect and any possible implementation thereof.

[0021] When the device is a chip within a terminal device, the processing unit can be an internal processing unit of the chip, and the input unit can be an output interface, pin, or circuit, etc.; the chip may also include a memory, which can be an internal memory of the chip (e.g., a register, cache, etc.) or an external memory (e.g., a read-only memory, random access memory, etc.); the memory is used to store computer program code, and when the processor executes the computer program code stored in the memory, the chip performs the method of the first aspect and any possible implementation of the first aspect.

[0022] Fourthly, a computer-readable storage medium is provided, the computer-readable storage medium storing computer program code, which, when executed by a media data recommendation device, causes the media data recommendation device to perform the method of the first aspect and any possible implementation thereof.

[0023] Fifthly, a computer program product is provided, the computer program product comprising: computer program code, which, when run by a media data recommendation device, causes the media data recommendation device to perform the method of the first aspect and any possible implementation thereof. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of a hardware system applicable to the apparatus of this application; Figure 2 This is a schematic diagram of a software system applicable to the apparatus of this application; Figure 3 This is a schematic diagram illustrating an application scenario provided in an embodiment of this application; Figures 4 to 6 This is a schematic diagram of the user interface provided in an embodiment of this application; Figure 7 This is a schematic flowchart of an image data recommendation method; Figure 8 This is a schematic flowchart illustrating an image data recommendation method provided in an embodiment of this application; Figure 9 This is a schematic flowchart illustrating another image data recommendation method provided in an embodiment of this application; Figure 10 This is a schematic structural diagram of an image data recommendation device provided in this application; Figure 11 This is a schematic structural diagram of an electronic device for recommending image data, as provided in this application. Detailed Implementation

[0025] The technical solutions in the embodiments of this application will now be described with reference to the accompanying drawings.

[0026] Figure 1 A hardware system suitable for the apparatus of this application is shown.

[0027] Device 100 may be a mobile phone, smart screen, tablet computer, wearable electronic device, in-vehicle electronic device, augmented reality (AR) device, virtual reality (VR) device, laptop computer, ultra-mobile personal computer (UMPC), netbook, personal digital assistant (PDA), projector, etc. This application embodiment does not limit the specific type of device 100.

[0028] Device 100 may include a processor 110, an external memory interface 120, an internal memory 121, a universal serial bus (USB) interface 130, a charging management module 140, a power management module 141, a battery 142, an antenna 1, an antenna 2, a mobile communication module 150, a wireless communication module 160, an audio module 170, a speaker 170A, a receiver 170B, a microphone 170C, a headphone jack 170D, a sensor module 180, buttons 190, a motor 191, an indicator 192, a camera 193, a display screen 194, and a subscriber identification module (SIM) card interface 195, etc. The sensor module 180 may include a pressure sensor 180A, a gyroscope sensor 180B, a barometric pressure sensor 180C, a magnetic sensor 180D, an accelerometer sensor 180E, a distance sensor 180F, a proximity sensor 180G, a fingerprint sensor 180H, a temperature sensor 180J, a touch sensor 180K, an ambient light sensor 180L, a bone conduction sensor 180M, etc.

[0029] It should be noted that, Figure 1 The structure shown does not constitute a specific limitation on device 100. In other embodiments of this application, device 100 may include a... Figure 1 The components shown may include more or fewer components, or the device 100 may include... Figure 1 The components shown may be a combination of certain components, or the device 100 may include... Figure 1 Sub-components of some of the components shown. Figure 1 The components shown can be implemented in hardware, software, or a combination of software and hardware.

[0030] Processor 110 may include one or more processing units. For example, processor 110 may include at least one of the following processing units: application processor (AP), modem processor, graphics processing unit (GPU), image signal processor (ISP), controller, video codec, digital signal processor (DSP), baseband processor, and neural network processing unit (NPU). These different processing units may be independent devices or integrated devices.

[0031] The controller can generate operation control signals based on the instruction opcode and timing signals to complete the control of instruction fetching and execution.

[0032] The processor 110 may also include a memory for storing instructions and data. In some embodiments, the memory in the processor 110 is a cache memory. This memory can store instructions or data that the processor 110 has just used or that are used repeatedly. If the processor 110 needs to use the instruction or data again, it can retrieve it directly from the memory. This avoids repeated accesses, reduces the waiting time of the processor 110, and thus improves the efficiency of the system.

[0033] In some embodiments, processor 110 may include one or more interfaces. For example, processor 110 may include at least one of the following interfaces: an inter-integrated circuit (I2C) interface, an inter-integrated circuit sound (I2S) interface, a pulse code modulation (PCM) interface, a universal asynchronous receiver / transmitter (UART) interface, a mobile industry processor interface (MIPI), a general-purpose input / output (GPIO) interface, a SIM interface, and a USB interface.

[0034] The I2C interface is a bidirectional synchronous serial bus, including a serial data line (SDA) and a serial clock line (SCL). In some embodiments, the processor 110 may include multiple I2C buses. The processor 110 can couple to the touch sensor 180K, charger, flash, camera 193, etc., through different I2C bus interfaces. For example, the processor 110 can couple to the touch sensor 180K through the I2C interface, enabling the processor 110 and the touch sensor 180K to communicate through the I2C bus interface, thereby realizing the touch function of the device 100.

[0035] The I2S interface can be used for audio communication. In some embodiments, the processor 110 may include multiple I2S buses. The processor 110 can be coupled to the audio module 170 via the I2S bus to enable communication between the processor 110 and the audio module 170. In some embodiments, the audio module 170 can transmit audio signals to the wireless communication module 160 via the I2S interface to enable the function of answering phone calls through a Bluetooth headset.

[0036] The PCM interface can also be used for audio communication, sampling, quantizing, and encoding analog signals. In some embodiments, the audio module 170 and the wireless communication module 160 can be coupled via the PCM interface.

[0037] In some embodiments, the audio module 170 can also transmit audio signals to the wireless communication module 160 via the PCM interface to enable the function of answering phone calls through a Bluetooth headset. Both the I2S interface and the PCM interface can be used for audio communication.

[0038] The UART interface is a universal serial data bus used for asynchronous communication. This bus can be a bidirectional communication bus. It converts the data to be transmitted between serial and parallel communication. In some embodiments, the UART interface is typically used to connect the processor 110 and the wireless communication module 160. For example, the processor 110 communicates with the Bluetooth module in the wireless communication module 160 via the UART interface to implement Bluetooth functionality.

[0039] In some embodiments, the audio module 170 can transmit audio signals to the wireless communication module 160 via the UART interface to enable the function of playing music through Bluetooth headphones.

[0040] The MIPI interface can be used to connect the processor 110 to peripheral devices such as the display screen 194 and the camera 193. The MIPI interface includes a camera serial interface (CSI) and a display serial interface (DSI). In some embodiments, the processor 110 and the camera 193 communicate via the CSI interface to enable the shooting function of the device 100. The processor 110 and the display screen 194 communicate via the DSI interface to enable the display function of the device 100.

[0041] The GPIO interface can be configured via software. It can be configured as a control signal interface or a data signal interface.

[0042] In some embodiments, the GPIO interface can be used to connect the processor 110 to the camera 193, the display screen 194, the wireless communication module 160, the audio module 170, and the sensor module 180. The GPIO interface can also be configured as an I2C interface, an I2S interface, a UART interface, or a MIPI interface.

[0043] USB interface 130 is a USB standard compliant interface, such as a Mini USB, Micro USB, or USB Type-C interface. USB interface 130 can be used to connect a charger to charge device 100, to transfer data between device 100 and peripheral devices, and to connect headphones for audio playback. USB interface 130 can also be used to connect other devices 100, such as AR devices.

[0044] Figure 1 The connection relationships between the modules shown are merely illustrative and do not constitute a limitation on the connection relationships between the modules of device 100. Optionally, the modules of device 100 may also adopt a combination of various connection methods described in the above embodiments.

[0045] The charging management module 140 receives power from a charger. The charger can be a wireless charger or a wired charger. In some wired charging embodiments, the charging management module 140 receives current from the wired charger via the USB interface 130. In some wireless charging embodiments, the charging management module 140 receives electromagnetic waves (the current path is shown as dashed lines) via the wireless charging coil of the device 100. While charging the battery 142, the charging management module 140 can also supply power to the device 100 via the power management module 141.

[0046] The power management module 141 connects the battery 142, the charging management module 140, and the processor 110. The power management module 141 receives input from the battery 142 and / or the charging management module 140, providing power to the processor 110, internal memory 121, display screen 194, camera 193, and wireless communication module 160, etc. The power management module 141 can also monitor parameters such as battery capacity, battery cycle count, and battery health status (e.g., leakage current, impedance). Optionally, the power management module 141 can be located within the processor 110, or the power management module 141 and the charging management module 140 can be located in the same device.

[0047] The wireless communication function of device 100 can be implemented by devices such as antenna 1, antenna 2, mobile communication module 150, wireless communication module 160, modem processor and baseband processor.

[0048] Antennas 1 and 2 are used to transmit and receive electromagnetic wave signals. Each antenna in device 100 can be used to cover one or more communication frequency bands. Different antennas can also be multiplexed to improve antenna utilization. For example, antenna 1 can be multiplexed as a diversity antenna for a wireless local area network. In some other embodiments, the antennas can be used in conjunction with a tuning switch.

[0049] Mobile communication module 150 can provide a wireless communication solution applied to device 100, such as at least one of the following: second generation (2G) th Generation 2G mobile communication solutions, third generation (3G) th Generation 3G mobile communication solutions, fourth generation (4G) th 5G mobile communication solutions, fifth generation (5G) th (5G) mobile communication solutions. The mobile communication module 150 may include at least one filter, switch, power amplifier, low noise amplifier (LNA), etc. The mobile communication module 150 can receive electromagnetic waves via antenna 1, and perform filtering and amplification on the received electromagnetic waves before transmitting them to a modem processor for demodulation. The mobile communication module 150 can also amplify the signal modulated by the modem processor, and the amplified signal is converted into electromagnetic waves and radiated by antenna 1. In some embodiments, at least some functional modules of the mobile communication module 150 may be housed in processor 110. In some embodiments, at least some functional modules of the mobile communication module 150 and at least some modules of the processor 110 may be housed in the same device.

[0050] The modem processor may include a modulator and a demodulator. The modulator modulates a low-frequency baseband signal to be transmitted into a mid-to-high frequency signal. The demodulator demodulates a received electromagnetic wave signal into a low-frequency baseband signal. The demodulator then transmits the demodulated low-frequency baseband signal to the baseband processor for processing. After processing by the baseband processor, the low-frequency baseband signal is transmitted to the application processor. The application processor outputs sound signals through audio devices (e.g., speaker 170A, receiver 170B) or displays images or videos through the display screen 194. In some embodiments, the modem processor may be a separate device. In other embodiments, the modem processor may be independent of the processor 110 and may be housed in the same device as the mobile communication module 150 or other functional modules.

[0051] Similar to the mobile communication module 150, the wireless communication module 160 can also provide a wireless communication solution applied to the device 100, such as at least one of the following: wireless local area networks (WLAN), Bluetooth (BT), Bluetooth Low Energy (BLE), ultra-wideband (UWB), global navigation satellite system (GNSS), frequency modulation (FM), near field communication (NFC), and infrared (IR) technology. The wireless communication module 160 can be one or more devices integrating at least one communication processing module. The wireless communication module 160 receives electromagnetic waves via antenna 2, performs frequency modulation and filtering of the electromagnetic wave signal, and sends the processed signal to processor 110. The wireless communication module 160 can also receive signals to be transmitted from processor 110, perform frequency modulation and amplification on them, and convert the signals into electromagnetic waves that are radiated via antenna 2.

[0052] In some embodiments, antenna 1 of device 100 is coupled to mobile communication module 150, and antenna 2 of device 100 is coupled to wireless communication module 160, enabling device 100 to communicate with networks and other electronic devices via wireless communication technology. This wireless communication technology may include at least one of the following communication technologies: Global System for Mobile Communications (GSM), General Packet Radio Service (GPRS), Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), Time Division Code Division Multiple Access (TD-SCDMA), Long Term Evolution (LTE), BT, GNSS, WLAN, NFC, FM, and IR technologies. The GNSS may include at least one of the following positioning technologies: Global Positioning System (GPS), Global Navigation Satellite System (GLONASS), BeiDou Navigation Satellite System (BDS), Quasi-Zenith Satellite System (QZSS), and Satellite Based Augmentation System (SBAS).

[0053] Device 100 can implement display functions via a GPU, a display screen 194, and an application processor. The GPU is a microprocessor for image processing, connected to the display screen 194 and the application processor. The GPU is used to perform mathematical and geometric calculations for graphics rendering. Processor 110 may include one or more GPUs, which execute program instructions to generate or modify display information.

[0054] The display screen 194 can be used to display images or videos. The display screen 194 includes a display panel. The display panel can be a liquid crystal display (LCD), an organic light-emitting diode (OLED), an active-matrix organic light-emitting diode (AMOLED), a flexible light-emitting diode (FLED), a mini light-emitting diode (Mini LED), a micro light-emitting diode (Micro LED), a micro OLED, or a quantum dot light-emitting diode (QLED). In some embodiments, the device 100 may include one or N displays 194, where N is a positive integer greater than 1.

[0055] The device 100 can perform shooting functions through an ISP, camera 193, video codec, GPU, display 194, and application processor.

[0056] The ISP is used to process data fed back from the camera 193. For example, when taking a picture, the shutter is opened, and light is transmitted through the lens to the camera's photosensitive element. The light signal is converted into an electrical signal, and the camera's photosensitive element transmits the electrical signal to the ISP for processing, transforming it into an image visible to the naked eye. The ISP can perform algorithmic optimization of image noise, brightness, and color. The ISP can also optimize parameters such as exposure and color temperature of the shooting scene. In some embodiments, the ISP can be set in the camera 193.

[0057] Camera 193 is used to capture still images or videos. An object is projected onto a photosensitive element by generating an optical image through the lens. The photosensitive element can be a charge-coupled device (CCD) or a complementary metal-oxide-semiconductor (CMOS) phototransistor. The photosensitive element converts the light signal into an electrical signal, which is then passed to an ISP for conversion into a digital image signal. The ISP outputs the digital image signal to a DSP for processing. The DSP converts the digital image signal into a standard red-green-blue (RGB), YUV, or other image signal format. In some embodiments, device 100 may include one or N cameras 193, where N is a positive integer greater than 1.

[0058] A digital signal processor (DSP) is used to process digital signals. Besides digital image signals, it can also process other digital signals. For example, when device 100 selects a frequency point, the DSP can perform Fourier transforms on the frequency energy.

[0059] Video codecs are used to compress or decompress digital video. Device 100 may support one or more video codecs. Thus, device 100 can play or record video in various encoding formats, such as Moving Picture Experts Group (MPEG) 1, MPEG2, MPEG3, and MPEG4.

[0060] An NPU (Neural Processing Unit) is a processor that borrows from the structure of biological neural networks, such as the transmission patterns between neurons in the human brain, to quickly process input information and continuously learn. Through an NPU, device 100 can achieve intelligent cognitive functions, such as image recognition, facial recognition, speech recognition, and text understanding.

[0061] The external storage interface 120 can be used to connect an external storage card, such as a secure digital (SD) card, to enable the storage capacity of the expansion device 100. The external storage card communicates with the processor 110 through the external storage interface 120 to perform data storage functions. For example, music, video, and other files can be stored on the external storage card.

[0062] Internal memory 121 can be used to store computer executable program code, which includes instructions. Internal memory 121 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function (e.g., sound playback and image playback). The data storage area may store data created during the use of device 100 (e.g., audio data and phonebook). Furthermore, internal memory 121 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, and universal flash storage (UFS). Processor 110 executes various processing methods of device 100 by running instructions stored in internal memory 121 and / or instructions stored in memory disposed in the processor.

[0063] The device 100 can implement audio functions, such as music playback and recording, through an audio module 170, a speaker 170A, a receiver 170B, a microphone 170C, a headphone jack 170D, and an application processor.

[0064] The audio module 170 is used to convert digital audio information into analog audio signals for output, and can also be used to convert analog audio input into digital audio signals. The audio module 170 can also be used for encoding and decoding audio signals. In some embodiments, the audio module 170, or some functional modules of the audio module 170, can be located in the processor 110.

[0065] The speaker 170A, also known as a loudspeaker, is used to convert audio electrical signals into sound signals. The device 100 can listen to music or make hands-free calls through the speaker 170A.

[0066] The receiver 170B, also known as the earpiece, is used to convert audio electrical signals into sound signals. When a user uses device 100 to answer a phone call or voice message, they can listen to the voice by bringing the receiver 170B close to their ear.

[0067] Microphone 170C, also known as a microphone or transducer, is used to convert sound signals into electrical signals. When a user makes a phone call or sends a voice message, they can input a sound signal into microphone 170C by speaking close to it. Device 100 may include at least one microphone 170C. In some embodiments, device 100 may include two microphones 170C to achieve noise reduction. In other embodiments, device 100 may also include three, four, or more microphones 170C to achieve functions such as sound source identification and directional recording. Processor 110 can process the electrical signals output by microphone 170C. For example, audio module 170 and wireless communication module 160 can be coupled through a PCM interface. After microphone 170C converts ambient sound into an electrical signal (such as a PCM signal), it transmits the electrical signal to processor 110 through the PCM interface. Processor 110 performs volume and frequency analysis on the electrical signal to determine the volume and frequency of the ambient sound.

[0068] The 170D headphone jack is used to connect wired headphones. The 170D headphone jack can be a USB 130 interface or a 3.5mm Open Mobile Terminal Platform (OMTP) standard interface, a CTIA (Cellular Telecommunications Industry Association of the USA) standard interface.

[0069] Pressure sensor 180A is used to sense pressure signals and can convert the pressure signals into electrical signals. In some embodiments, pressure sensor 180A can be disposed on display screen 194. Pressure sensor 180A can be of many types, such as resistive pressure sensor, inductive pressure sensor, or capacitive pressure sensor. Capacitive pressure sensor can include at least two parallel plates with conductive material. When force is applied to pressure sensor 180A, the capacitance between the electrodes changes, and device 100 determines the pressure intensity based on the change in capacitance. When a touch operation is applied to display screen 194, device 100 detects the touch operation based on pressure sensor 180A. Device 100 can also calculate the touch position based on the detection signal from pressure sensor 180A. In some embodiments, touch operations applied to the same touch position but with different touch operation intensities can correspond to different operation commands. For example: when a touch operation with an intensity less than a first pressure threshold is applied to the SMS application icon, a command to view an SMS is executed; when a touch operation with an intensity greater than or equal to the first pressure threshold is applied to the SMS application icon, a command to create a new SMS is executed.

[0070] The gyroscope sensor 180B can be used to determine the motion attitude of the device 100. In some embodiments, the gyroscope sensor 180B can determine the angular velocity of the device 100 around three axes (i.e., the x-axis, y-axis, and z-axis). The gyroscope sensor 180B can be used for image stabilization. For example, when the shutter is pressed, the gyroscope sensor 180B detects the angle of the device 100's shake, calculates the distance that the lens module needs to compensate based on the angle, and allows the lens to counteract the shake of the device 100 through reverse movement, thus achieving image stabilization. The gyroscope sensor 180B can also be used in scenarios such as navigation and motion-sensing games.

[0071] The barometric pressure sensor 180C is used to measure air pressure. In some embodiments, the device 100 calculates altitude using the air pressure value measured by the barometric pressure sensor 180C to assist in positioning and navigation.

[0072] The magnetic sensor 180D includes a Hall sensor. The device 100 can use the magnetic sensor 180D to detect the opening and closing of the flip cover. In some embodiments, when the device 100 is a flip phone, the device 100 can detect the opening and closing of the flip cover based on the magnetic sensor 180D. The device 100 can set features such as automatic flip unlocking based on the detected opening and closing state of the cover or the flip cover.

[0073] The accelerometer 180E can detect the magnitude of acceleration of device 100 in various directions (typically the x-axis, y-axis, and z-axis). When device 100 is stationary, it can detect the magnitude and direction of gravity. The accelerometer 180E can also be used to identify the attitude of device 100, serving as input parameters for applications such as screen orientation switching and pedometers.

[0074] Distance sensor 180F is used to measure distance. Device 100 can measure distance via infrared or laser. In some embodiments, such as in a shooting scenario, device 100 can utilize distance sensor 180F for distance measurement to achieve fast focusing.

[0075] The proximity sensor 180G may include, for example, a light-emitting diode (LED) and a light detector, such as a photodiode. The LED may be an infrared LED. Device 100 emits infrared light outward through the LED. Device 100 uses the photodiode to detect infrared reflected light from nearby objects. When reflected light is detected, device 100 can determine that an object is nearby. When no reflected light is detected, device 100 can determine that no object is nearby. Device 100 can use the proximity sensor 180G to detect whether the user is holding device 100 close to their ear for a call, so as to automatically turn off the screen to save power. The proximity sensor 180G can also be used for automatic unlocking and automatic screen locking in folding or pocket modes.

[0076] The ambient light sensor 180L is used to sense the ambient light intensity. The device 100 can adaptively adjust the brightness of the display screen 194 according to the sensed ambient light intensity. The ambient light sensor 180L can also be used to automatically adjust the white balance when taking pictures. The ambient light sensor 180L can also be used in conjunction with the proximity sensor 180G to detect whether the device 100 is in a pocket to prevent accidental activation.

[0077] The fingerprint sensor 180H is used to collect fingerprints. The device 100 can use the characteristics of the collected fingerprints to unlock the device, access app locks, take photos, and answer calls.

[0078] Temperature sensor 180J is used to detect temperature. In some embodiments, device 100 uses the temperature detected by temperature sensor 180J to execute a temperature handling strategy. For example, when the temperature reported by temperature sensor 180J exceeds a threshold, device 100 performs thermal protection by reducing the performance of a processor located near temperature sensor 180J to reduce power consumption. In other embodiments, when the temperature is below another threshold, device 100 heats battery 142 to prevent abnormal shutdown of device 100 due to low temperature. In still other embodiments, when the temperature is below yet another threshold, device 100 boosts the output voltage of battery 142 to prevent abnormal shutdown due to low temperature.

[0079] Touch sensor 180K, also known as a touch device, can be disposed on display screen 194. The touch sensor 180K and display screen 194 together form a touchscreen, also known as a touch display. Touch sensor 180K is used to detect touch operations applied to or near it. Touch sensor 180K can transmit the detected touch operation to the application processor to determine the type of touch event. Visual output related to the touch operation can be provided through display screen 194. In other embodiments, touch sensor 180K may also be disposed on the surface of device 100, and in a different location from display screen 194.

[0080] The bone conduction sensor 180M can acquire vibration signals. In some embodiments, the bone conduction sensor 180M can acquire vibration signals from the vibrating bone segments of the human vocal cords. The bone conduction sensor 180M can also contact the human pulse to receive blood pressure signals. In some embodiments, the bone conduction sensor 180M can also be incorporated into headphones to form bone conduction headphones. The audio module 170 can parse the voice signals from the vibrating bone segments of the vocal cords acquired by the bone conduction sensor 180M to realize voice functionality. The application processor can parse heart rate information from the blood pressure signals acquired by the bone conduction sensor 180M to realize heart rate detection functionality.

[0081] Button 190 includes a power button and volume buttons. Button 190 can be a mechanical button or a touch button. Device 100 can receive button input signals and implement functions related to the button input signals.

[0082] Motor 191 can generate vibrations. Motor 191 can be used for incoming call notifications or for haptic feedback. Motor 191 can produce different vibration feedback effects for touch operations applied to different applications. Motor 191 can also produce different vibration feedback effects for touch operations applied to different areas of the display screen 194. Different application scenarios (e.g., time reminders, receiving messages, alarm clocks, and games) can correspond to different vibration feedback effects. The touch vibration feedback effect can also be customized.

[0083] Indicator 192 can be an indicator light, which can be used to indicate charging status and power changes, or to indicate messages, missed calls and notifications.

[0084] The SIM card interface 195 is used to connect a SIM card. The SIM card can be inserted into the SIM card interface 195 to make contact with the device 100, or it can be removed from the SIM card interface 195 to detach from the device 100. The device 100 can support one or N SIM card interfaces, where N is a positive integer greater than 1. Multiple cards can be inserted into the same SIM card interface 195 simultaneously; the multiple cards can be of the same or different types. The SIM card interface 195 is also compatible with external memory cards. The device 100 interacts with the network through the SIM card to realize functions such as voice calls and data communication. In some embodiments, the device 100 uses an embedded SIM (eSIM) card, which can be embedded in the device 100 and cannot be separated from it.

[0085] The hardware system of device 100 has been described in detail above. The software system of device 100 is described below. The software system can adopt a layered architecture, event-driven architecture, microkernel architecture, microservice architecture, or cloud architecture. This application embodiment takes a layered architecture as an example to exemplarily describe the software system of device 100.

[0086] like Figure 2 As shown, a layered architecture software system is divided into several layers, each with a clear role and function. Layers communicate with each other through software interfaces. In some embodiments, the software system can be divided into four layers, from top to bottom: the application layer, the application framework layer, the Android Runtime and system libraries, and the kernel layer.

[0087] The application layer can include applications such as camera, gallery, calendar, call, map, navigation, WLAN, Bluetooth, music, video, and SMS.

[0088] The application framework layer provides application programming interfaces (APIs) and programming frameworks for applications in the application layer. The application framework layer may include some predefined functions.

[0089] For example, the application framework layer includes a window manager, content providers, a view system, a phone manager, a resource manager, and a notification manager.

[0090] The window manager is used to manage windowed applications. It can obtain the screen size, determine if a status bar is present, lock the screen, and capture the screen.

[0091] Content providers store and retrieve data, making that data accessible to applications. This data may include videos, images, audio, made and received phone calls, browsing history and bookmarks, and phone books.

[0092] A view system includes visual controls, such as controls for displaying text and controls for displaying images. View systems can be used to build applications. A display interface can consist of one or more views; for example, a display interface including a text notification icon can include views for displaying text and views for displaying images.

[0093] The phone manager is used to provide communication functions for device 100, such as the management of call status (connected or disconnected).

[0094] The file explorer provides applications with various resources, such as localized strings, icons, images, layout files, and video files.

[0095] The notification manager allows applications to display notifications in the status bar. These notifications can be used to deliver informational messages and can disappear automatically after a short pause, requiring no user interaction. For example, the notification manager can be used to notify users of download completions and message alerts. The notification manager can also manage notifications that appear as icons or scrollbar text in the system's top status bar, such as notifications from background applications. Furthermore, the notification manager can manage notifications that appear on the screen as dialog boxes, such as text messages in the status bar, sound alerts, electronic device vibrations, and indicator light flashing.

[0096] The Android Runtime consists of core libraries and a virtual machine. The Android Runtime is responsible for the scheduling and management of the Android system.

[0097] The core library consists of two parts: one part is the functionalities that need to be called by the Java language, and the other part is the Android core library.

[0098] The application layer and application framework layer run in a virtual machine. The virtual machine executes the Java files of the application layer and application framework layer as binary files. The virtual machine is used to perform functions such as object lifecycle management, stack management, thread management, security and exception management, and garbage collection.

[0099] The system library may include multiple functional modules, such as: surface manager, media libraries, 3D graphics processing libraries (e.g., OpenGL ES for embedded systems) and 2D graphics engines (e.g., Skia graphics library (SGL)).

[0100] The Surface Manager is used to manage the display subsystem and provides the blending of 2D and 3D layers for multiple applications.

[0101] The media library supports playback and recording of various audio and video formats, as well as still image files. It supports multiple audio and video encoding formats, such as MPEG4, H.264, Moving Picture Experts Group Audio Layer III (MP3), Advanced Audio Coding (AAC), Adaptive Multi-rate (AMR), Joint Photographic Experts Group (JPG), and Portable Network Graphics (PNG).

[0102] The 3D graphics processing library can be used to implement 3D graphics drawing, image rendering, compositing, and layer processing.

[0103] A 2D graphics engine is a graphics engine for 2D drawing.

[0104] The kernel layer is the layer between hardware and software. It can include driver modules such as display drivers, camera drivers, audio drivers, and sensor drivers.

[0105] It should be noted that the above is passed Figure 1 An example is given to illustrate the structure of an electronic device. Figure 2 An example of a software architecture diagram for an electronic device is provided; however, this application does not impose any limitations on it.

[0106] The application scenarios of the embodiments of this application will be briefly described below.

[0107] For example, such as Figure 3 As shown, the media data recommendation method of this application can be applied to the field of media data processing. In this method, media data information from multiple media datasets in a media library is analyzed and processed to determine recommended media data, which is then recommended to the user. The media data information reflects the user's preference for each media dataset.

[0108] A media library can include one or more of the following: an audio library, an image library, a video library, and a document library. An audio library, also known as a music library, contains multiple audio files. An image library, also known as a photo album, contains multiple image files. The video and document libraries contain video and document data, respectively.

[0109] The following uses a media library as an example to illustrate the application scenarios of this application.

[0110] The functionality of cameras on mobile devices has continuously evolved, from single-lens to multi-lens, from fixed focal length to zoom, from low-pixel to high-pixel, and with the emergence of super night scene mode. This has provided users with the convenience to record and share beautiful moments in life anytime, anywhere, resulting in a growing amount of image data in the image libraries of mobile devices. As intelligence has become a key focus for mobile device manufacturers, more and more manufacturers are implementing features that automatically recommend selected pictures or generate short videos for users based on information such as time, location, and people in photos and videos, making it easier for users to browse and share image content.

[0111] When a user opens their media library, recommended media data can be displayed to the user.

[0112] Figure 4 A graphical user interface (GUI) of a terminal device is shown, which is the desktop 410 of the terminal device. When the terminal device detects that the user clicks the icon 411 of the photo album application (APP) on the desktop 410, it can launch the photo album application and display the following: Figure 5 Another GUI is shown. Figure 5 The GUI shown can be referred to as the album interface 420. The album interface 420 may include multiple recommended image data 421 to 426.

[0113] Or, such as Figure 6 As shown, the icon 431 of the photo album application on the desktop 430 of the terminal device may include one or more of a plurality of recommended image data 421 to 426. In some embodiments, the plurality of recommended image data in icon 431 may change over time.

[0114] Alternatively, the terminal device can generate an image display video based on multiple recommended image data, and display the image display video through the user interface to show the multiple recommended image data to the user.

[0115] It should be understood that the above are illustrative examples of application scenarios and do not limit the application scenarios of this application in any way.

[0116] However, different user groups have different preferences for image categories. For example, outdoor enthusiasts may prefer nature-related images, while pet lovers may prefer images of cute animals. Therefore, when determining recommended image data, it's crucial to consider not only the aesthetic appeal of the images but also the user's personal preferences. Identifying images with special meaning for users from the vast library makes it easier to resonate with them. Therefore, in the process of image selection, comprehensively considering image aesthetics, user semantic preferences, and user preferences for specific images is particularly important.

[0117] Figure 7 A schematic flowchart of an image data recommendation method is shown. Figure 7 The image data recommendation method shown includes steps S701 to S706.

[0118] Step S701: Normalize the amount of user interaction with each image data in the gallery across multiple interaction categories.

[0119] For a given interaction category, the user's interaction volume for each image data point across multiple interaction categories can be divided by the maximum value among those interaction volumes to obtain the normalized result for each image data point within that interaction category. Through normalization, the interaction volume is normalized to a scale of 0 to 1.

[0120] Step S702: Calculate the interaction score for each image data. The interaction score for each image data is the sum of the normalized results of multiple interaction quantities of that image data.

[0121] The interaction score for image data takes the image as the dimension and comprehensively considers the amount of interaction the user has with the image data under the interaction category, representing the user's degree of preference for the image data.

[0122] Step S703: Add the interaction amounts of image data containing the same semantics to the corresponding interaction amounts of each interaction category to obtain the interaction amount corresponding to that semantics.

[0123] Semantic recognition of image data can determine the semantics contained within the content recorded in the image data. The semantics contained in image data can include various objects, scenes, and / or people, etc. The semantics contained within the content recorded in the image data can also be understood as the semantics contained within the image data itself.

[0124] Step S704: Calculate the semantic score for each image data. The semantic score for each image data is the sum of the interaction quantities corresponding to at least one semantic contained in the image data.

[0125] Different users have different preferences for different semantics. To better protect consumer privacy, it is necessary to provide users with adaptive image selection results without using personal information (such as gender, age, etc.). Through steps S703 to S704, the semantic score of the image data is determined based on the user's interaction volume with image data containing a certain semantic meaning in each interaction category. The semantic score of the image data represents the user's overall preference for the semantics of that image data.

[0126] Step S705: Calculate the image recommendation score for each image data. The image recommendation score for each image data is the sum of the aesthetic score, semantic score, and interaction score of that image data.

[0127] The aesthetic score for each image is used to indicate how pleasing the image is to the eye. In other words, the aesthetic score of an image reflects its aesthetic appeal to the user.

[0128] Step S706: Determine recommended image data from multiple image data in the image library according to the image recommendation score.

[0129] Recommended image data can be a preset number of images from the gallery that have the highest recommendation scores.

[0130] Through steps S701 to S706, the interaction score, semantic score, and aesthetic score of the image data are determined, and the image recommendation score of the image data is determined based on the interaction score, semantic score, and aesthetic score of the image data. Image data is then recommended to the user so that the recommended image data is more in line with the user's preferences.

[0131] However, some abnormal and extreme data can have a significant impact on image recommendation scores, thereby affecting the accuracy of the image data recommended to users.

[0132] For example, some image data contains identity verification information, and users frequently browse this image data to authenticate their identities. In other words, this image data is used very frequently, but users may not have a high level of interest in it.

[0133] This situation leads to users interacting with a significant amount of some image data in a particular interaction category compared to other image data, with the interaction volume for other image data potentially differing by at least an order of magnitude. Step S701 normalizes the interaction volume for this particular interaction category, and the normalized result for the interaction volume of the other image data is extremely small, close to 0. The differences in the interaction scores among these other image data are very small, and the interaction scores of the image data cannot accurately reflect the user's preference for the image data.

[0134] Similar to the interaction score, if a user interacts with some image data far more than with other image data in a certain interaction category, it will also have a significant impact on the semantic score, making the semantic score unable to reflect the user's degree of semantic preference.

[0135] Therefore, image data recommended to users based on interaction scores and semantic scores cannot accurately reflect user preferences and cannot achieve accurate image recommendations.

[0136] To address the aforementioned issues, this application provides a media data recommendation method that can reduce the impact of extreme data and make the media data recommended to users more in line with their preferences.

[0137] The following is combined Figures 8 to 9 The media data recommendation method provided in the embodiments of this application will be described in detail.

[0138] Figure 8 This is a schematic flowchart illustrating a media data recommendation method provided in an embodiment of this application. The image data recommendation method may include steps S801 to S804, which are described in detail below.

[0139] Step S801: Obtain parameter values ​​for multiple parameter categories for each media data information in the multiple media data information, where each media data information corresponds to at least one media data.

[0140] Multiple media data information corresponds to multiple media data, which can have the same data type. For example, the multiple media data can all be image data, all be audio data, all be text data, or all be video data, etc.

[0141] The parameter value for each parameter category of media data information is used to represent the degree of user preference for media data information under that parameter category.

[0142] Step S802: Determine the first ranking of each media data information in the plurality of parameter categories according to the order of the parameter values.

[0143] For each parameter category, the media data information is sorted according to the magnitude of the parameter values. The first ranking of a media data piece within a certain parameter category is its sequence number within that parameter category's sorting results.

[0144] Step S803: Determine the first priority of the media data information based on the first ranking of each media data information in the multiple parameter categories.

[0145] It should be understood that the same sorting method should be used for different parameter categories, that is, sorting them in descending order or ascending order. Therefore, the sorting results for different parameter categories can reflect changes in users' interactive preferences for media data information.

[0146] Step S804: Based on the first priority of the plurality of media data information, determine recommended media data from the plurality of media data corresponding to the plurality of media data information.

[0147] In some embodiments, different parameter categories can be different interaction categories. The parameter value of a certain parameter category of media data information can be the amount of interaction between the user and at least one piece of media data corresponding to that media data information in that interaction category.

[0148] Interaction categories can include the number of views (or plays), the duration of views (or plays), the number of edits, the number of shares, etc.

[0149] Different media data can be different media data in general. The first priority of media data can be understood as the interaction score of the media data, which represents the degree of user preference for each media data in terms of interaction volume. Based on the interaction scores of multiple media data, recommended media data can be determined from among them.

[0150] The units of measurement for interaction volume may differ across different interaction categories, and there can be significant numerical differences in interaction volume between the same or different interaction categories. By sorting media data in multiple interaction categories according to the magnitude of interaction volume, and determining the interaction score based on the first ranking of each media data in each interaction category, the significant impact of the interaction volume of individual media data with large numerical differences from other media data in a particular interaction category on the accuracy of the interaction score can be avoided or reduced, thereby improving the accuracy of recommended media data.

[0151] Different media data information can also be classified into different semantic categories.

[0152] Each piece of media data includes one or more semantic categories. Semantic categories include object, scene, time, location, etc.

[0153] As a semantic category, an object refers to the content recorded in image data, including but not limited to people, furniture, toys, food, sports equipment, tools, books, architectural elements (e.g., doors, windows, houses, buildings, pillars, etc.), vehicles, natural elements (e.g., trees, grass, flowers, plants, waterfalls, waves, etc.), and celestial phenomena and planetary bodies (e.g., the sun, clouds, rain, snow, the earth, stars, comets, etc.).

[0154] Scene, as a semantic category, refers to the location or positional information recorded in image data, also known as a point of information (POI). Scenes include, but are not limited to, beaches, forests, oceans, rivers, cityscapes, concert and performance venues, indoor spaces, professional and amateur sports venues, commercial or office spaces, residences, shops, bars, gas stations, hospitals, and train stations.

[0155] The objects and scenes in a given image data can be obtained by identifying that image data.

[0156] Time, as a semantic category, refers to the time information generated by image data, including holidays, weekdays, rest days, and time periods of the day (e.g., morning, noon, afternoon, night, early morning, etc.).

[0157] Time, as a semantic category, refers to the location information of the terminal device when image data is generated, used to indicate the recording or shooting location, etc.

[0158] The first priority of a semantic category can be understood as the degree of user preference for that semantic category. The interaction volume of a semantic category within a certain interaction category can be obtained by adding the user's interaction volume with at least one image data containing that semantic category within that interaction category.

[0159] After determining the first priority of the semantic category, i.e., determining the user's preference for the semantic category, in step S804, for each piece of media data, the first priorities of multiple semantic categories contained in the media data can be weighted and accumulated to obtain the semantic score of the media data. Therefore, recommended media data is determined based on the semantic score of the media data.

[0160] The weights used in calculating the semantic score of each media data point can be preset or determined based on the theme, etc.

[0161] Topics are used to represent the type of recommended media data. Topics may include, for example, weekend moments, afternoon relaxation, hiking, gatherings, the allure of flowers, and recommendations based on location.

[0162] The correspondence between topics and weight combinations can record the weights corresponding to multiple semantic types under different topics. The weights corresponding to the same semantic type can be the same or different in weight combinations corresponding to different topics. For example, compared with the topic "the charm of flowers", in the weight combination corresponding to the topic "hiking", the semantic categories "mountains" and "rivers" that are related to the outdoors have greater weights.

[0163] After obtaining the recommended topics, for each piece of media data, the first priority of at least one semantic category contained in the media data can be weighted and accumulated according to the weight combination corresponding to the recommended topic to obtain the semantic score of the media data.

[0164] Recommended topics can be obtained in two ways: first, by selecting one or more topics from a set of preset options; second, by using a user-selected topic as the recommended topic. The user's selection can be made on a user interface that includes multiple topics.

[0165] Designers can update the mapping between themes and weight combinations based on recent popular activities and topics. This new mapping can be set in the cloud, communicating with the terminal devices to update the mapping between themes and weight combinations on the devices.

[0166] The different weight combinations corresponding to different themes reflect the varying needs when recommending media data. Providing users with recommended media data based on the recommended themes increases the diversity of recommended media data and improves the user experience.

[0167] Similar to determining the interaction score of media data, determining the semantic score of media data requires using the amount of user interaction with the media data across multiple interaction categories. The units of measurement for interaction volume may differ between different interaction categories, and there can be significant numerical differences between interaction volumes within the same or different interaction categories. By sorting the semantic categories according to the magnitude of interaction volume across multiple interaction categories, and determining the first priority of the semantic category based on its first ranking in each interaction category, we can avoid or reduce the significant impact of the interaction volume of individual media data with large numerical differences from other media data in a particular interaction category on the accuracy of the first priority of the semantic category. This improves the accuracy of the semantic score of the media data, thereby improving the accuracy of the determined recommended media data.

[0168] When media data information corresponds to different media data, after determining the first priority of the media data, the recommended media data can be determined from multiple media data based on the first priority and the second priority of the media data.

[0169] During step S804, the second priority of each media data can be obtained.

[0170] The second priority of media data is used to indicate the degree of user preference for the media data in other categories besides interaction. There can be one or more second priorities for media data. For example, the second priority of media data may include semantic scores, aesthetic scores, etc. That is, the categories can be semantic, aesthetic, etc.

[0171] The aesthetic score of media data is used to represent the aesthetic appeal of the media data.

[0172] The semantic score of media data can be obtained by weighted summation of the interaction volume of media data containing the semantics in each interaction category, or by ranking the interaction volume of media data containing the semantics in each interaction category in order of magnitude, and then weighted summation of the rankings of the media data in each interaction category.

[0173] For media data with a first priority and at least one category as a second priority, an image recommendation model can be used to determine whether the media data is recommended. The image recommendation model is a trained neural network model. For details on how to train an image recommendation model, please refer to [link to documentation / reference]. Figure 9 Explanation.

[0174] Alternatively, a recommendation score can be determined for each media data point based on a first priority of multiple media data points and a second priority of at least one category, thereby allowing the selection of recommended media data based on the recommendation score.

[0175] The recommendation score for media data can be obtained by weighted summing of the media data's first priority and at least one category's second priority. Alternatively, the recommendation score for media data can be determined by ranking the media data's first priority among multiple media data's first priorities, and ranking the media data's second priority for each category among multiple media data's second priorities for that category.

[0176] The second ranking of the multiple media data can be determined according to the order of their second priority. Having obtained the second priority of the media data in multiple categories, the multiple media data can be ranked according to the second priority of each category to determine the second ranking of the media data in each category.

[0177] Based on the order of priority, the third ranking of media data can be determined.

[0178] Based on the second and third rankings of each media data point, recommended media data can be determined.

[0179] For the third-ranked and at least one second-ranked media data point each day, a weighted sum can be applied, or a weighted average of the third-ranked and at least one second-ranked media point can be calculated. This result is then used as the recommendation score for that media data point. Recommended media data is determined based on the recommendation score.

[0180] In determining the recommendation score of media data based on its priority across various categories, compared to simply weighting and summing the priorities of different media data, determining the recommendation score based on the priority ranking of a particular media data across multiple categories avoids the impact of anomalies such as individual priority values ​​in a category significantly exceeding those of others, thus improving the accuracy of the recommendation score in reflecting user preferences for media data.

[0181] In other embodiments, the multiple parameter categories for each media data information include at least two of the following: an interaction score, a semantic score, and an aesthetic score. The interaction score is determined based on the amount of interaction between the user and the media data; the semantic score represents the user's preference for the semantics of the media data; and the aesthetic score represents the aesthetic appeal of the media data.

[0182] Through steps S801 to S804, in the process of determining recommended media data, the parameter values ​​of each media data in multiple parameter types are no longer weighted and accumulated. Instead, the recommended media data is determined from the multiple media data according to the ranking of the parameter values ​​of each media data in multiple parameter types in order of magnitude. This avoids the impact of individual parameter values ​​with large numerical differences on the accuracy of the determined recommended media data, so that the recommended media data can accurately reflect the user's preferences.

[0183] Figure 9 This is a schematic flowchart of a media data recommendation method provided in an embodiment of this application, with image data as an example for illustration. Figure 9 The media data recommendation method shown may include steps S901 to S904, which are described in detail below.

[0184] Step S901: Based on the user's interaction volume with each image data in the image library across multiple interaction categories, sort the multiple image data in the image library to obtain the interaction score for each image data.

[0185] The image data in the gallery may include image data captured and stored by the terminal device through a camera, or image data obtained by the terminal device from the cloud.

[0186] User interactions with images include clicking to view, editing, and sharing. The frequency and duration of these interactions reveal user preferences for the images.

[0187] Table 1 is a statistical table of user interaction behavior with image data, recording the statistical results of user interaction volume with each image data in the gallery under various interaction categories. Table 1 provides statistics on user interaction volume with image data under various interaction categories, including click count, edit count, share count, and browsing duration.

[0188] Table 1. Statistics on User Interaction Behavior with Image Data

[0189] It should be understood that the statistical results in Table 1 can be statistics on user interaction behavior with image data within a preset time period. For example, user preferences may change over time, and the preset time period could be the time period corresponding to the most recent preset duration; weekday user use of the image library is more likely due to work needs, so the preset time period could also be a weekend within the most recent preset duration period. Setting the preset time period based on time length, time of day, and whether it is a rest day ensures that the statistics on user interaction behavior align with user preferences.

[0190] The dimensions of interaction quantities across different interaction categories may differ. Normalization helps avoid the influence of these dimensions on the calculation. However, calculating user preferences for image data by summing the interaction quantities across all categories may result in significant errors.

[0191] In some cases, image data contains identity verification information, and users frequently browse this image data for authentication. In other words, this image data is used very frequently, but users may not have a strong interest in or use it. In other cases, a user's browsing of image data on their device is abruptly interrupted, resulting in the device displaying a particular image for an extended period. Statistics show that the user's interaction with this image in terms of browsing time far exceeds that with other image data.

[0192] These circumstances lead to a situation where, in a certain interaction category, the amount of user interaction with some image data far exceeds the amount of interaction with other image data, with the interaction volume of other image data potentially differing by at least an order of magnitude. If the user's preference for image data is calculated by summing the interaction volumes of image data across all categories, then in a certain interaction category, the interaction volume of other image data becomes negligible compared to the interaction volume of the aforementioned image data. The sum of the interaction volumes of the aforementioned image data across all interaction categories is greater than the sum of the interaction volumes of the other image data across all interaction categories, resulting in a deviation between the determined preference level and the actual situation.

[0193] To reduce this bias, the image data are sorted according to the amount of interaction within each interaction category. If the number of image data in the library is N and the number of interaction categories is K, then the ranking of the nth image data in the kth interaction category can be represented as: N, K, n, and k are all positive integers. The sorting result can be represented as a matrix. :

[0194] Rank the nth image data among the K interaction categories. to The interaction scores are summed to obtain the interaction score for each image data point. The interaction score represents the user's preference for the image data. The interaction score for the nth image data point is... It can be represented as:

[0195] Different user actions on specific image data imply user preferences for that image data. In step S901, taking the image as the dimension, the interaction volume of users with different interaction categories on the image data is comprehensively considered, and an interaction score is used to represent the degree of user preference for that image data.

[0196] Based on the number of interactions users have with each image in the gallery across multiple interaction categories, the images in the gallery are sorted, and an interaction score is obtained based on the ranking of the images. Compared to the method of accumulating the interaction volume of image data across multiple interaction categories, this method avoids the situation where the interaction score of a particular image is necessarily higher than the interaction scores of other image data in the same interaction category, even if the user's interaction volume for a particular image in a certain interaction category far exceeds the user's interaction volume for other image data in that category. This improves the accuracy of the interaction score in representing the degree of user preference.

[0197] Step S902: Based on the amount of user interaction with the image data in multiple interaction categories, the semantic categories of the image data are sorted to obtain the semantic score of the image data.

[0198] For newly added image data in the gallery, the number of user interactions with this image data is relatively small, and the statistical data shows that the number of user interactions with this image data in each interaction category is also small. Therefore, it is necessary to use more generalized user interaction data to select the best image data.

[0199] For image data, semantic categories include objects, scenes, time, etc.

[0200] An object, as a semantic category, refers to the content recorded in image data. A scene, as a semantic category, refers to the location or positional information recorded in image data. Time, as a semantic category, refers to the time information when the image data was generated. The objects and scenes of a given image can be identified by recognizing that image data.

[0201] The interaction amount of a semantic category is obtained by summing the interaction amounts of a user on at least one image data containing a certain semantic category in a certain interaction category.

[0202] Table 2 is a statistical table of user interaction behavior across multiple semantic categories, recording the statistical results of user interaction volume across various semantic categories and interaction types. Table 2 provides statistics on user interaction volume with image data containing certain semantics across multiple interaction categories. These multiple interaction categories include click counts, edit counts, share counts, and browsing duration.

[0203] Table 2. Statistics on User Interaction Behaviors Across Multiple Semantic Categories

[0204] For M semantic categories in the library, if the number of interaction categories is K, then the rank of the m-th semantic category in the j-th interaction category can be represented as: M and m are both positive integers. The sorting result can be represented as a matrix. :

[0205] Rank the m-th semantic category among the K interaction categories. to The interaction preferences for each semantic category are summed to obtain the interaction preference for that semantic category. The interaction preference for a semantic category represents the degree of user preference for that semantic category. The interaction preference parameter for the m-th semantic category... It can be represented as:

[0206] Based on the interaction preference parameters of each semantic category, the semantic scores of N image data in the image library can be determined. The semantic score of the nth image data is obtained by summing the interaction preference parameters of the semantic categories contained in the nth image data. For example, the semantic score of the nth image data can be expressed as the ratio of the sum of the interaction preference parameters of the semantic categories included in the nth image data to the number of semantic categories included in the nth image data.

[0207] When the weights corresponding to each semantic category are equal, the semantic score of the nth image data can be expressed as the average of the interaction preference parameters of the semantic categories included in the nth image data. For example, if the nth image data includes the 2nd and 3rd semantic categories out of M semantic categories, then the semantic score of the nth image data is:

[0208] Multiple themes can correspond to multiple weight combinations. Under a target theme, the interaction preference parameters of the semantic categories included in the image data can be accumulated using the weight combination corresponding to that target theme, thereby obtaining the semantic score of the image data under that target theme.

[0209] In other words, semantic categories can be assigned weights appropriate to different themes. The semantic score of the nth image data under the target theme can be represented as the weighted average of the interaction preference parameters of the semantic categories included in the nth image data.

[0210] For example, under a certain topic, the weights corresponding to the second and third semantic categories are respectively... and Then, the semantic score of the nth image data packet, which includes the second and third semantic categories out of M semantic categories, is:

[0211] The mapping between topics and weight combinations can be set based on prior knowledge. This mapping can be configured by developers on the terminal device, or developers can configure it in the cloud, with the terminal device retrieving the mapping from the cloud.

[0212] Different users have different semantic preferences. Through S902, considering the interaction volume of users with multiple image data sets in each interaction category, as well as the semantic categories within those image data sets, interaction preference parameters for multiple semantic categories are determined. Based on the semantic categories contained in a particular image data set, a semantic score for that image data set is determined. The semantic score of an image data set represents the degree of user preference for the semantic categories within that image data set.

[0213] Based on the amount of user interaction with each image in the gallery across multiple interaction categories, the semantic categories of the image data are ranked, and a semantic score is obtained based on the ranking of the semantic categories. Compared to the method of accumulating the interaction volume of image data across multiple interaction categories, this reduces the impact of the interaction volume of a particular image in a certain interaction category on the interaction score of multiple image data when the user's interaction volume with a particular image in a certain interaction category far exceeds the user's interaction volume with other image data in that interaction category.

[0214] Step S903: Obtain the aesthetic score for each image data.

[0215] With the development of computer vision-related technologies, computers can simulate human aesthetic abilities to score images from an aesthetic perspective.

[0216] The aesthetic scores of multiple image data can be obtained by calling the aesthetic scoring module to score each of the multiple image data separately to obtain the aesthetic score of each image data; or, the aesthetic scores of multiple image data can be obtained by reading the aesthetic score of each image data from the memory.

[0217] Step S904: Determine recommended image data from the plurality of image data based on the interaction score, semantic score, and aesthetic score of each image data.

[0218] In some embodiments, an image recommendation score can be determined for each image data based on interaction scores, semantic scores, and aesthetic scores of multiple image data, thereby determining recommended image data based on the image recommendation scores.

[0219] For any image data, the interaction score, semantic score, and aesthetic score of the image data are weighted and summed. The result can be used as the image recommendation score for that image data.

[0220] Interaction score, semantic score, and aesthetic score can all be positively correlated with the user's level of preference. The recommended image data can be a preset number of images with the highest possible recommended image score.

[0221] Alternatively, for any given image data, multiple images can be sorted according to their interaction score, semantic score, and aesthetic score. The image recommendation score for any given image can be obtained by weighted summation of its ranking in the order of interaction score, semantic score, and aesthetic score.

[0222] Interaction score, semantic score, and aesthetic score can all be positively correlated with user preference. When sorting each image data point based on its interaction score, semantic score, and aesthetic score, the sorting can be done in descending or ascending order. If sorted in ascending order, the recommended image data can be a preset number of images with the highest recommended scores.

[0223] The image recommendation score is determined by ranking multiple image data according to their interaction score, semantic score, and aesthetic score. This avoids the significant impact of individual scores that differ greatly from the majority of scores in any of the three categories on the image recommendation score, thus improving the accuracy of the image recommendation score in reflecting the user's preference for the image data.

[0224] In other embodiments, inputting the interaction score, semantic score, and aesthetic score of image data into an image recommendation model can determine whether the image data is recommended image data.

[0225] An image recommendation model is a trained neural network model. By training an initial image recommendation model using training data, an image recommendation model can be obtained.

[0226] During neural network training, to ensure the output closely approximates the desired predicted value, we compare the network's prediction with the target value and update the weight vector of each layer based on the difference. (Of course, there's usually an initialization process before the first update, where parameters are pre-configured for each layer.) For example, if the network's prediction is too high, the weight vector is adjusted to predict a lower value. This adjustment continues until the neural network can predict the target value or a value very close to it. Therefore, we need to predefine "how to compare the difference between the predicted and target values," which is the loss function or objective function. These are important equations used to measure the difference between the predicted and target values. Taking the loss function as an example, a higher output value (loss) indicates a greater difference, so training the neural network becomes a process of minimizing this loss.

[0227] Neural networks can employ backpropagation (BP) to correct the parameters of the initial neural network model during training, thereby reducing the reconstruction error loss. Specifically, forward propagation of the input signal to the output generates error loss; this error loss information is then propagated back to update the parameters of the initial neural network model, leading to convergence of the error loss. The backpropagation algorithm is an error-loss-driven backpropagation process aimed at obtaining the optimal parameters of the neural network model, such as the weight matrix.

[0228] The training data includes a training score and training label for each training image in a dataset of multiple training images. The training scores for each training image include interaction scores, semantic scores, and aesthetic scores. The training labels for each training image indicate whether it is suitable for training recommendation images.

[0229] The training scores of the training image data can be obtained in the same way as the interaction scores, semantic scores, and aesthetic scores recorded in steps S901 to S903.

[0230] Training labels for training image data can be obtained based on whether the user clicks on the training image data. Multiple training image data are displayed using the user interface; the training label of a clicked training image data indicates that it is a recommended training image data, while the clicked training label indicates that it is not a recommended training image data.

[0231] The initial image recommendation model is used to process the training scores of the training image data to obtain the training output. Based on the training output, the parameters of the initial image recommendation model are adjusted to minimize the difference between the training output and the training labels, thereby bringing the difference between the training output and the training labels to converge. The initial image recommendation model with adjusted parameters is the trained image recommendation model.

[0232] For example, the training output can be represented by a numerical value in the range [0, 1]. The training output of the training image data represents the probability that the training image data is the recommended image data. If the training output of the training image data is greater than or equal to a preset value, it indicates that the training image data is the recommended image data for training.

[0233] Therefore, by using training image data, the semantic score, interaction score, and aesthetic score of the image data can be processed to obtain the probability that the image data is recommended image data. If this probability is greater than a certain value, the image data can be considered recommended image data.

[0234] Alternatively, using training image data, the semantic score, interaction score, and aesthetic score of multiple image data in the image library can be processed separately to obtain the probability that each image data is a recommended image data. The recommended image data can be a preset number of images with the highest probability.

[0235] Step S904 comprehensively considers the aesthetic appeal of the image, the user's personal semantic preferences, and the user's preference for specific images, thereby integrating aesthetic scores, semantic scores, and interaction scores to make the recommended image data more in line with user preferences.

[0236] Through steps S901 to S904, the impact of outlier data with significant differences from other data values ​​on the calculation results is reduced during the calculation of interaction and semantic scores. When determining the image recommendation score using the ranking of interaction, semantic, and aesthetic scores of image data, the influence of outlier data with large numerical differences on the image recommendation score is also reduced. Therefore, the accuracy of the determined recommended image data is improved, making the recommended image data more in line with user preferences.

[0237] It should be understood that the above examples are provided to help those skilled in the art understand the embodiments of this application, and are not intended to limit the embodiments of this application to the specific values ​​or scenarios illustrated. Those skilled in the art can obviously make various equivalent modifications or changes based on the above examples, and such modifications or changes also fall within the scope of the embodiments of this application.

[0238] The above text combined Figures 1 to 9 The media data recommendation method of this application embodiment is described in detail below, and will be combined with Figure 10 and Figure 11 This document describes in detail the apparatus embodiments of this application. It should be understood that the media data recommendation apparatus in the embodiments of this application can execute the various media data recommendation methods described in the foregoing embodiments of this application. That is, the specific working processes of the various products described below can be referred to the corresponding processes in the foregoing method embodiments.

[0239] Figure 10 This is a schematic diagram of the media data recommendation device provided in the embodiments of this application.

[0240] It should be understood that the media data recommendation device 1000 can perform... Figures 8 to 9 The media data recommendation method shown; the media data recommendation device 1000 includes: an acquisition unit 1010 and a processing unit 1020.

[0241] The acquisition unit 1010 is used to acquire parameter values ​​of multiple parameter categories for each media data information in multiple media data information, where each media data information corresponds to at least one media data.

[0242] The processing unit 1020 is used to determine the first ranking of each media data information in the plurality of parameter categories according to the order of the parameter values.

[0243] The processing unit 1020 is further configured to determine the first priority of the media data information based on the first ranking of each media data information in the plurality of parameter categories.

[0244] The processing unit 1020 is further configured to determine recommended media data from the multiple media data corresponding to the multiple media data information based on the first priority of the multiple media data information.

[0245] Optionally, different parameter categories are different interaction categories, and the parameter value of each parameter category of each media data information is the amount of interaction between the user and at least one media data corresponding to the media data information in that parameter category.

[0246] Alternatively, different media data information may fall into different semantic categories.

[0247] The processing unit 1020 is specifically configured to: for each piece of media data, weighted summation of the first priority of at least one semantic category contained in the media data to obtain a semantic score of the media data; and determine the recommended media data based on the semantic scores of the multiple pieces of media data.

[0248] Optionally, the acquisition unit 1010 is also used to acquire recommended topics, different topics correspond to different weight combinations, and each weight combination includes the weights corresponding to each semantic category.

[0249] The processing unit 1020 is specifically used to: for each piece of media data, according to the weight combination corresponding to the recommended topic, weight and accumulate the first priority of at least one semantic category contained in the media data to obtain the semantic score of the media data.

[0250] Alternatively, different media data information can be different media data.

[0251] The acquisition unit 1010 is also used to acquire the second priority of each media data in the plurality of media data corresponding to the plurality of media data information.

[0252] The processing unit 1020 is specifically configured to: determine a second ranking of the plurality of media data according to the second priority order; determine a third ranking of the plurality of media data according to the first priority order; and determine the recommended media data based on the second ranking and the third ranking of each media data.

[0253] Optionally, the multiple parameter categories of each media data information include at least two of the media data's interaction score, semantic score, and aesthetic score. The interaction score is determined based on the amount of interaction between the user and the media data. The semantic score is used to represent the user's preference for semantic categories in the media data, and the aesthetic score is used to represent the aesthetic appeal of the media data.

[0254] It should be noted that the aforementioned media data recommendation device 1000 is embodied in the form of a functional unit. The term "unit" here can be implemented in software and / or hardware, without specific limitations.

[0255] For example, a "unit" can be a software program, a hardware circuit, or a combination of both that implements the above functions. The hardware circuit may include an application-specific integrated circuit (ASIC), electronic circuitry, a processor (e.g., a shared processor, a proprietary processor, or a group processor) and memory for executing one or more software or firmware programs, integrated logic circuitry, and / or other suitable components that support the described functions.

[0256] Therefore, the units of the various examples described in the embodiments of this application can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0257] Figure 11 A schematic diagram of the structure of an electronic device provided in this application is shown. Figure 11 The dashed lines in the diagram indicate that the unit or module is optional. Electronic device 1100 can be used to implement the media data recommendation method described in the above method embodiments.

[0258] Electronic device 1100 includes one or more processors 1101 that support the media data recommendation method in the method embodiments of electronic device 1100. Processor 1101 can be a general-purpose processor or a special-purpose processor. For example, processor 1101 can be a central processing unit (CPU), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices such as discrete gates, transistor logic devices, or discrete hardware components.

[0259] The processor 1101 can be used to control the electronic device 1100, execute software programs, and process data from the software programs. The electronic device 1100 may also include a communication unit 1105 for inputting (receiving) and outputting (transmitting) signals.

[0260] For example, electronic device 1100 may be a chip, communication unit 1105 may be the input and / or output circuit of the chip, or communication unit 1105 may be the communication interface of the chip, and the chip may be a component of terminal device or other electronic device.

[0261] For example, electronic device 1100 can be a terminal device, communication unit 1105 can be the transceiver of the terminal device, or communication unit 1105 can be the transceiver circuit of the terminal device.

[0262] The electronic device 1100 may include one or more memories 1102, which store a program 1104. The program 1104 can be executed by the processor 1101 to generate instructions 1103, causing the processor 1101 to execute the media data recommendation method described in the above method embodiments according to the instructions 1103.

[0263] Optionally, the memory 1102 may also store data. Optionally, the processor 1101 may also read the data stored in the memory 1102, which may be stored at the same memory address as the program 1104, or the data may be stored at a different memory address than the program 1104.

[0264] The processor 1101 and memory 1102 can be configured separately or integrated together; for example, integrated on the system on chip (SOC) of the terminal device.

[0265] For example, the memory 1102 can be used to store the relevant program 1104 of the media data recommendation method provided in the embodiments of this application. The processor 1101 can be used to call the relevant program 1104 of the media data recommendation method stored in the memory 1102 when performing image restoration on the terminal device, and execute the media data recommendation method of the embodiments of this application. For example, it can obtain the parameter values ​​of multiple parameter categories of each media data information in multiple media data information, each media data information corresponding to at least one media data; determine the first ranking of each media data information in the multiple parameter categories according to the order of the parameter values; determine the first priority of the media data information according to the first ranking of each media data information in the multiple parameter categories; and determine recommended media data among the multiple media data corresponding to the multiple media data information according to the first priority of the multiple media data information.

[0266] This application also provides a computer program product that, when executed by processor 1101, implements the media data recommendation method described in any of the method embodiments of this application.

[0267] The computer program product can be stored in memory 1102, for example, program 1104. Program 1104 is finally converted into an executable object file that can be executed by processor 1101 after processing such as preprocessing, compilation, assembly and linking.

[0268] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a computer, implements the media data recommendation method described in any of the method embodiments of this application. The computer program may be a high-level language program or an executable object program.

[0269] The computer-readable storage medium is, for example, memory 1102. Memory 1102 can be volatile memory or non-volatile memory, or memory 1102 can include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM).

[0270] In this application, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0271] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0272] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0273] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0274] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for example, the division of units is merely a logical functional division, and other division methods may exist in actual implementation; for example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0275] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0276] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0277] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A media data recommendation method, characterized in that, include: Obtain parameter values ​​for multiple parameter categories for each media data information from multiple media data information, where each media data information corresponds to at least one media data; For each of the multiple parameter categories, the multiple media data information is sorted according to the order of the parameter values ​​to determine the first ranking of each media data information in the multiple parameter categories; The first priority of the media data information is determined based on the first ranking of each media data information in the multiple parameter categories; Based on the first priority of the plurality of media data information, recommended media data is determined from the plurality of media data corresponding to the plurality of media data information.

2. The method according to claim 1, characterized in that, Different parameter categories represent different interaction categories. The parameter value for each parameter category of each media data information is the amount of interaction between the user and at least one media data corresponding to the media data information in that parameter category.

3. The method according to claim 2, characterized in that, Different media data information belongs to different semantic categories; the step of determining recommended media data from the multiple media data corresponding to the multiple media data information according to the first priority of the multiple media data information includes: For each piece of media data, the first priority of at least one semantic category contained in the media data is weighted and accumulated to obtain the semantic score of the media data; The recommended media data is determined based on the semantic scores of the multiple media data.

4. The method according to claim 3, characterized in that, The method further includes: obtaining recommended topics, where different topics correspond to different weight combinations, and each weight combination includes the weights corresponding to each semantic category; The step of weighted summing of the first priorities of at least one semantic category contained in the media data for each media data to obtain a semantic score for the media data includes: for each media data, weighted summing of the first priorities of at least one semantic category contained in the media data according to the weight combination corresponding to the recommended topic to obtain a semantic score for the media data.

5. The method according to claim 2, characterized in that, Different media data are different media data; The method further includes: obtaining the second priority of each media data among the multiple media data corresponding to the multiple media data information; The step of determining recommended media data from the multiple media data corresponding to the multiple media data information based on the first priority of the multiple media data information includes: The second ranking of the multiple media data is determined according to the size order of the second priority. The third ranking of the multiple media data is determined according to the order of the first priority. The recommended media data is determined based on the second ranking and the third ranking of each media data.

6. The method according to claim 1, characterized in that, Each media data information has multiple parameter categories, including at least two of the following: an interaction score, a semantic score, and an aesthetic score. The interaction score is determined based on the amount of interaction between the user and the media data. The semantic score is used to represent the user's preference for semantic categories in the media data. The aesthetic score is used to represent the aesthetic appeal of the media data.

7. A media data recommendation device, characterized in that, The media data recommendation device includes a processor and a memory, the memory being used to store a computer program, and the processor being used to call and run the computer program from the memory, causing the media data recommendation device to perform the media data recommendation method according to any one of claims 1 to 6.

8. A chip, characterized in that, Includes a processor, which, when executing instructions, performs the media data recommendation method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, causes the processor to perform the media data recommendation method according to any one of claims 1 to 6.

10. A computer program product, characterized in that, The computer program product includes: computer program code, which, when executed by a processor, causes the processor to perform the media data recommendation method according to any one of claims 1 to 6.

Images