Intelligent video recommendation method and device and storage medium

Abstract

The invention provides an intelligent video recommendation method and device and a storage medium, and relates to the technical field of video recommendation. Comprising the following steps: S1, labeling multiple label groups for a video; s2, constructing a dominant label preference library for the user, wherein the preference degree of the user for each dominant label is stored in the dominant label preference library; s3, in response to the historical video browsing record of the user, updating the preference degree of the user for each dominant tag; s4, obtaining a priority value of the to-be-recommended video in combination with the current popularity of the to-be-recommended video and a product of the preference degree of the user on the dominant tag of the to-be-recommended video and the weight factor; and S5, performing priority ranking on the to-be-recommended videos according to a descending order of the priority values, and selecting a first number of to-be-recommended videos ranked in front of the priority for recommendation. In the video recommendation process, the preference degree of the user to the event and different viewpoints of the public to the same event are considered, and the phenomenon that information cocoon rooms are generated due to the fact that the public excessively pays attention to the same viewpoint is avoided.

Description

Technical Field

[0001] The present application relates to the field of video recommendation technology, and in particular to an intelligent video recommendation method, device, and storage medium. Background Art

[0002] "Information cocoon" refers to people's obsession with a certain hobby in the information field, thus limiting their understanding and confining themselves to a "cocoon" like a silkworm cocoon, and becoming ignorant of other aspects that they are not interested in.

[0003] Nowadays, when short videos are popular, the algorithms used by most video recommendation platforms analyze the user's historical video browsing records, find topics that the user is interested in, and recommend corresponding videos, which can easily lead to the phenomenon of information cocoon.

[0004] The existing technology has also proposed some solutions that can avoid information cocoons. For example, the Chinese invention patent application with publication number CN117150081A proposes a video recommendation method, device, electronic device and storage medium. Based on the user's own interests, it places a certain degree of emphasis on the user's unplayed videos, thereby ensuring that the recommended videos are in line with the user's interests and have a certain degree of novelty, effectively avoiding the information cocoon effect, and thus improving the user experience.

[0005] However, the key to breaking down information cocoons isn't simply to recommend more points of interest, such as videos that cover specific events or expose social phenomena. More importantly, video publishers often express their own perspectives on the events, which are often subjective and biased. If video recommendation methods focus solely on videos that cover more events, they risk exposing users to only the same or similar perspectives on the same event, ultimately impacting society's understanding and judgment of specific events.

[0006] Therefore, how to propose a method that can balance different perspectives on the same event in a video is one of the difficult problems that needs to be solved urgently. Summary of the Invention

[0007] The present application provides an intelligent video recommendation method, device and storage medium to at least solve the above technical problems existing in the prior art.

[0008] According to a first aspect of the present application, there is provided an intelligent video recommendation method, comprising the following steps: S1, labeling a video with a multiple label group; the multiple label group includes explicit labels and implicit labels; S2, building an explicit tag preference library for the user, wherein the explicit tag preference library stores the user's preference for each explicit tag; S3, in response to the user's historical video browsing record, updating the user's preference for each explicit tag stored in the explicit tag preference library; S4, combining the current popularity of the video to be recommended and the user's preference for the explicit tag of the recommended video with the product of a weight factor to obtain a priority value for the video to be recommended, wherein the weight factor is negatively correlated with the number of times the implicit tag of the video to be recommended has been viewed in historical video browsing records; S5: sort the priority of the videos to be recommended in descending order of priority value, and select the videos to be recommended that are ranked first in priority for recommendation.

[0009] In certain embodiments of the first aspect of the present application, in S2, when constructing the explicit tag preference library, an initial preference degree is assigned to each explicit tag; and each time a new explicit tag is generated, the new explicit tag is updated to the explicit tag preference library and assigned an initial preference degree at the same time.

[0010] In certain embodiments of the first aspect of the present application, in S3, the method for updating the user's preference for each explicit tag stored in the explicit tag preference library in response to the user's historical video browsing record is as follows: S301, setting a first time threshold and a second time threshold, wherein the first time threshold is smaller than the second time threshold; S302, extracting the user's historical video browsing records within a specified time period; S303, counting the browsing time of each video and the explicit tags in the video's multiple tag groups in the historical video browsing records; S304, if the viewing time of the video is greater than the first time threshold, then in the explicit tag preference library, the user's preference for the explicit tags is increased; if the viewing time of the video is less than the first time threshold, then in the explicit tag preference library, the user's preference for the explicit tags is reduced; if the viewing time of the video is between the first time threshold and the second time threshold, then in the explicit tag preference library, the user's preference for the explicit tags is kept unchanged.

[0011] In certain embodiments of the first aspect of the present application, in S3, the method for updating the user's preference for each explicit tag stored in the explicit tag preference library in response to the user's historical video browsing record is as follows: S310 , each video is configured with a feedback mechanism, which is triggered during or after video playback to display explicit tags of the video to the user for positive or negative feedback; S311, setting a first time threshold and a second time threshold, wherein the first time threshold is smaller than the second time threshold; S312, extracting the user's historical video browsing records within a specified time period; S313, collecting statistics on the browsing time of each video, the explicit tags in the video's multiple tag groups, and the user's feedback on the explicit tags in the historical video browsing records; S314: If the video viewing time is greater than the first time threshold, then in the explicit tag preference library, the user's first preference for the explicit tags is increased; if the video viewing time is less than the first time threshold, then in the explicit tag preference library, the user's second preference for the explicit tags is decreased; if the video viewing time is between the first time threshold and the second time threshold, then in the explicit tag preference library, the user's preference for the explicit tags remains unchanged; S315, if the user provides positive feedback on the explicit tags, the user's third preference for the explicit tags will be increased; conversely, if the user provides negative feedback on the explicit tags, the user's fourth preference for the explicit tags will be reduced; if the user does not provide any feedback on the explicit tags, the user's preference for the explicit tags will remain unchanged.

[0012] In certain embodiments of the first aspect of the present application, in S4, the method for obtaining the priority value of the video to be recommended by combining the current popularity of the video to be recommended and the product of the user's preference for the explicit tag of the recommended video and the weight factor is as follows: S401, constructing a weight factor that is negatively correlated with the implicit tag of the video to be recommended; S402, calculating the current popularity of the video to be recommended; S403: Calculate the current popularity of the video to be recommended and the sum of the product of the user's preference for the explicit tag of the recommended video and the weight factor as the priority value of the video to be recommended.

[0013] In certain implementations of the first aspect of the present application, in S401, the method for constructing a weight factor that is negatively correlated with the implicit tag of the video to be recommended is as follows: Extract the user's historical video browsing records within a specified time period, count the videos containing the explicit tags corresponding to the implicit tags of the video to be recommended, and record them as the first videos; Counting the total number of first videos and the number of first videos containing the implicit tag of the video to be recommended; The ratio of the number of first videos containing the implicit tag of the video to be recommended to the total number of first videos is calculated, and the weight factor is obtained by subtracting the ratio from a constant.

[0014] In certain implementations of the first aspect of the present application, in S402, the method for calculating the current popularity of the video to be recommended is as follows: Extract the explicit tag of the video to be recommended, count the number of times all users have viewed the video to be recommended within the specified time period, and record it as the first time; The current popularity of the video to be recommended is obtained by dividing the first number of times by the set number threshold.

[0015] In certain embodiments of the first aspect of the present application, the basis for determining whether the user has viewed the video to be recommended is as follows: when the user's browsing time for the video to be recommended is greater than a first time threshold, and / or the user provides positive feedback on the video to be recommended, it is deemed that the user has viewed the video to be recommended.

[0016] In certain embodiments of the first aspect of the present application, if a user browses a video tagged with a certain explicit tag for a time greater than a first time threshold and the number of times reaches a set number threshold, and / or the number of times a user provides positive feedback on a certain explicit tag reaches a set number threshold, the user will be marked as a follower of the explicit tag; In S1, the notification videos for events published by the authenticated video publishers are marked with corresponding explicit tags and notification tags; and when priority is sorted in S5, if one of the recommended videos is marked with a notification tag, when it is recommended to the following users of the corresponding explicit tag, the priority of the recommended video marked with the notification tag is set to the highest and is ranked first.

[0017] According to a second aspect of the present application, an intelligent video recommendation system is provided, comprising: A labeling module is used to label a video with a multiple label group; the multiple label group includes explicit labels and implicit labels; A preference building module builds an explicit tag preference library for the user, wherein the explicit tag preference library stores the user's preference for each explicit tag; A preference updating module, in response to the user's historical video browsing record, updates the user's preference for each explicit tag stored in the explicit tag preference library; A priority calculation module calculates the priority value of the video to be recommended by combining the current popularity of the video to be recommended and the product of the user's preference for the explicit tag of the recommended video and a weight factor. The weight factor is negatively correlated with the number of times the implicit tag of the video to be recommended has been viewed in historical video browsing records. The sorting recommendation module sorts the priority of the recommended videos in descending order of priority value, and selects the first number of recommended videos to be recommended for recommendation.

[0018] According to a third aspect of the present application, an electronic device is provided, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor. The instructions are executed by the at least one processor to enable the at least one processor to perform the method described in this application.

[0019] According to a fourth aspect of the present application, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the method described in the present application.

[0020] Compared with the prior art, this application has the following beneficial effects: This application is a video tagging multiple tag group. On the one hand, it combines the user's historical video browsing history to update the user's preference for each explicit tag, so that videos with explicit tags with higher preference are recommended first in the future. On the other hand, it combines the weight factor for weighted summation. The weight factor is negatively correlated with the number of times the implicit tag has been viewed in the historical video browsing history. For videos with the same explicit tag, videos with different implicit tags are recommended first. In this way, in the process of video recommendation, the user's preference for the event and the public's different views on the same event are taken into account, avoiding the phenomenon of excessive focus on the same viewpoint and the formation of information cocoons.

[0021] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The above and other objects, features and advantages of the exemplary embodiments of the present application will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present application are shown in an illustrative and non-limiting manner, in which: In the drawings, the same or corresponding reference numerals denote the same or corresponding parts.

[0023] Figure 1 The figure shows the overall flow chart of the intelligent video recommendation method of the present application.

[0024] Figure 2 Shown is a diagram of the multi-tag group architecture of the present application.

[0025] Figure 3 A flow chart of a method for updating the preference of an explicit tag in the present application is shown.

[0026] Figure 4 A flow chart of another method for updating the preference of an explicit tag of the present application is shown.

[0027] Figure 5 The figure shows the principle diagram of the intelligent video recommendation system of the present application.

[0028] Figure 6 A schematic diagram of the structure of an electronic device according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0029] In order to make the purpose, features, and advantages of this application more obvious and easy to understand, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of this application.

[0030] Example 1: Please refer to Figure 1 This embodiment provides an intelligent video recommendation method, comprising the following steps: S1, annotate multiple label groups for the video; please refer to Figure 2 , the multiple label group includes explicit labels and implicit labels; The explicit tag represents the event in the video, and the implicit tag includes the viewpoint of the explicit tag, which represents the video publisher's opinion on the event in the video. and Mark to express the corresponding relationship between the two.

[0031] The explicit tag is visible to the user, while the implicit tag is invisible to the user.

[0032] Generally speaking, an event usually includes at least three different perspectives: the affirmative, the negative, and the neutral. If an event involves more parties, the number of perspectives will also increase.

[0033] That is to say, each explicit tag corresponds to several implicit tags, but it should be noted that for a single video, its explicit tag only corresponds to one implicit tag.

[0034] Explicit tags represent specific events in the video, such as an entertainment video revealing the latest developments of a public figure.

[0035] As for the latest developments of the public figure, some video publishers will express their own biased opinions, the most typical of which are support or opposition to the latest developments of the public figure, or neutrality; it is worth mentioning that if the video publisher does not express his or her own biased opinion, he or she is also regarded as neutral.

[0036] The explicit labels and implicit labels can be manually annotated or realized by analyzing the multimodal information of the video such as text and images through artificial intelligence and big data.

[0037] Text analysis: Tags can be generated by analyzing textual information such as video titles and descriptions. For example, a CRF model can be used to extract candidate tags. A semantic similarity model based on an attention mechanism can then be used to rank the candidate tags and select appropriate ones for labeling. Alternatively, a Transformer-based generative and extractive approach can be employed, where the generative model first generates the tags. If no results are found, the extractive approach is used to improve the accuracy of labeling.

[0038] Image recognition: For cover images or video frames, use pre-trained image models such as Xception for feature extraction. Select high-frequency abstract labels as image classification labels, build an image classification model, and perform fine-tuning. Then, extract intermediate layer vectors as image representations and generate relevant labels based on image features.

[0039] It is worth noting that the biased opinions on events in the video represented by implicit labels are mostly displayed in text or voice. Therefore, it is preferred to directly label the implicit labels based on text analysis or to convert the voice into text information first, and then label them based on text analysis.

[0040] S2, constructing an explicit tag preference library for the user, wherein the explicit tag preference library stores the user's preference for each explicit tag.

[0041] When constructing the explicit tag preference library, each explicit tag is assigned an initial preference degree. It should be noted that each time a new explicit tag is generated, the new explicit tag is updated to the explicit tag preference library and assigned an initial preference degree.

[0042] S3: In response to the user's historical video browsing record, the user's preference for each explicit tag stored in the explicit tag preference library is updated.

[0043] For details, please refer to Figure 3 , including the following steps: S301, setting a first time threshold and a second time threshold, wherein the first time threshold is smaller than the second time threshold; S302, extracting the user's historical video browsing records within a specified time period; the specified time period is preferably the latest 10 days or 30 days.

[0044] S303, counting the browsing time of each video and the explicit tags in the video's multiple tag groups in the historical video browsing records; S304: If the video viewing time is greater than the first time threshold, then in the explicit tag preference library, the user's preference for the explicit tags is increased; if the video viewing time is less than the first time threshold, then in the explicit tag preference library, the user's preference for the explicit tags is decreased; if the video viewing time is between the first time threshold and the second time threshold, then in the explicit tag preference library, the user's preference for the explicit tags remains unchanged; The time a user spends browsing a video largely represents whether the user likes the video. The longer the browsing time, the more the user likes the video.

[0045] In this embodiment, the user's preference for explicit tags is updated according to the above step S304 for all videos viewed by the user within the specified time period. In other words, if the user views five videos containing the same explicit tag within the specified time period, the user's preference for the explicit tag in the explicit tag preference database needs to be reduced five times.

[0046] With explicit label For example, its initial preference in the specified period is , and each time the user's preference for these explicit tags is increased by , each time the user's preference for these explicit tags is reduced by ; and They can be equal or unequal.

[0047] In the explicit tag preference library, users have Like Updated to:

[0048] in, Represents the number of times the user's preference for the explicit tag is increased. Represents the number of times the user's preference for the explicit tag is reduced.

[0049] In other embodiments, the method for updating the user's preference for each explicit tag stored in the explicit tag preference library may adopt the following scheme.

[0050] In response to the user's historical video browsing record, the method for updating the user's preference for each explicit tag stored in the explicit tag preference library is as follows, please refer to Figure 4 , including the following steps: S310 , each video is configured with a feedback mechanism, which is triggered during or after video playback to display explicit tags of the video to the user for positive or negative feedback; Only explicit tags are provided for user feedback. This is why explicit tags are visible to users, while implicit tags are invisible. This is to prevent users from intentionally or unintentionally providing feedback on implicit tags, which could lead to only pushing videos that focus on a specific perspective on the event, creating an information cocoon.

[0051] The triggering of the feedback mechanism includes but is not limited to the following methods: triggering by long touching the video playback interface during video playback, automatic triggering after the video playback ends, etc.; positive feedback / negative feedback includes like / dislike, follow / not follow, etc.

[0052] S311, setting a first time threshold and a second time threshold, wherein the first time threshold is smaller than the second time threshold; S312, extracting the user's historical video browsing records within a specified time period; S313, collecting statistics on the browsing time of each video, the explicit tags in the video's multiple tag groups, and the user's feedback on the explicit tags in the historical video browsing records; S314: If the video browsing time is greater than the first time threshold, then in the explicit tag preference library, increase the user's first preference for the explicit tags. If the video viewing time is less than the first time threshold, then in the explicit tag preference library, reduce the user's second preference for these explicit tags. If the video viewing time is between the first time threshold and the second time threshold, then the user's preference for the explicit tags in the explicit tag preference library is kept unchanged; S315: If the user gives positive feedback on the explicit tags, the user's third preference for the explicit tags is increased. On the contrary, the negative feedback of users on the explicit tags will reduce the fourth preference of users on these explicit tags. If the user does not provide feedback on the explicit tag, the user's preference for the explicit tag remains unchanged. Similarly, the user's preference for the explicit tag needs to be updated according to the above steps S314 and S315 for all videos viewed by the user within the specified time period. In S315, if the user provides feedback on the same explicit tag multiple times, the preference for the explicit tag needs to be reduced or increased multiple times accordingly.

[0053] Or explicit label For example, its initial preference in the specified period is After the method of steps S310 to S315, the user has a preference for these explicit tags in the explicit tag preference library. Like Updated to:

[0054] in, Represents the number of times the user's preference for the explicit tag is increased. Represents the number of times the user's preference for the explicit tag is reduced; Represents the number of times users have given positive feedback to the explicit tag; Represents the number of times users have given negative feedback to the explicit tag.

[0055] It should be noted that the first preference , second preference , third preference , Fourth preference The values ​​of can be different, but must meet the following conditions: Greater than first preference , and the fourth preference Greater than the second preference The reason is that the third preference , Fourth preference It is the reward / punishment for users’ self-feedback videos, and the first preference , second preference It is a reward / penalty made by analyzing the user's viewing time on the video. The reliability of the former should be greater than the latter, so the values ​​should be distinguished.

[0056] S4, combining the current popularity of the video to be recommended and the user's preference for the explicit tag of the recommended video with the product of a weight factor to obtain a priority value for the video to be recommended, wherein the weight factor is negatively correlated with the number of times the implicit tag of the video to be recommended has been viewed in historical video browsing records; Specifically, the following methods are included: S401, constructing a weight factor that is negatively correlated with the implicit tag of the video to be recommended; Extract the user's historical video browsing records within the specified time period and count the implicit tags of the videos to be recommended The corresponding explicit label The video is recorded as the first video; Count the total number of first videos , and the implicit tag containing the video to be recommended The number of first videos ; Calculate the number of first videos that contain the implicit tag of the video to be recommended Total with the first video The ratio of the constant is subtracted from the ratio to obtain the weight factor The constant is usually 1, that is:

[0057] S402, calculating the current popularity of the video to be recommended; Extract the explicit tag of the video to be recommended, count the number of times all users have viewed the video to be recommended within the specified time period, and record it as the first time. ; It should be noted that the basis for determining whether a user has viewed the video to be recommended is as follows: When the user's browsing time for the video to be recommended is greater than a first time threshold, and / or the user provides positive feedback to the video to be recommended, it is deemed that the user has browsed the video to be recommended.

[0058] The first number Divide by the set number of thresholds , realize the normalization of data, that is, get the video to be recommended Current popularity .

[0059] S403: Calculate the current popularity of the video to be recommended , and the user's preference for the explicit tags of the recommended videos With weight factor The sum of the products is used as the priority value of the video to be recommended .

[0060] .

[0061] S5, according to the priority value The priority of the recommended videos is sorted from large to small, and the videos ranked first in priority are selected for recommendation.

[0062] What this means is that before the first number of videos to be recommended are played, the next round of videos to be recommended are prioritized and recommended, thereby ensuring smooth viewing of videos by users.

[0063] Through the above steps, due to the weight factor It is negatively correlated with the number of times the implicit tag of the video to be recommended is viewed in the historical video browsing records. Therefore, if the implicit tag of the video to be recommended accounts for a higher proportion in the historical video browsing records, the priority value of the video to be recommended with the same implicit tag will be lowered in subsequent recommendations; on the contrary, if the implicit tag of the video to be recommended accounts for a lower proportion in the historical video browsing records, the priority value of the video to be recommended with the same implicit tag will be increased in subsequent recommendations.

[0064] This solution can push corresponding videos to users based on the popularity of the videos and the preference of the explicit tags, and can also combine the weight factor that is negatively correlated with the number of views of the implicit tags of the recommended videos in the historical video browsing records, so that when users browse videos with the same explicit tags in the future, they are more likely to see videos with different implicit tags. That is, for the same event, different views of the public can be obtained, avoiding the cognition of the same event being limited to a certain viewpoint. This is especially true for children, the elderly and other groups, who are less likely to fall into the dilemma of information cocoons.

[0065] Although various video publishers may offer different perspectives on the same event, a significant portion of their videos offer incomplete details and biased perspectives. This makes it difficult for viewers to fully understand the event, and they may even be misled by the views of some video publishers.

[0066] To avoid this situation, this solution also includes the following methods: If the number of times a user browses a video tagged with a certain explicit tag for a time period greater than a first time threshold reaches a set number threshold, and / or the number of times a user provides positive feedback on a certain explicit tag reaches a set number threshold, the user will be marked as a follower of the explicit tag; In S1, the notification video for the event published by the authenticated video publisher is labeled with corresponding explicit tags and notification tags; Certified video publishers are usually government agencies at all levels, such as public security departments, courts and other administrative agencies.

[0067] When performing priority sorting in S5, if one of the recommended videos is marked with a notification tag, when recommending to the following user with the corresponding explicit tag, the priority of the recommended video marked with the notification tag is set to the highest and ranked first.

[0068] Example 2: This embodiment 2 provides an intelligent video recommendation system, please refer to Figure 5 , including the following functional modules: A labeling module is used to label a video with a multiple label group; the multiple label group includes explicit labels and implicit labels; A preference building module builds an explicit tag preference library for the user, wherein the explicit tag preference library stores the user's preference for each explicit tag; A preference updating module, in response to the user's historical video browsing record, updates the user's preference for each explicit tag stored in the explicit tag preference library; A priority calculation module calculates the priority value of the video to be recommended by combining the current popularity of the video to be recommended and the product of the user's preference for the explicit tag of the recommended video and a weight factor. The weight factor is negatively correlated with the number of times the implicit tag of the video to be recommended has been viewed in historical video browsing records. The sorting recommendation module sorts the priority of the recommended videos in descending order of priority value, and selects the first number of recommended videos to be recommended for recommendation.

[0069] The specific implementation methods / principles of the functional modules in the second embodiment are the same as those in the first embodiment. For details, please refer to steps S1 to S5 of the first embodiment, which will not be repeated here.

[0070] Example 3 This third embodiment also provides an electronic device and a readable storage medium.

[0071] Figure 6 A schematic block diagram of an example electronic device that can be used to implement an embodiment of the present application is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described and / or required herein.

[0072] like Figure 6 As shown, the device includes a computing unit that can perform various appropriate actions and processes based on a computer program stored in a read-only memory (ROM) or loaded from a storage unit into a random access memory (RAM). The RAM can also store various programs and data required for device operation. The computing unit, ROM, and RAM are interconnected via a bus. An input / output (I / O) interface is also connected to the bus.

[0073] Many components in a device are connected to the I / O interface, including: input units, such as a keyboard and mouse; output units, such as various types of displays and speakers; storage units, such as magnetic disks and optical disks; and communication units, such as network cards, modems, and wireless communication transceivers. The communication unit allows the device to exchange information / data with other devices via computer networks such as the Internet and / or various telecommunication networks.

[0074] The computing unit can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of computing units include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit performs the various methods and processes described above, such as the intelligent video recommendation method described in Example 1. For example, in some embodiments, the intelligent video recommendation method can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device via ROM and / or a communication unit. When the computer program is loaded into RAM and executed by the computing unit, one or more steps of the intelligent video recommendation method described above can be performed. Alternatively, in other embodiments, the computing unit can be configured to perform the intelligent video recommendation method through any other suitable means (e.g., via firmware).

[0075] Various embodiments of the systems and techniques described above can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0076] The program code for implementing the methods of the present application can be written in any combination of one or more programming languages. Such program code can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the program code is executed by the processor or controller, the functions / operations specified in the flow charts and / or block diagrams are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0077] In the context of this application, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or apparatus. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fibers, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0078] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0079] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0080] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises through computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.

[0081] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this application can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of this application can be achieved. This is not limited herein.

[0082] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. Throughout the description of this application, "plurality" means two or more, unless otherwise specifically defined.

[0083] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. An intelligent video recommendation method, characterized in that: The following steps are involved: S1, labeling a video with a multiple label group; the multiple label group includes explicit labels and implicit labels; S2, building an explicit tag preference library for the user, wherein the explicit tag preference library stores the user's preference for each explicit tag; S3, in response to the user's historical video browsing record, updating the user's preference for each explicit tag stored in the explicit tag preference library; S4, combining the current popularity of the video to be recommended and the user's preference for the explicit tag of the recommended video with the product of a weight factor to obtain a priority value for the video to be recommended, wherein the weight factor is negatively correlated with the number of times the implicit tag of the video to be recommended has been viewed in historical video browsing records; S5: sort the priority of the videos to be recommended in descending order of priority value, and select the videos to be recommended that are ranked first in priority for recommendation.

2. The intelligent video recommendation method according to claim 1, characterized in that In said S2, when constructing the explicit tag preference library, an initial preference degree is assigned to each explicit tag; and each time a new explicit tag is generated, the new explicit tag is updated to the explicit tag preference library and assigned an initial preference degree at the same time.

3. The intelligent video recommendation method according to claim 1, characterized in that In S3, in response to the user's historical video browsing record, the method for updating the user's preference for each explicit tag stored in the explicit tag preference library is as follows: S301, setting a first time threshold and a second time threshold, wherein the first time threshold is smaller than the second time threshold; S302, extracting the user's historical video browsing records within a specified time period; S303, counting the browsing time of each video and the explicit tags in the video's multiple tag groups in the historical video browsing records; S304, if the viewing time of the video is greater than the first time threshold, then in the explicit tag preference library, the user's preference for the explicit tags is increased; if the viewing time of the video is less than the first time threshold, then in the explicit tag preference library, the user's preference for the explicit tags is reduced; if the viewing time of the video is between the first time threshold and the second time threshold, then in the explicit tag preference library, the user's preference for the explicit tags is kept unchanged.

4. The intelligent video recommendation method according to claim 1, characterized in that In S3, in response to the user's historical video browsing record, the method for updating the user's preference for each explicit tag stored in the explicit tag preference library is as follows: S310 , each video is configured with a feedback mechanism, which is triggered during or after video playback to display explicit tags of the video to the user for positive or negative feedback; S311, setting a first time threshold and a second time threshold, wherein the first time threshold is smaller than the second time threshold; S312, extracting the user's historical video browsing records within a specified time period; S313, collecting statistics on the browsing time of each video, the explicit tags in the video's multiple tag groups, and the user's feedback on the explicit tags in the historical video browsing records; S314: If the video viewing time is greater than the first time threshold, then in the explicit tag preference library, the user's first preference for the explicit tags is increased; if the video viewing time is less than the first time threshold, then in the explicit tag preference library, the user's second preference for the explicit tags is decreased; if the video viewing time is between the first time threshold and the second time threshold, then in the explicit tag preference library, the user's preference for the explicit tags remains unchanged; S315, if the user provides positive feedback on the explicit tags, the user's third preference for the explicit tags will be increased; conversely, if the user provides negative feedback on the explicit tags, the user's fourth preference for the explicit tags will be reduced; if the user does not provide any feedback on the explicit tags, the user's preference for the explicit tags will remain unchanged.

5. The intelligent video recommendation method according to claim 1, characterized in that: In S4, the method for obtaining the priority value of the video to be recommended by combining the current popularity of the video to be recommended and the product of the user's preference for the explicit tag of the recommended video and the weight factor is as follows: S401, constructing a weight factor that is negatively correlated with the implicit tag of the video to be recommended; S402, calculating the current popularity of the video to be recommended; S403: Calculate the current popularity of the video to be recommended and the sum of the product of the user's preference for the explicit tag of the recommended video and the weight factor as the priority value of the video to be recommended.

6. The intelligent video recommendation method according to claim 5, characterized in that: In S401, the method for constructing a weight factor that is negatively correlated with the implicit tag of the video to be recommended is as follows: Extract the user's historical video browsing records within a specified time period, count the videos containing the explicit tags corresponding to the implicit tags of the video to be recommended, and record them as the first videos; Counting the total number of first videos and the number of first videos containing the implicit tag of the video to be recommended; The ratio of the number of first videos containing the implicit tag of the video to be recommended to the total number of first videos is calculated, and the weight factor is obtained by subtracting the ratio from a constant.

7. The intelligent video recommendation method according to claim 5, characterized in that: In S402, the method for calculating the current popularity of the video to be recommended is as follows: Extract the explicit tag of the video to be recommended, count the number of times all users have viewed the video to be recommended within the specified time period, and record it as the first time; The current popularity of the video to be recommended is obtained by dividing the first number of times by the set number threshold.

8. The intelligent video recommendation method according to claim 7, characterized in that: The basis for determining whether the user has viewed the video to be recommended is as follows: when the user's browsing time for the video to be recommended is greater than a first time threshold, and / or the user provides positive feedback on the video to be recommended, it is deemed that the user has viewed the video to be recommended.

9. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 8.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 8.

Images