CN112528150A - Live broadcast recommendation method and device, electronic equipment and storage medium - Google Patents
Live broadcast recommendation method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN112528150A CN112528150A CN202011496650.1A CN202011496650A CN112528150A CN 112528150 A CN112528150 A CN 112528150A CN 202011496650 A CN202011496650 A CN 202011496650A CN 112528150 A CN112528150 A CN 112528150A
- Authority
- CN
- China
- Prior art keywords
- user
- current
- live
- live broadcast
- content
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 67
- 230000008451 emotion Effects 0.000 claims abstract description 61
- 230000006399 behavior Effects 0.000 claims description 43
- 238000004458 analytical method Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 12
- 238000001514 detection method Methods 0.000 claims description 5
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000036651 mood Effects 0.000 claims description 2
- 238000004422 calculation algorithm Methods 0.000 description 15
- 238000007405 data analysis Methods 0.000 description 10
- 238000004364 calculation method Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 7
- 230000006870 function Effects 0.000 description 5
- 238000013135 deep learning Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 238000013480 data collection Methods 0.000 description 3
- 238000013500 data storage Methods 0.000 description 3
- 230000008921 facial expression Effects 0.000 description 3
- 230000019771 cognition Effects 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 210000000988 bone and bone Anatomy 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000008909 emotion recognition Effects 0.000 description 1
- 230000002996 emotional effect Effects 0.000 description 1
- 230000001815 facial effect Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000003205 muscle Anatomy 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/906—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
- G06F21/36—User authentication by graphic or iconic representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0269—Targeted advertisements based on user profile or attribute
- G06Q30/0271—Personalized advertisement
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0277—Online advertisement
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/174—Facial expression recognition
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- General Engineering & Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Computer Security & Cryptography (AREA)
- Entrepreneurship & Innovation (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
The application discloses a live broadcast recommendation method, a live broadcast recommendation device, electronic equipment and a storage medium, and relates to the technical field of computer multimedia, wherein the method comprises the following steps: when detecting that a live application program is started, acquiring a user image of a current user; acquiring attribute information and current emotion of the current user according to the user image; determining at least one preference tag of the current user according to the attribute information and the current emotion; acquiring a current live broadcast list, wherein each live broadcast content in the current live broadcast list corresponds to at least one classification label; and selecting target live broadcast content from the current live broadcast list for recommendation according to the relevance of the at least one preference label and the at least one classification label, wherein the method can effectively realize live broadcast recommendation.
Description
Technical Field
The present application relates to the field of computer multimedia technologies, and in particular, to a live broadcast recommendation method and apparatus, an electronic device, and a storage medium.
Background
Recommendation System (Recommendation System) is an important tool that can help users to discover content and overcome information overload. The user interest is modeled by analyzing the user behavior, so that the user interest is predicted and recommended to the user. With the development of the mobile internet entertainment industry, related streaming media such as live broadcast and short video are more and more deep into the life of people, and the quality requirements of people on live broadcast and short video are higher and higher. At present, most live broadcast recommendation systems recommend programs for users only based on viewing records, which results in single program recommendation and cannot achieve the effect of personalized recommendation.
Disclosure of Invention
The application provides a live broadcast recommendation method and device, electronic equipment and a storage medium.
In a first aspect, an embodiment of the present application provides a live broadcast recommendation method, including: when detecting that a live application program is started, acquiring a user image of a current user; acquiring attribute information and current emotion of the current user according to the user image; determining at least one preference tag of the current user according to the attribute information and the current emotion; acquiring a current live broadcast list, wherein each live broadcast content in the current live broadcast list corresponds to at least one classification label; and selecting target live broadcast content from the current live broadcast list for recommendation according to the relevance of the at least one preference label and the at least one classification label.
In a second aspect, an embodiment of the present application provides a live broadcast recommendation apparatus, where the apparatus includes: the detection module is used for acquiring a user image of a current user when detecting that the live application program is started; the first acquisition module is used for acquiring the attribute information and the current emotion of the current user according to the user image; the determining module is used for determining at least one preference tag of the current user according to the attribute information and the current emotion; the second acquisition module is used for acquiring a current live broadcast list, wherein each live broadcast content in the current live broadcast list corresponds to at least one classification label; and the recommending module is used for selecting the target live broadcast content from the current live broadcast list for recommending according to the relevance of the at least one preference label and the at least one classification label.
In a third aspect, an embodiment of the present application provides an electronic device, including: one or more processors; a memory; one or more cameras, wherein a user of the camera acquires an image and transmits the image to a processor for processing; one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, the one or more applications being configured to perform the live recommendation method provided in the first aspect above.
In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, where a program code is stored in the computer-readable storage medium, and the program code may be called by a processor to execute the live broadcast recommendation method provided in the first aspect.
According to the live broadcast recommendation method and device, the electronic equipment and the storage medium, when the live broadcast application program is detected to be started, the user image of the current user is obtained, the attribute information and the current emotion of the current user are obtained according to the user image, and at least one preference label of the current user is further determined. And meanwhile, acquiring at least one classification label corresponding to each live content in the current live list, and selecting the target live content from the current live list for recommendation according to the obtained correlation degree of at least one preference label and at least one classification label. Therefore, after the relevance of the preference data of the live broadcast content and the live broadcast content of the user is obtained, the electronic equipment can recommend live broadcast programs to the user according to the relevance sequence, and therefore the accuracy of personalized recommendation of different users is improved.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 shows a flow chart of a live recommendation method according to an embodiment of the application.
Fig. 2 shows a flowchart of a live recommendation method according to another embodiment of the present application.
Fig. 3 shows a flowchart of step S260 in a live recommendation method according to another embodiment of the present application.
Fig. 4 shows a flowchart of step S280 in a live recommendation method according to still another embodiment of the present application.
Fig. 5 shows a flowchart of a live recommendation method according to a further embodiment of the present application.
Fig. 6 shows a flow diagram of a live recommendation method according to an embodiment of the application.
Fig. 7 shows an architecture diagram of a live recommendation system according to an embodiment of the present application.
Fig. 8 shows a block diagram of a live recommender in accordance with an embodiment of the present application.
Fig. 9 is a block diagram of an electronic device for executing a live broadcast recommendation method according to an embodiment of the present application.
Fig. 10 is a storage unit, according to an embodiment of the present application, configured to store or carry program code for implementing a live broadcast recommendation method according to an embodiment of the present application.
Detailed Description
In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application.
The network live broadcast gradually has new expansion due to the development of internet technology at any time. Users can watch various programs provided by sports events, entertainment and entertainment, news and anchor broadcasts on line through network live broadcast. Generally, when a user watches live broadcasting by using a live broadcasting platform, the live broadcasting platform can recommend live broadcasting programs which the user may like to the user according to the preference of the user. At present, most live broadcast recommendation methods acquire the relevant information of user interest only when live broadcast is in an end state, so that the purpose of real-time recommendation of users cannot be achieved. In addition, when a program is recommended, data processing is generally performed only on an original live program list to obtain a target recommended live program list, which results in a single type of recommended program in the target recommended live program list.
Therefore, in order to overcome the above-mentioned defects, an embodiment of the present application provides a live broadcast recommendation method, please refer to fig. 1, and fig. 1 shows a live broadcast recommendation method provided by the embodiment of the present application. The method is applied to electronic equipment, which can be a camera device for acquiring images and various electronic equipment supporting data storage and computing capacity. Specifically, the method comprises the following steps: s110 to S150.
S110: and when the live broadcast application program is detected to be started, acquiring a user image of the current user.
In order to improve the real-time performance of the electronic equipment for live broadcast recommendation of the user, the electronic equipment can recommend live broadcast programs in real time in the process of watching the live broadcast programs by the user. Therefore, when the user starts to use the live broadcast application, the electronic equipment can start live broadcast recommendation operation so as to ensure the real-time performance of live broadcast recommendation. In some embodiments, when the electronic device detects that a live application used is started, a user image of a current user may be acquired, where the user image refers to a picture or video data acquired by the electronic device through a camera at the current time, and the picture or video data may be used to identify user identity information and facial expressions.
As an implementation manner, the electronic device may start the face detection program when detecting that the user opens the live broadcast application, or start the face detection program when detecting that the user clicks the operation behavior of the live broadcast video file after the live broadcast application is opened. For example, currently, on a smart phone, a user opens live application software to watch live programs, at this time, the smart phone detects that the user clicks a live program file, and then a camera can be used to capture face pictures and clothing pictures of the user.
As another embodiment, the face detection program may be set as a resident application program, and defaults to a background process that is running all the time. Specifically, when the electronic device is powered on, the face detection program can be started and enters a standby state in the background, and when the user clicks the live application software, the face recognition program enters a starting state from the standby state.
S120: and acquiring the attribute information and the current emotion of the current user according to the user image.
In the embodiment of the application, after the electronic device collects the user image of the user, the electronic device can determine the identity information of the user by analyzing the user image, so that the attribute information of the user can be obtained by combining the user identity information, and further, personalized recommendation service can be provided for the user by combining the attribute information of the user and the current emotion. The attribute information is at least basic information that can be used to confirm the identity of the user, and may include the gender, age, and geographic area of residence of the user, which is not limited herein. The current emotion refers to an emotional state, including anger, fear, surprise, sadness and happiness, exhibited by facial expressions or body movements of the user when the user watches a live program, and is not limited herein.
In some embodiments, the electronic device may extract the current emotion of the acquired user image based on a Face Recognition (FR) algorithm and/or a Facial Expression Recognition (FER) algorithm. The FR algorithm and the FER algorithm may be deep learning algorithms. As an implementation manner, after acquiring a user image of a user, live broadcast application software on an electronic device may extract, through a Multi-task Convolutional Neural Network (MTCNN) and a Deep Belief Network (DBN), a limb feature and/or a facial feature of the user from the acquired user image, respectively, so as to obtain a current emotion of the user through feature analysis. As another embodiment, the electronic device may obtain social data of the user through the data collection agent, and combine the collected user image with the social data to analyze the current emotion used. Considering that the emotion of the user has volatility, the recent emotion condition of the user can be known through social data of the user in a recent period of time, and then data analysis is carried out together with the collected user images to obtain the current emotion, so that the accuracy analysis of the current emotion of the user is improved. Specifically, the data collection agent captures the speech, the internet surfing trace and the like of the user on the network in a web crawler mode. Such as capturing logs of records left by users on social networks, online shopping platforms, web search platforms, web service platforms, and the like. For another example, information related to the user published through various channels, such as social media content of friends in the user's social circle, information related to the user in a log, and the like, is captured.
In other embodiments, after acquiring the user image, the electronic device may determine the current user identity through an image recognition technology, and further acquire attribute information of the corresponding user according to the user identity. Therefore, recommendation operation can be performed according to different types of user identities so as to improve the accuracy of personalized live broadcast recommendation. As an embodiment, after the electronic device determines the identity of the user through the user image, the electronic device may obtain corresponding attribute information from a database in which the user is stored. For example, when the smartphone determines that the user currently using the live application software is a registered user based on the face recognition program, the smartphone may acquire the attribute information from the database in which the user attribute information is stored through the database interface. As another embodiment, when the electronic device determines that the user is a new user through the user image, user feature extraction may be performed according to the collected user image of the user, and further, the extracted user feature may be analyzed by using a big data analysis technique to obtain attribute information of the user. In addition, the social data of the user can be acquired by the data collection agent according to the identity information of the user, so that data analysis is performed, and the attribute information of the user is obtained.
S130: and determining at least one preference tag of the current user according to the attribute information and the current emotion.
In the embodiment of the application, in order to improve the accuracy of recommending different users by the electronic equipment, after the attribute information and the current emotion of the user are acquired, the electronic equipment can analyze the preference tag of the user according to the attribute information and the current emotion, and further perform targeted personalized recommendation by combining live broadcast content. The preference tag is used for describing the preference of the user for watching the live broadcast content, can effectively express the requirement of the user for watching the live broadcast content and the live broadcast watching interest of the user, and plays a guiding role in personalized recommendation of the user for watching the live broadcast.
As an implementation manner, the electronic device may perform data preprocessing on the acquired attribute information, and further, may perform feature extraction on the preprocessed attribute information through feature engineering, specifically, the algorithm used by the feature engineering may include a characterization learning algorithm, for example, the electronic device may perform feature extraction on the attribute information by using a graph probability model, and perform classification to obtain a preference pre-label corresponding to the attribute information. Furthermore, the electronic equipment can perform matching calculation on the preference pre-label and the emotion of the user at the current moment, and generate the preference label matched with the emotion of the user at the current moment from the preference pre-label, so that the preference label can have the characteristic information of the emotion of the user at the current moment, and the real-time performance of personalized recommendation of the live program is improved.
S140: and acquiring a current live broadcast list, wherein each live broadcast content in the current live broadcast list corresponds to at least one classification label.
In the embodiment of the application, after the electronic equipment acquires the preference tag of the user, the preference tag and the classification tag of the live broadcast content can be calculated, so that the live broadcast program is recommended, the classification tag abstracts, summarizes and analyzes the live broadcast content to obtain the most valuable and representative live broadcast content information, and the cognition and understanding of the user on the live broadcast content can be further simplified and accurate. In some embodiments, a specific live content list and one or more classification tags corresponding to each program are stored in the live list, where the classification tags are tags of live content information in multiple dimensions, and may include, but are not limited to, content type, genre theme, field, applicable group, validity period, video subject, shooting location, and anchor characteristics.
As an implementation manner, the electronic device may obtain a live broadcast list from a WebRTC (Web Real-Time Communication, WebRTC) server, and it should be understood that when new live broadcast content is added to the WebRTC server or when live broadcast content currently being live broadcast is ended, the WebRTC server may notify the electronic device to dynamically update the current live broadcast list. Furthermore, when a certain live content finishes playing, the end user may take the time to mask the live content in order to prevent repeated recommendations. The classification tags in the live broadcast list can be obtained by performing data analysis on live broadcast content through a deep learning algorithm by the WebRTC server, for example, voice recognition and image recognition can be performed on live broadcast video content, so that feature extraction is performed on the live broadcast content, and the classification tags are obtained by analyzing further according to the extracted feature information.
S150: and selecting target live broadcast content from the current live broadcast list for recommendation according to the relevance of the at least one preference label and the at least one classification label.
In the embodiment of the application, after the electronic equipment acquires the preference label of the user and the classification label of the live content, the relevance of the preference label and the classification label can be calculated, so that the live content can be recommended according to the calculated relevance. Specifically, the electronic device can sort the relevance calculated by the preference tag and the classification tag according to the sequence of the relevance from large to small, and select the first live content in the sorting as the target live content each time to recommend the live content. In addition, platform behavior characteristics of the user can be added in the calculation of the relevance, wherein the platform behavior characteristics refer to information of the live broadcast content on a live broadcast platform, such as history of clicking, praise, collection, forwarding and the like, the electronic equipment can take the platform behavior characteristics as one of the weight coefficients to participate in the calculation of the relevance, and the platform behavior characteristics really reflect the attention condition of the user to the live broadcast content, so that the accuracy of a calculation result of the relevance can be improved by adding the platform behavior characteristics in the calculation of the relevance.
As an embodiment, the electronic device may perform matching calculation on the preference tag of the user and the category tag of the live content, so as to obtain target live content that can be recommended. Specifically, for a certain user, the electronic device may match a category tag of one live content in the live content list with a preference tag of the user to obtain a tag number related to the preference tag of the user in the category tag of the live content. Further, the tag numbers are sorted from large to small, and then sequentially recommended according to the sorting order, specifically, the electronic device may acquire the sorted Top-N live content as the target live content to sequentially recommend.
According to the live broadcast recommendation method provided by the embodiment of the application, when the live broadcast application program is detected to be started, the user image of the current user is obtained, the attribute information and the current emotion of the current user are obtained according to the user image, and then at least one preference label of the current user is determined. And meanwhile, acquiring at least one classification label corresponding to each live content in the current live list, and selecting the target live content from the current live list for recommendation according to the obtained correlation degree of at least one preference label and at least one classification label. Therefore, after the relevance of the preference data of the live broadcast content of the user and the live broadcast content is obtained, the electronic equipment can recommend live broadcast programs to the user according to the relevance sequence, and therefore the accuracy of personalized recommendation to the user is improved.
Referring to fig. 2, fig. 2 illustrates another live broadcast recommendation method provided in an embodiment of the present application. The method is applied to electronic equipment, which can be a camera device for acquiring images and various electronic equipment supporting data storage and computing capacity. Specifically, the method comprises the following steps: s210 to S280.
S210: and when the live broadcast application program is detected to be started, acquiring a user image of the current user.
In the embodiment of the present application, the content in the foregoing embodiment may be referred to for a specific description of step S210, and is not repeated herein.
S220: and recognizing the face information in the user image, and judging whether the current user is a new user or not based on the recognition result.
In the embodiment of the application, after the electronic equipment acquires the user image of the current user, the user image can be identified so as to confirm the identity of the user, so that personalized recommendation can be more accurately performed and the recommendation efficiency is improved. As an implementation manner, the electronic device may perform face recognition on the acquired user image through a deep learning algorithm, so as to determine whether the user is a new user. For example, when a user starts a live application by using a smart phone, the smart phone may acquire a face image of the user through a camera, and then recognize the face image by using an lbph (local Binary Patterns databases) algorithm based on Opencv, so as to obtain face information of the user and compare the face information with a database in which the face information of the user is stored, thereby determining whether the user is a new user.
S230: and when the current user is a new user, analyzing and processing the physical state of the current user in the user image to obtain the attribute information and the current emotion of the current user.
In the embodiment of the application, in order to accurately recommend live programs to a new user, after the electronic device judges that the user is the new user, the electronic device can acquire attribute information of the new user and watch the current emotion of live programs in a targeted manner, so that effective support is provided for live program recommendation. The physical state may refer to state data expressed by the user in aspects of face, body state, bone, posture, muscle, and the like. For example, a human face image is the state data expressed by the human body in the aspect of face appearance. The specific physical appearance data is not limited in the embodiment of the application, and the physical appearance data only needs to restore the physical appearance of the user.
As an embodiment, after obtaining the physical state of the user, the electronic device may obtain attribute information corresponding to the physical state and a current mood based on big data analysis. The electronic device may further store the analyzed attribute information and the current emotion, and optionally, the electronic device may upload the attribute information of the new user to the user data server for storage.
S240: and when the current user is not a new user, acquiring the attribute information of the current user, which is stored in advance.
In the embodiment of the application, when the electronic device determines that the current user is not a new user, the specific identity information of the user can be determined through a face recognition technology, and further, the attribute information of the current user, which is stored in advance, is obtained according to the identity information. In one embodiment, when determining that a user watching a live broadcast is not a new user, the electronic device may directly retrieve attribute information that has been previously stored by the user from a server for storing the attribute information of the user. For example, when the electronic device determines that the user currently using the live application software is an old user, the electronic device may directly obtain attribute information corresponding to the user from the user data server. The preference label of the user can be directly determined by directly acquiring the attribute information of the old user, so that the live broadcast content recommendation step is simplified, and the recommendation efficiency is improved.
S250: and performing emotion analysis on the face information in the user image to obtain the current emotion of the current user.
In the embodiment of the application, after the electronic device acquires the user image, emotion analysis can be performed on face information in the user image through a face recognition algorithm, so that the current emotion obtained through analysis is combined with attribute information to generate the preference label of the user. As an embodiment, after acquiring a face image of a user, the electronic device may perform emotion recognition on the face image through at least one of the following algorithms: the method comprises a supervised learning algorithm, an unsupervised learning algorithm and a deep learning algorithm. For the detailed description, reference may be made to the contents in the foregoing embodiments, which are not described herein again.
S260: and determining at least one preference tag of the current user according to the attribute information and the current emotion.
Because the interest and emotion of the user have great influence on the preference of watching different live broadcast contents in a short term and a long term, the attribute information and the current emotion of the user are fused and jointly used for live broadcast recommendation, and the attribute characteristics and the emotion characteristics of the user which are personalized as a core can be effectively utilized, so that the personalization and the real-time performance of the live broadcast recommendation are improved. In some embodiments, the electronic device determines at least one preference tag of the current user according to the attribute information and the current emotion, and in particular, referring to fig. 3, step S260 may include:
s261: and acquiring at least one candidate label corresponding to the attribute information.
As an implementation manner, after acquiring attribute information of a user, the electronic device may obtain a corresponding candidate tag from the attribute information by a data analysis method, and further obtain a preference tag, so as to formulate a personalized live content recommendation policy according to the preference tag. The candidate tag may refer to a type of live program content that is of great probability of interest to the user. Specifically, the electronic device may perform big data analysis on the attribute information based on the acquired attribute information to obtain a candidate tag of the user. For example, after acquiring attribute information of the user, such as age, gender, makeup style, dressing style, and the like, the electronic device may obtain candidate tags of the preference of the user for live content through big data analysis, for example, a female youth user, and the candidate tags obtained through analysis may be shopping, makeup, and travel.
As another embodiment, the electronic device may further analyze the candidate tag based on the social data and the attribute information by obtaining the social data of the user from the data agent. Therefore, the candidate label is more in line with the cognition and interest of the user in the actual life, and the recommendation accuracy is improved.
S262: selecting at least one preference tag from the at least one candidate tag that matches the current emotion.
As an implementation manner, after acquiring at least one candidate tag corresponding to the attribute information, the electronic device may select at least one preference tag matching the current emotion from the at least one candidate tag in order to obtain a preference tag according with the current emotion state of the user, and thus, the obtained user preference tag may have real-time performance, so that the recommendation has real-time performance. Specifically, the electronic device may match n (n >1 or n ═ 1) candidate tags with the current emotion, and select l (l < n or l ═ n) preference tags that match the current emotion from the n candidate tags.
S270: and acquiring a current live broadcast list, wherein each live broadcast content in the current live broadcast list corresponds to at least one classification label.
S280: and selecting target live broadcast content from the current live broadcast list for recommendation according to the relevance of the at least one preference label and the at least one classification label.
In some embodiments, the electronic device selects, according to the relevance between the at least one preference tag and the at least one category tag, a target live content from the current live list for recommendation, and specifically, referring to fig. 4, step S280 may include:
s281: and aiming at each live content in the current live list, matching at least one classification label corresponding to the live content with the at least one preference label.
As an implementation manner, the electronic device may perform matching calculation on preference tags of the user and classification tags of all live content in the live list one by one, so that preference tag data matched with the preference tags of the user may be obtained, and recommendation may be performed according to a tag number sorting result. For example, for a certain user, the electronic device may select n (n >1 or n ═ 1) preference tags of the user, and similarly select m (m >1 or m ═ 1) classification tags of a certain live content in the live list, and further match the n preference tags with the m classification tags.
S282: and acquiring the number of the preference labels successfully matched with each live content.
As an implementation manner, after performing matching calculation on at least one classification tag corresponding to live content and at least one preference tag, the electronic device obtains the number of the preference tags successfully matched with each live content. For example, when h of the user's n preference tags match some of the m category tags of the live content, h is taken as the number of preference tags for which the matching is successful. And similarly, matching preference labels and classification labels with all live broadcast contents in the live broadcast list and the user to obtain the number of the preference labels successfully matched with each live broadcast content.
S283: and sequencing the live broadcast contents in the current live broadcast list according to the sequence of the preference labels from large to small to obtain a sequenced live broadcast recommendation list.
As an implementation manner, after obtaining the number of preference tags, the electronic device may sort the number of preference tags successfully matched with all live broadcast contents in a descending order, and finally obtain a sorted live broadcast recommendation list.
S284: and selecting target live broadcast content for recommendation based on the live broadcast recommendation list.
As an implementation manner, the electronic device may select Top-N target live broadcast contents from a live broadcast recommendation list for recommendation, or sequentially recommend the target live broadcast contents according to the order of the number of preference tags from large to small.
According to the live broadcast recommendation method provided by the embodiment of the application, the face information in the user image is identified, whether the current user is a new user is judged based on the identification result, when the current user is the new user, the physical and appearance state of the current user in the user image is analyzed and processed, the attribute information and the current emotion of the current user are obtained, when the current user is not the new user, the attribute information of the current user stored in advance is obtained, and the electronic equipment can perform different recommendation operations according to different user types, so that the accuracy of personalized recommendation is improved, and the recommendation efficiency is improved.
Referring to fig. 5, fig. 5 shows a live broadcast recommendation method according to an embodiment of the present application. The method is applied to electronic equipment, which can be a camera device for acquiring images and various electronic equipment supporting data storage and computing capacity. Specifically, the method comprises the following steps: s310 to S380.
S310: and when the live broadcast application program is detected to be started, acquiring a user image of the current user.
S320: and acquiring the attribute information and the current emotion of the current user according to the user image.
S330: and determining at least one preference tag of the current user according to the attribute information and the current emotion.
S340: and acquiring a current live broadcast list, wherein each live broadcast content in the current live broadcast list corresponds to at least one classification label.
S350: and selecting target live broadcast content from the current live broadcast list for recommendation according to the relevance of the at least one preference label and the at least one classification label.
In the embodiment of the present application, the content in the foregoing embodiment can be referred to for the specific description of step S310 to step S350, and is not repeated herein.
S360: and acquiring the behavior information of the current user when watching the live broadcast content.
When the electronic equipment carries out personalized recommendation on live broadcast content, differentiation processing needs to be carried out in combination with actual use conditions of users, namely the live broadcast content is adjusted according to behavior characteristics of the users when watching the live broadcast, and therefore instantaneity of live broadcast watching recommendation of the users is improved. Therefore, in the embodiment of the application, after the electronic equipment carries out live broadcast recommendation operation on the user, the behavior information of the user when the user watches live broadcast content currently can be detected in real time, and when the behavior of the user is different from the normal watching behavior, the user can be recommended again. The behavior information may include a viewing duration of the user viewing the recommended live content and a number of times of switching different live contents, which is not limited herein.
As an embodiment, for a recommended live content, the electronic device may detect, in real time, a time when a user views the recommended live content and a number of times of switching different live contents. The time for the user to watch the live content is the time length from when the user actually starts watching the live content to when the user switches to different live content. It should be understood that, when the electronic device detects that the user clicks the live content file but does not actually watch the live content, the watching duration is not counted, specifically, for a live program, in the process of watching the live program, when the electronic device cannot detect the user's physical form, the duration recording may be suspended, and when the electronic device detects the user again, the duration recording is continued.
S370: and when the behavior information meets a first preset condition, updating the current live broadcast list.
In the embodiment of the application, after the electronic device acquires the behavior information of the current user watching the live broadcast content, whether the behavior information meets a first preset condition or not can be detected, and when the behavior information meets the first preset condition, the current live broadcast list is updated, so that the real-time performance of live broadcast recommendation is improved. The first preset condition is that the electronic equipment watches the time length of recommended live content and switches the times of different live content under abnormal conditions.
As an implementation manner, when the electronic device detects that a duration of viewing the recommended live content by the user meets a preset first time long condition and a number of times of switching while viewing the live content reaches a preset switching threshold, a current live list may be updated, where the first time long condition is a time threshold set in advance by the electronic device. Specifically, the electronic device may retrieve the current live list and delete live content viewed by the user from the live list.
S380: and selecting target live broadcast content from the updated current live broadcast list for recommendation according to the relevance of the at least one preference label and the at least one classification label.
In the embodiment of the present application, the content in the foregoing embodiment may be referred to for a specific description of step S380, and is not repeated herein.
Exemplarily, referring to fig. 6, fig. 6 shows an overall flowchart of a live broadcast recommendation method provided by an embodiment of the present application. Specifically, when detecting that a user starts the live application software, the electronic device may start the camera to shoot a face image, and then perform face detection according to the shot face image to determine whether the user is a new user or an old user. And when the user is detected to be a new user, extracting and classifying the characteristics of the new user. The extracted features include but are not limited to gender, age, nationality, dressing style, emotion and the like, and further, user data is generated according to the extracted features, and preference information of the user for live content is obtained based on big data analysis. When it is detected that the user is an old user, user information and preference information of the user for live content may be acquired from a user data server.
For example, if the new user is detected to be a baby, by analyzing the baby's preferences may be enlightening animations, children's songs, toys trying to play live-like programs; if the new user is a teenager male, analyzing the preference of the teenager male to possibly play games, sports and online education live broadcast programs; if the user is a young woman, the user may be shopping, make-up, or a tour live program by analyzing the preference of the young woman; if the user is an old man, the preference of the old man is analyzed to be fishing, health preserving and drama live broadcasting programs.
After the electronic device performs feature analysis on the new user to obtain the preference information, the identity information and the preference information of the new user can be uploaded to a user data server for storage. Further, the electronic device may acquire a live list of a current live program from the WebRTC server, where the live list includes a category tag corresponding to the live program. The electronic equipment can match the acquired live list with the preference information of the user, and sort the matching marks obtained after matching according to the sequence from the top to the bottom. Further, according to the sequencing sequence, the live program sequenced to be the first is taken as the recommended live content, and meanwhile, the live content is played in a pull stream mode from the WebRTC server.
When the electronic equipment finds that the user frequently switches the live broadcast content, ranking can be performed again with the current user preference data according to the obtained latest live broadcast content list, the recommended live broadcast content is removed, the matching mark number is sorted from most to least, when the user carelessly watches the live broadcast, user behavior analysis is performed on the user, namely, the operation behavior of the user is analyzed, when the user slides up and down, left and right to switch the live broadcast content, the live broadcast content with the first ranking is recommended again, and when the user slides next time, the recommended content is played by pulling from the WebRTC server.
Referring to fig. 7, fig. 7 is a diagram illustrating an overall architecture of a live recommendation system 300 provided by the present application. Specifically, the live recommendation system 300 includes: the live broadcast application module 310 is used for a user to watch live broadcast application programs, and can be deployed on electronic equipment with multimedia capability, such as a mobile phone, a computer, a tablet computer, a PC-side computer and the like; the WebRTC service module 320 is configured to forward live content, classify each live slave content, the anchor gender, the age, the style, the industry to which the content belongs, the live location, and the like, and store a current live content list. When a new anchor joins in live broadcast or the current live broadcast content is finished, refreshing a current content live broadcast list, and informing each user side to update the live broadcast content list; a user data module 330 for storing user information and preference data;
the live application module 310 is configured in a specific electronic device, and configured for live recommendation, which may include: the camera shooting unit is used for acquiring images of live users; the identification unit is used for analyzing the gender, age, dressing style and emotion identification of the user; the feature unit is used for extracting features of the user according to the face recognition result and analyzing whether the user is a new user or an old user; and if the user is a new user, creating a new user record, storing user information, predicting user preference, and uploading the new user data record to a user data server. If the user is an old user, pulling the user record on the user data server; and the live broadcast classification unit is used for acquiring a live broadcast content list on the current WebRTC server, when the WebRTC server has new live broadcast content added or the current live broadcast content is ended, when the live broadcast application of the user is started, the WebRTC server should inform the user side, the user side updates the current live broadcast content list, and meanwhile, the live broadcast content which is watched by the user at the time is shielded, and the live broadcast time range is used for closing the live broadcast application when the live broadcast application is opened by the user. Matching marks in the live content list by using the analyzed user preference information, and ranking according to the number of the matched marks; the behavior analysis unit is used for analyzing whether the current live broadcast watching user frequently switches the live broadcast content or not, and informing the live broadcast program classification list module to recommend the live broadcast content again when the fact that the user frequently switches the live broadcast content is found; and the playing unit is used for acquiring the live broadcast list with good ranking of the live broadcast program classification list module when the user is watching the live broadcast content, and pulling the corresponding live WebRTC live broadcast multimedia stream for playing when the user switches the live broadcast content next time.
S390: and recommending the associated content corresponding to the current live content when the behavior information meets a second preset condition.
The electronic equipment can recommend the associated content except the live broadcast by analyzing the behavior trend of the user while performing live broadcast recommendation on the user, so that the experience degree of watching the live broadcast by the user is improved.
As an implementation manner, when the electronic device detects that a duration of live viewing of the live content by the user meets a second duration condition, the associated content corresponding to the current live content may be recommended, where the second duration condition is a time threshold preset in advance by the electronic device, and the time threshold may be an average duration of viewing of the live content by the user who generally prefers the live content. For example, when the electronic device detects that a user stays watching a live shopping time for a time corresponding to a threshold of watching the live shopping time, an item advertisement associated with the live content in the live can be recommended.
As another embodiment, the second preset condition may also be a behavior characteristic of the user watching the live content under a normal condition, where the behavior characteristic refers to information that the live content is clicked, praised, collected, and forwarded on the live platform historically. The electronic device may set a preset behavior threshold, and may recommend an item advertisement associated with the live content to the user in live broadcast when the behavior characteristics of the user meet the preset behavior threshold through data analysis.
The live broadcast recommendation method provided by the embodiment of the application acquires behavior information of a current user when the current user watches live broadcast content after live broadcast recommendation is performed on the user, updates the current live broadcast list when the behavior information meets a first preset condition, updates the current live broadcast list when the live broadcast watching duration meets a first time long condition and the live broadcast switching frequency reaches a preset threshold value, and selects target live broadcast content from the updated current live broadcast list for recommendation according to the relevance of at least one preference label and at least one classification label. In addition, when the behavior information meets a second preset condition, the associated content corresponding to the current live broadcast content is recommended, so that the recommendation instantaneity and the live broadcast watching viscosity of the user are improved.
Referring to fig. 8, a block diagram of a live broadcast recommendation apparatus 400 according to an embodiment of the present application is shown, where the apparatus may include: a detection module 410, a first acquisition module 420, a determination module 430, a second acquisition module 440, and a recommendation module 450. The detection module 410 is configured to, when detecting that a live application is started, obtain a user image of a current user; a first obtaining module 420, configured to obtain attribute information and a current emotion of the current user according to the user image; a determining module 430, configured to determine at least one preference tag of the current user according to the attribute information and the current emotion; a second obtaining module 440, configured to obtain a current live broadcast list, where each live broadcast content in the current live broadcast list corresponds to at least one category label; a recommending module 450, configured to select a target live content from the current live list to recommend according to the relevance of the at least one preference tag and the at least one classification tag
In some embodiments, the first obtaining module 420 may include: the identification unit is used for identifying the face information in the user image and judging whether the current user is a new user or not based on an identification result; and the first analysis unit is used for analyzing and processing the physical state of the current user in the user image to obtain the attribute information and the current emotion of the current user when the current user is a new user.
In some embodiments, the first obtaining module 420 may further include: the acquiring unit is used for acquiring the attribute information of the current user which is stored in advance when the current user is not a new user; and the second analysis unit is used for carrying out emotion analysis on the face information in the user image to obtain the current emotion of the current user.
In some embodiments, the determining module 430 may include: a candidate acquiring unit configured to acquire at least one candidate tag corresponding to the attribute information; a preference matching unit for selecting at least one preference tag matching the current emotion from the at least one candidate tag.
In some embodiments, recommendation module 450 may include: a matching unit, configured to match, for each live content in the current live content list, at least one category tag corresponding to the live content with the at least one preference tag; the quantity obtaining unit is used for obtaining the quantity of the preference labels successfully matched with each live broadcast content; the sorting unit is used for sorting the live broadcast contents in the current live broadcast list according to the sequence of the preference tag number from large to small to obtain a sorted live broadcast recommendation list; and the recommending unit is used for selecting target live broadcast content to recommend based on the live broadcast recommendation list.
In some embodiments, the live recommendation device 400 may further include: the behavior information acquisition module is used for acquiring the behavior information of the current user when watching the live broadcast content; the updating module is used for updating the current live broadcast list when the behavior information meets a first preset condition; and the re-recommendation module is used for selecting the target live broadcast content from the updated current live broadcast list for recommendation according to the relevance of the at least one preference label and the at least one classification label.
In some embodiments, the updating module may be specifically configured to update the current live broadcast list when the live broadcast watching time length satisfies a first time long condition and the live broadcast switching times reach a preset threshold.
In some embodiments, the live broadcasting recommendation apparatus 400 may further include an association recommendation module configured to recommend associated content corresponding to the current live broadcasting content when the behavior information satisfies a second preset condition.
In some embodiments, the association recommendation module may be specifically configured to recommend the association content corresponding to the current live content when the live viewing duration satisfies the second duration condition.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described apparatuses and modules may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, the coupling between the modules may be electrical, mechanical or other type of coupling.
In addition, functional modules in the embodiments of the present application may be integrated into one processing module, or each of the modules may exist alone physically, or two or more modules are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode.
Referring to fig. 9, a block diagram of an electronic device according to an embodiment of the present application is shown. The electronic device 100 may be a PC computer, a mobile terminal, or other electronic device capable of running an application. The electronic device 100 in the present application may include one or more of the following components: a processor 110, a memory 120, a camera 130, and one or more applications, wherein the one or more applications may be stored in the memory 120 and configured to be executed by the one or more processors 110, the one or more programs configured to perform the method as described in the aforementioned method embodiments, wherein the camera 130 is used to capture images and is passed to the processor 110 for processing.
The Memory 120 may include a Random Access Memory (RAM) or a Read-Only Memory (Read-Only Memory). The memory 120 may be used to store instructions, programs, code sets, or instruction sets. The memory 120 may include a stored program area and a stored data area, wherein the stored program area may store instructions for implementing an operating system, instructions for implementing at least one function (such as a touch function, a sound playing function, an image playing function, etc.), instructions for implementing various method embodiments described below, and the like. The storage data area may also store data created by the terminal 100 in use, such as a phonebook, audio-video data, chat log data, and the like.
Referring to fig. 10, a block diagram of a computer-readable storage medium according to an embodiment of the present application is shown. The computer-readable medium 800 has stored therein a program code that can be called by a processor to execute the method described in the above-described method embodiments.
The computer-readable storage medium 800 may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read only memory), an EPROM, a hard disk, or a ROM. Alternatively, the computer-readable storage medium 800 includes a non-volatile computer-readable storage medium. The computer readable storage medium 800 has storage space for program code 810 to perform any of the method steps of the method described above. The program code can be read from or written to one or more computer program products. The program code 810 may be compressed, for example, in a suitable form.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not necessarily depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.
Claims (12)
1. A live recommendation method, characterized in that the method comprises:
when detecting that a live application program is started, acquiring a user image of a current user;
acquiring attribute information and current emotion of the current user according to the user image;
determining at least one preference tag of the current user according to the attribute information and the current emotion;
acquiring a current live broadcast list, wherein each live broadcast content in the current live broadcast list corresponds to at least one classification label;
and selecting target live broadcast content from the current live broadcast list for recommendation according to the relevance of the at least one preference label and the at least one classification label.
2. The method according to claim 1, wherein the obtaining of the attribute information and the current emotion of the current user according to the user image comprises:
recognizing face information in the user image, and judging whether the current user is a new user or not based on a recognition result;
and when the current user is a new user, analyzing and processing the physical state of the current user in the user image to obtain the attribute information and the current emotion of the current user.
3. The method according to claim 2, wherein after the identifying the face information in the user image and determining whether the current user is a new user based on the identification result, the method further comprises:
when the current user is not a new user, acquiring attribute information of the current user, which is stored in advance;
and performing emotion analysis on the face information in the user image to obtain the current emotion of the current user.
4. The method of claim 1, wherein determining at least one preference tag of the current user based on the attribute information and a current mood comprises:
acquiring at least one candidate label corresponding to the attribute information;
selecting at least one preference tag from the at least one candidate tag that matches the current emotion.
5. The method of claim 1, wherein the selecting target live content from the current live list for recommendation according to the relevance of the at least one preference tag and the at least one category tag comprises:
for each live content in the current live list, matching at least one classification tag corresponding to the live content with the at least one preference tag;
acquiring the number of preference labels successfully matched with each live broadcast content;
sequencing the live broadcast contents in the current live broadcast list according to the sequence of the preference labels from large to small to obtain a sequenced live broadcast recommendation list;
and selecting target live broadcast content for recommendation based on the live broadcast recommendation list.
6. The method of any of claims 1-5, wherein after the selecting target live content from the current live list for recommendation according to the relevance of the at least one preference tag and the at least one category tag, the method further comprises:
acquiring behavior information of the current user when watching live content;
when the behavior information meets a first preset condition, updating the current live broadcast list;
and selecting target live broadcast content from the updated current live broadcast list for recommendation according to the relevance of the at least one preference label and the at least one classification label.
7. The method according to claim 6, wherein the behavior information includes a live viewing duration and a live switching number, and when the behavior information satisfies a first preset condition, updating the current live list includes:
and when the live broadcast watching time length meets a first time long condition and the live broadcast switching times reach a preset threshold value, updating the current live broadcast list.
8. The method of claim 6, wherein the behavior information comprises a live viewing duration, and after the obtaining of the behavior information of the current user viewing live content, the method further comprises:
and recommending the associated content corresponding to the current live content when the behavior information meets a second preset condition.
9. The method according to claim 8, wherein the behavior information includes a live viewing duration, and recommending associated content corresponding to current live content when the behavior information satisfies a second preset condition includes:
and recommending the associated content corresponding to the current live broadcast content when the live broadcast watching time length meets a second time length condition.
10. A live recommendation apparatus, the apparatus comprising:
the detection module is used for acquiring a user image of a current user when detecting that the live application program is started;
the first acquisition module is used for acquiring the attribute information and the current emotion of the current user according to the user image;
the determining module is used for determining at least one preference tag of the current user according to the attribute information and the current emotion;
the second acquisition module is used for acquiring a current live broadcast list, wherein each live broadcast content in the current live broadcast list corresponds to at least one classification label;
and the recommending module is used for selecting the target live broadcast content from the current live broadcast list for recommending according to the relevance of the at least one preference label and the at least one classification label.
11. An electronic device, comprising:
a memory;
one or more processors coupled with the memory;
one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, the one or more applications configured to perform the method of any of claims 1-9.
12. A computer-readable storage medium, having stored thereon program code that can be invoked by a processor to perform the method according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011496650.1A CN112528150A (en) | 2020-12-17 | 2020-12-17 | Live broadcast recommendation method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011496650.1A CN112528150A (en) | 2020-12-17 | 2020-12-17 | Live broadcast recommendation method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112528150A true CN112528150A (en) | 2021-03-19 |
Family
ID=75001079
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011496650.1A Pending CN112528150A (en) | 2020-12-17 | 2020-12-17 | Live broadcast recommendation method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112528150A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113190758A (en) * | 2021-05-21 | 2021-07-30 | 聚好看科技股份有限公司 | Server and media asset recommendation method |
| CN113207010A (en) * | 2021-06-02 | 2021-08-03 | 清华大学 | Model training method, live broadcast recommendation method, device and program product |
| CN113297953A (en) * | 2021-05-21 | 2021-08-24 | 北京读我网络技术有限公司 | Information processing method, device, electronic equipment and storage medium |
| CN113315988A (en) * | 2021-05-28 | 2021-08-27 | 北京中指讯博数据信息技术有限公司 | Live video recommendation method and device |
| CN113852861A (en) * | 2021-09-23 | 2021-12-28 | 深圳Tcl数字技术有限公司 | Program pushing method and device, storage medium and electronic equipment |
| CN114265777A (en) * | 2021-12-23 | 2022-04-01 | 北京百度网讯科技有限公司 | Application program testing method and device, electronic equipment and storage medium |
| CN116012118A (en) * | 2023-02-28 | 2023-04-25 | 荣耀终端有限公司 | Product recommendation method and device |
| CN116668788A (en) * | 2023-07-28 | 2023-08-29 | 中国标准化研究院 | Intelligent television content recommendation device and method |
| CN116761011A (en) * | 2023-08-21 | 2023-09-15 | 浙江印象软件有限公司 | Real-time loading method and system for special effect data of live virtual article |
| CN117725273A (en) * | 2023-09-21 | 2024-03-19 | 书行科技(北京)有限公司 | Sample label generation method, device, computer equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108509660A (en) * | 2018-05-29 | 2018-09-07 | 维沃移动通信有限公司 | A kind of broadcasting object recommendation method and terminal device |
| US10091554B1 (en) * | 2017-12-06 | 2018-10-02 | Echostar Technologies L.L.C. | Apparatus, systems and methods for generating an emotional-based content recommendation list |
| CN110175245A (en) * | 2019-06-05 | 2019-08-27 | 腾讯科技（深圳）有限公司 | Multimedia recommendation method, device, equipment and storage medium |
| CN111161035A (en) * | 2019-12-31 | 2020-05-15 | 北京三快在线科技有限公司 | Dish recommendation method and device, server, electronic equipment and storage medium |
-
2020
- 2020-12-17 CN CN202011496650.1A patent/CN112528150A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10091554B1 (en) * | 2017-12-06 | 2018-10-02 | Echostar Technologies L.L.C. | Apparatus, systems and methods for generating an emotional-based content recommendation list |
| CN108509660A (en) * | 2018-05-29 | 2018-09-07 | 维沃移动通信有限公司 | A kind of broadcasting object recommendation method and terminal device |
| CN110175245A (en) * | 2019-06-05 | 2019-08-27 | 腾讯科技（深圳）有限公司 | Multimedia recommendation method, device, equipment and storage medium |
| CN111161035A (en) * | 2019-12-31 | 2020-05-15 | 北京三快在线科技有限公司 | Dish recommendation method and device, server, electronic equipment and storage medium |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113190758A (en) * | 2021-05-21 | 2021-07-30 | 聚好看科技股份有限公司 | Server and media asset recommendation method |
| CN113297953A (en) * | 2021-05-21 | 2021-08-24 | 北京读我网络技术有限公司 | Information processing method, device, electronic equipment and storage medium |
| CN113315988A (en) * | 2021-05-28 | 2021-08-27 | 北京中指讯博数据信息技术有限公司 | Live video recommendation method and device |
| CN113315988B (en) * | 2021-05-28 | 2023-04-07 | 北京中指讯博数据信息技术有限公司 | Live video recommendation method and device |
| CN113207010A (en) * | 2021-06-02 | 2021-08-03 | 清华大学 | Model training method, live broadcast recommendation method, device and program product |
| CN113852861A (en) * | 2021-09-23 | 2021-12-28 | 深圳Tcl数字技术有限公司 | Program pushing method and device, storage medium and electronic equipment |
| CN114265777B (en) * | 2021-12-23 | 2023-01-10 | 北京百度网讯科技有限公司 | Application program testing method and device, electronic equipment and storage medium |
| CN114265777A (en) * | 2021-12-23 | 2022-04-01 | 北京百度网讯科技有限公司 | Application program testing method and device, electronic equipment and storage medium |
| CN116012118A (en) * | 2023-02-28 | 2023-04-25 | 荣耀终端有限公司 | Product recommendation method and device |
| CN116012118B (en) * | 2023-02-28 | 2023-08-29 | 荣耀终端有限公司 | Product recommendation method and device |
| CN116668788A (en) * | 2023-07-28 | 2023-08-29 | 中国标准化研究院 | Intelligent television content recommendation device and method |
| CN116668788B (en) * | 2023-07-28 | 2023-10-17 | 中国标准化研究院 | Intelligent television content recommendation device and method |
| CN116761011A (en) * | 2023-08-21 | 2023-09-15 | 浙江印象软件有限公司 | Real-time loading method and system for special effect data of live virtual article |
| CN116761011B (en) * | 2023-08-21 | 2023-11-14 | 浙江印象软件有限公司 | Real-time loading method and system for special effect data of live virtual article |
| CN117725273A (en) * | 2023-09-21 | 2024-03-19 | 书行科技(北京)有限公司 | Sample label generation method, device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112528150A (en) | Live broadcast recommendation method and device, electronic equipment and storage medium | |
| CN109769128B (en) | Video recommendation method, video recommendation device and computer-readable storage medium | |
| CN109819325B (en) | Hotspot video annotation processing method and device, computer equipment and storage medium | |
| WO2020221278A1 (en) | Video classification method and model training method and apparatus thereof, and electronic device | |
| CN110519617B (en) | Video comment processing method and device, computer equipment and storage medium | |
| CN109145784B (en) | Method and apparatus for processing video | |
| CN108769823B (en) | Direct broadcasting room display methods, device, equipment | |
| CN110166827B (en) | Video clip determination method and device, storage medium and electronic device | |
| CN110557659B (en) | Video recommendation method and device, server and storage medium | |
| CN111258435B (en) | Comment method and device for multimedia resources, electronic equipment and storage medium | |
| CN107563500B (en) | Video recommendation method and system based on user head portrait | |
| CN105635824B (en) | Personalized channel recommendation method and system | |
| CN104994426B (en) | Program video identification method and system | |
| US8457368B2 (en) | System and method of object recognition and database population for video indexing | |
| CN109862397B (en) | Video analysis method, device, equipment and storage medium | |
| CN110633669B (en) | Mobile terminal face attribute identification method based on deep learning in home environment | |
| CN109508406B (en) | Information processing method and device and computer readable storage medium | |
| CN111918130A (en) | Video cover determining method and device, electronic equipment and storage medium | |
| KR101895846B1 (en) | Facilitating television based interaction with social networking tools | |
| CN112348117A (en) | Scene recognition method and device, computer equipment and storage medium | |
| CN111597446A (en) | Content pushing method and device based on artificial intelligence, server and storage medium | |
| CN111432206A (en) | Video definition processing method and device based on artificial intelligence and electronic equipment | |
| CN111046209B (en) | Image clustering retrieval system | |
| KR20200077176A (en) | Apparatus and method for recommending keyword related to video | |
| CN112685596A (en) | Video recommendation method and device, terminal and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |