CN111353068A - Video recommendation method and device - Google Patents
Video recommendation method and device Download PDFInfo
- Publication number
- CN111353068A CN111353068A CN202010130419.4A CN202010130419A CN111353068A CN 111353068 A CN111353068 A CN 111353068A CN 202010130419 A CN202010130419 A CN 202010130419A CN 111353068 A CN111353068 A CN 111353068A
- Authority
- CN
- China
- Prior art keywords
- video
- videos
- user
- tag
- tags
- 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
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
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/73—Querying
- G06F16/735—Filtering based on additional data, e.g. user or group profiles
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses a video recommendation method, which comprises the following steps: after a video acquisition request of a user is detected, acquiring a history record of the user; the history record comprises play information of one or more videos, wherein the play information comprises a video label and play duration, and the video label is used for representing key information of the videos; acquiring a video set according to the video label; and recommending the videos in the video set to the user according to the playing time length. By the method and the device, the short video can be recommended by using music knowledge in the video, and the accuracy of video recommendation is improved.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a video recommendation method and device.
Background
With the continuous development of multimedia technology, the integration of music in short videos becomes a new social entertainment mode, and therefore, more and more music Applications (APPs) increase the short video functions and can increase the social and entertainment of the music APPs.
At present, a common music APP recommends short videos based on video background music, so that the music content of the recommended short videos is too single, and therefore, how to recommend the short videos according to the real requirements of users, and the improvement of the accuracy of short video recommendation becomes a problem to be solved urgently.
Disclosure of Invention
The embodiment of the application provides a recommendation method and device, which can recommend short videos by using music knowledge in the videos and improve the accuracy of video recommendation.
In a first aspect, an embodiment of the present application provides a video recommendation method, where the video recommendation method includes:
after a video acquisition request of a user is detected, acquiring a historical record of the user; the history record comprises play information of one or more videos, wherein the play information comprises video tags and play duration, and the video tags are used for representing key information of the videos;
acquiring a video set according to the video label;
and recommending the videos in the video set to the user according to the playing duration.
In a possible implementation manner, the obtaining a video set according to the video tag includes:
calculating an interest score of the video tag, wherein the interest score is used for representing the interest degree of the user in a preset time period;
selecting a label with the interest score of the video label higher than a first threshold value;
and acquiring videos from the video library according to the selected tags to obtain a video set.
In a possible implementation manner, the obtaining a video set according to the video tag includes:
and acquiring videos of which the similarity between the video tags corresponding to the videos in the video library and the video tags of the users is greater than a second threshold value to obtain a video set.
In a possible implementation manner, the obtaining a video set according to the video tag includes:
calculating the association degree of the user and each user in the user set;
selecting the user with the maximum association degree with the user in the user set;
extracting the video tags in the historical records of the users with the maximum relevance;
and acquiring a video from a video library according to the video tag of the user and the video tag of the user with the maximum association degree to obtain a video set.
In a possible implementation manner, the obtaining a video from a video library according to the video tag of the user and the video tag of the user with the largest association degree to obtain a video set includes:
acquiring videos from a video library according to the video tags of the users and the video tags of the users with the maximum relevance degree to obtain a plurality of videos;
and eliminating the videos of which the playing quantity is lower than a third threshold value from the plurality of videos to obtain a video set.
In a possible implementation manner, the recommending videos in the video set to the user according to the playing duration includes:
calculating the score of the video in the video set according to the playing time length;
sequencing the videos in the video set from high to low according to the corresponding scores to obtain a video list;
and displaying the videos to the user according to the sequence of the videos in the video list.
In a possible implementation manner, the calculating scores of videos in the video set according to the play duration includes:
calculating the watching proportion of the videos in the historical record of the user according to the playing time length and the total time length of the videos corresponding to the playing time length, and determining the score of the videos in the historical record according to the watching proportion;
determining a target video corresponding to a video tag of a video in the historical record in the video set;
and determining the score of the video in the historical record as the score of the target video.
In a possible implementation manner, the obtaining a video set according to the video tag includes:
the method comprises the steps of obtaining a preset video library, wherein videos in the video library are obtained according to any one or more of video parameters, video playing amount and an uploading account number of the videos;
and acquiring a video set from a preset video library according to the video label.
In one possible implementation, the method further includes:
extracting one or more music keywords from the title and/or content of a first video, wherein the first video is any video in the video library;
matching the one or more music keywords with vocabularies in a word bank;
and determining the corresponding music keywords of which the matching degree is greater than a fourth threshold value in the one or more music keywords as the video tags of the first video.
In one possible implementation, the history record further includes one or more audio playing information, and the audio playing information includes an audio tag; then, the obtaining a video set according to the video tag includes:
and acquiring a video set from a preset video library according to the video label and/or the audio label.
In a second aspect, an embodiment of the present invention provides a video recommendation apparatus, where the video recommendation apparatus includes:
the device comprises a first acquisition unit, a second acquisition unit and a third acquisition unit, wherein the first acquisition unit is used for acquiring the history of a user after detecting a video acquisition request of the user; the history record comprises play information of one or more videos, wherein the play information comprises video tags and play duration, and the video tags are used for representing key information of the videos;
the second acquisition unit is used for acquiring a video set according to the video label;
and the recommending unit is used for recommending the videos in the video set to the user according to the playing time length.
In a possible implementation manner, the second obtaining unit is specifically configured to:
calculating an interest score of the video tag, wherein the interest score is used for representing the interest degree of the user in a preset time period;
selecting a label with the interest score of the video label higher than a first threshold value;
and acquiring videos from the video library according to the selected tags to obtain a video set.
In a possible implementation manner, the second obtaining unit is specifically configured to:
and acquiring videos of which the similarity between the video tags corresponding to the videos in the video library and the video tags of the users is greater than a second threshold value to obtain a video set.
In a possible implementation manner, the second obtaining unit is specifically configured to:
calculating the association degree of the user and each user in the user set;
selecting the user with the maximum association degree with the user in the user set;
extracting the video tags in the historical records of the users with the maximum relevance;
and acquiring a video from a video library according to the video tag of the user and the video tag of the user with the maximum association degree to obtain a video set.
In a possible implementation manner, the obtaining, by the second obtaining unit, a video from a video library according to the video tag of the user and the video tag of the user with the largest association degree to obtain a video set includes:
acquiring videos from a video library according to the video tags of the users and the video tags of the users with the maximum relevance degree to obtain a plurality of videos;
and eliminating the videos of which the playing quantity is lower than a third threshold value from the plurality of videos to obtain a video set.
In a possible implementation manner, the recommending unit is specifically configured to:
calculating the score of the video in the video set according to the playing time length;
sequencing the videos in the video set from high to low according to the corresponding scores to obtain a video list;
and displaying the videos to the user according to the sequence of the videos in the video list.
In a possible implementation manner, the calculating, by the recommending unit, the score of the video in the video set according to the playing duration includes:
calculating the watching proportion of the videos in the historical record of the user according to the playing time length and the total time length of the videos corresponding to the playing time length, and determining the score of the videos in the historical record according to the watching proportion;
determining a target video corresponding to a video tag of a video in the historical record in the video set;
and determining the score of the video in the historical record as the score of the target video.
In a possible implementation manner, the second obtaining unit is specifically configured to:
the method comprises the steps of obtaining a preset video library, wherein videos in the video library are obtained according to any one or more of video parameters, video playing amount and an uploading account number of the videos;
and acquiring a video set from a preset video library according to the video label.
In one possible implementation, the apparatus further includes:
a third obtaining unit, configured to extract one or more music keywords from a title and/or content of a first video, where the first video is any video in the video library;
the matching unit is used for matching the one or more music keywords with vocabularies in a word bank;
a determining unit, configured to determine, as the video tag of the first video, a music keyword of which a matching degree is greater than a fourth threshold in the one or more music keywords.
In one possible implementation, the history record further includes one or more audio playing information, and the audio playing information includes an audio tag; the first obtaining unit is specifically configured to:
and acquiring a video set from a preset video library according to the video label and/or the audio label.
In a third aspect, an embodiment of the present application provides an electronic device, which includes an output device, an input device, a processor, and a memory, where the output device, the input device, the processor, and the memory are connected to each other. The memory is configured to store a computer program that supports the terminal device to execute the method provided by the first aspect and/or any one of the possible implementation manners of the first aspect, where the computer program includes program instructions, and the processor is configured to call the program instructions to execute the method provided by the first aspect and/or any one of the possible implementation manners of the first aspect.
In a fourth aspect, the present application provides a computer-readable storage medium, which stores a computer program, where the computer program includes program instructions, and the program instructions, when executed by a processor, cause the processor to execute the method provided by the first aspect and/or any one of the possible implementation manners of the first aspect.
In the embodiment of the invention, after responding to a video acquisition request for video recommendation, acquiring the history record of the user; the history record comprises play information of one or more videos, wherein the play information comprises a video label and play duration, and the video label is used for representing key information of the videos; acquiring a video set according to the video label; and recommending the videos in the video set to the user according to the playing time length. Through analyzing the playing time length and predicting the recommendation scores of the videos in the video recommendation set, and recommending from high to low according to the scores, the videos can be recommended individually, and the accuracy of video recommendation can be improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, 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 some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic diagram of a network architecture according to an embodiment of the present invention;
fig. 2 is a schematic flowchart of a video recommendation method according to an embodiment of the present invention;
FIG. 3a is a schematic diagram of a user interface for video recommendation provided by embodiments of the present invention;
FIG. 3b is a schematic diagram of another user interface for video recommendation provided by an embodiment of the present invention;
FIG. 3c is a schematic diagram of another user interface for video recommendation provided by an embodiment of the present invention;
FIG. 3d is a schematic diagram of yet another user interface for video recommendation provided by an embodiment of the present invention;
FIG. 3e is a schematic diagram of still another user interface for video recommendation provided by an embodiment of the present invention;
FIG. 4 is a schematic flow chart of determining a video tag according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a video recommendation apparatus according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of a terminal device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the invention provides a video recommendation method, which is used for recommending videos in music APP and improving the accuracy of recommendation by using music knowledge; the video recommendation method can be applied to a terminal, where the terminal device may include but is not limited to: smart phones, tablets, laptops, and desktops, among others. The terminal equipment can call the corresponding application client to execute the video recommendation method according to the actual service requirement. For example, the processing of short video recommendations by music-like software clients may be invoked, where music-like software application clients may include, but are not limited to: mobile client, PC (personal computer) client, and the like; the mobile client is a music software client running in the terminal equipment, and the PC client is a music software client running in the PC; but also video recommendations of web pages running on browsers, etc.
The following explains a video recommendation method provided by an embodiment of the present invention, taking the application of the video recommendation method in the network architecture shown in fig. 1 and taking a music software client in a terminal device as an example: referring to fig. 1, the network architecture diagram may include: server, terminal equipment and user. Each terminal device corresponds to one or more users, a client of music class software can be installed in the terminal device, and each user can watch videos of music classes through the installed client. The server is used for storing videos in the platform, and the terminal equipment acquires one or more videos from the videos in the platform and recommends the videos to the user. For example, after the terminal device detects a video acquisition request input by a user, a historical play record of the user may be acquired, where the historical record includes a video tag corresponding to each video in a plurality of videos watched by the user, a play duration corresponding to each video, and an audio tag corresponding to each audio in a plurality of audios played by the user, the video tag is a music keyword extracted from each video, and the audio tag is a music keyword extracted from the audio. Further, the video set is obtained from the video library according to the video tags and the audio tags, the video set is divided into two aspects, on one hand, videos with the similarity between the video tags corresponding to videos in the video library and the video tags extracted from the user history record and/or the tags in the audio tags larger than a threshold value are obtained from the video library, on the second hand, the users with the maximum relevance with the users are obtained according to the video tags, the video tags and/or the audio tags in the user history record with the maximum relevance are extracted, and videos corresponding to the video tags and/or the audio tags in the user history record with the maximum relevance are obtained and extracted from the video library. After the video set is obtained, the history record further includes the playing time of each video, the playing proportion of the videos in the user history record can be calculated according to the playing time, the playing proportion is input into a preset classification model, the model includes the incidence relation between the watching proportion and the score, so that the output score can be obtained according to the model, the score can represent the score of the label corresponding to the watched video, each label corresponds to one or more videos in the video set, the score of each video in the set can be obtained, and the videos in the video set are recommended to the user according to the sequence of the scores from high to low. The terminal may include a mobile phone, a tablet computer, a notebook computer, a palm computer, a Mobile Internet Device (MID), and the like, which is not limited herein.
Referring to fig. 2, fig. 2 is a flowchart illustrating a video recommendation method according to an embodiment of the present invention. The video recommendation method provided by the embodiment of the invention can comprise the following steps 201-203:
201. and after a video acquisition request of the user is detected, acquiring the history of the user.
Specifically, the video acquisition request is triggered by the user based on the terminal device and is used for acquiring the recommended video. The video acquisition request may be triggered when the user performs an operation indicated in the client. The client can instruct the user to perform sliding, clicking and other operations to acquire the video. It can be understood that when each user uses the client, an account needs to be registered, and in the recommendation process, recommendation is performed on the history of the account corresponding to each user.
After detecting a video acquisition request of a user, acquiring music and video records played in a history record of the user, specifically including music listened to by the user and videos watched by the user, wherein each song in the listened music includes one or more corresponding music tags and one or more corresponding video tags for each video watched, and a playing time length of each video, it should be noted that, in order to better recommend through the playing record of the user, the tags of the videos and the tags of the music are of the same type, and include one or more tags, such as singer information, song information, and genre information. The tags for video or music may be obtained from the title or from the video content.
It should be noted that the music tag can solve the problem of the cold start of the short video, that is, under the condition that there is no history of the video, the music tag can acquire a video set according to the audio tag and recommend the video set to the user.
202. And acquiring a video set according to the video label.
In one possible implementation, videos may be obtained from a video library from two aspects, resulting in a video set.
One aspect is by calculating interest scores for individual tags that represent the user's recent level of interest, which takes into account a time decay process to ensure that videos that are of recent interest to the user are recommended. Specifically, the closer the current time is to the time of the tag appearing in the history, the higher the calculated score, and the longer the current time is from the time of the tag appearing in the history, the lower the calculated score, and the degree of current interest of the user in the tag is represented by the calculated attenuation value of interest at the current time.
An example of an interest score calculation formula for a tag is as follows:
e(-0.6931472*(now_t-last_t)/3600/24/30)
taking a tag as an example, now _ t refers to the current time, last _ t refers to the time of the video corresponding to the tag being played last time in the history play record, where the time is in seconds, it can be understood that there are one or more tags corresponding to one video, where last _ t denotes the time when the tag has appeared last time, may be the time of the tag corresponding to the video played last time in the history play record, or may be the time corresponding to the video played last time in the videos corresponding to the tag. Wherein, if the score is calculated according to the formula, the score is set to 1 as a full score. Optionally, N may also be set0And if the score is full, the full score represents an initial interest score, the initial interest score is multiplied by the formula to obtain the score of each label, and another interest score calculation formula of the label is as follows:
N0e(-0.6931472*(now_t-last_t)/3600/24/30)
and selecting the labels with the scores higher than a first threshold value from the labels, wherein the first threshold value can be set according to the number of the acquired videos, and the videos corresponding to the labels are acquired from a video library according to the selected labels to obtain a video set without limitation.
In the second aspect, the video is obtained according to the similarity of the tags, specifically, the similarity is divided into two parts, and the first part can obtain the video with the similarity higher than the threshold value with the tag text from the video library according to the tags. The tags can be selected from the titles of the videos, and are information such as keywords, singers, songs, genres and the like of the videos extracted from the titles of the videos, and can also be music information extracted from the videos. Calculating the similarity of the labels with the same type corresponding to each video in the two videos to obtain the similarity of the two videos, for example, dividing the label types extracted from the title in the video A into title keywords, songs, singers, genres, manually labeled words and the like of the videos, wherein one word is regarded as one label. Calculating similarity of the corresponding type words extracted from the video B, specifically, calculating the text similarity may be calculating the text similarity by using cosine similarity algorithm (cosine), calculating the text similarity by using euclidean distance, calculating the text similarity by using manhattan distance (manhattan distance), and not limited herein. Under the condition that the label similarity of title keyword labels, song labels, singer labels and manually marked words in the A video and the B video is obtained respectively, different weights can be set respectively, so that the text similarity is obtained.
Wherein, the weight of the singer label can be set to be 0.5, the weight of the song label is 0.25, the weight of the keyword label is 1, and the weight of the manually labeled word is 1, and then the text similarity can be calculated according to the following formula:
content_sim＝0.5*singer_sim+0.25*song_sim+
title _ word _ sim + tag _ sim. Wherein content _ sim refers to text similarity, singer _ sim refers to similarity of singer labels, song _ sim refers to similarity of song labels, title _ word _ sim refers to similarity of title keywords, and tag _ sim refers to similarity of manually labeled word labels. It is to be understood that the above-mentioned weights are only examples, and may be specifically set according to a specific scene or effect, and the text similarity is calculated.
Obtaining a part of videos according to the similarity, which may be obtained according to the collaborative similarity, includes obtaining videos and music tags played in a current history record of a user a, obtaining a user set having a plurality of same video tags as the user a, calculating a user B having a maximum similarity with the user a in the user set, obtaining tags in the history record of the user B, obtaining at least one tag, searching a corresponding video in a video library for the obtained tag, and obtaining a video set as an alternative of the video set of the user a. The user B with the maximum label similarity to the user A in the user set can be calculated by respectively comparing a plurality of labels of the user A and the user B, wherein each label corresponds to one label identifier, when the label identifiers of the two users are the same, the label identifiers are scored, and the similarity of the two users is determined according to the number of the labels which are the same. For example, when the comparison shows that the identifiers of the N tags in the tags of the user B are the same as the identifiers of the N tags in the tags of the user a, the score is recorded by N, and the score can be used to represent the similarity between the user a and the user B. The collaborative similarity may also obtain a user set having the same video tag as the user a, and a video played in a history of one or more users in the user set may be used as an alternative to the user a's video set. The collaborative similarity can be calculated by searching a video list after keyword aggregation and the like as a sequence and applying a graph embedding algorithm (such as a node2vec algorithm). The calculation method is not limited here.
Further, in the calculation process, in consideration of the viewability of the video, the video with the playing amount lower than the third threshold value can be removed from the acquired video, and in the calculation method, in the process of designing the weight, the playing amount can be set to be prone to hot video, so as to ensure the recommended hot video. And recommending the obtained video set to the user.
Specifically, in the process of obtaining, for a large amount of videos in the platform, in order to obtain videos corresponding to the tags more quickly and conveniently, each video in the video library obtained by the video in the platform or through primary screening, the corresponding tag and the serial number of the video may be input into an existing index construction tool, and construction of the index is completed, that is, the inverted list is obtained, wherein the construction tool of the index may adopt an elastic search and the like. The method for constructing the index and acquiring the video is not limited herein.
In a possible implementation manner, when the video set is obtained, the video set may be obtained from a video in the platform, or may be a video obtained by performing preliminary screening on the video in the platform. And after the preliminary screening, obtaining a video library, and acquiring videos from the video library to obtain a video set. Specifically, the video may be filtered from three dimensions: firstly, screening basic parameters of video parameters, such as video release time, video code rate and video definition, and leaving videos meeting the quality of basic contents, namely videos with video parameters larger than or equal to standard parameters; secondly, screening the playing amount of the video, selecting the video with the playing amount in the platform within a preset time period, and selecting the video with the playing amount larger than a playing amount threshold value, wherein the preset time period can be within a week and 15 days, and the method is not limited; and thirdly, a white list of video uploading accounts can be provided, at least one high-quality video is uploaded by the accounts on the white list, and the corresponding operator who has experience in operating the video can directly enter the video library. The three conditions can be satisfied, wherein one of the three conditions can enter the video library, and the three conditions can also enter the video library when any two conditions are satisfied simultaneously.
In the embodiment of the invention, the videos are acquired and recommended by calculating the interest scores of the users to the videos along with the time, so that the videos which are interested recently can be recommended to the users, and the accuracy of video recommendation is improved. The videos similar to videos played in user history records can be obtained through the text similarity, videos similar to the videos played in the user history records can be obtained through calculating the user similarity, the videos similar to the users and liked by the users can be used as recommendations, the recommendation accuracy can be improved, meanwhile, the videos are obtained through the tags, not only music based on the videos but also music knowledge contained in the videos are recommended, and therefore the music knowledge in the videos can be understood more deeply.
203. And recommending the videos in the video set to the user according to the playing time length.
In a possible implementation manner, after the video set is obtained, the videos in the set need to be scored, and the score is used for indicating whether the predicted video in the video set is a video really interested by the user. A higher score indicates a higher degree of interest. And according to the scores, the videos in the video set can be sorted from high to low according to the corresponding scores to obtain the recommendation sequence of the videos, and the videos are displayed to the user according to the sorting sequence.
The playing time length can be input into a preset classification model to obtain the score of the video label of the video corresponding to the playing time length. The process of the classification model to calculate the video tag score may be: calculating the watching proportion of the video according to the playing time length of the video in the user history record and the total time length of the video, setting the video with the watching proportion larger than or equal to a preset proportion threshold value as a positive example, setting the video with the watching proportion smaller than the preset proportion as a negative example, and determining the score of the video corresponding to the watching proportion according to the incidence relation between the watching proportion and the score, wherein the video is provided with one or more video labels, and therefore the score of the video is the score of the one or more video labels. Since each tag corresponds to one or more videos in the video set, the scores of the videos in the video set can be determined accordingly.
And sequencing the videos in the set according to the scores to obtain a video list, and selecting the top N videos in the video list as recommended videos to be displayed to a user. N is an integer greater than or equal to 1. The display to the user may be direct playing or recommendation to the user in sequence, and the display mode is not specified here.
In a possible implementation manner, the operation process of the user using the terminal device can be seen in fig. 3a to 3 e. Fig. 3a to 3e are schematic diagrams of user interfaces for video recommendation provided in an embodiment of the present invention. As shown in fig. 3a, fig. 3a illustrates one way of recommending an interface on the client side of music software. In the interface diagram in fig. 3a, a user may select songs and videos recommended according to a history to listen on trial or watch, where a specific music or video selection mode may be implemented in a touch mode (clicking a position where any music or video is located), a voice control mode (inputting any song name or video name by voice), and the like, which is not limited herein. At this time, the video a and the video B are displayed in the interface, and the video a and the video B are only two videos as an example, and more videos can be obtained by sliding downwards.
In a possible implementation manner, when a user wants to obtain a new recommended video or music, the new recommendation may be pulled, specifically, the new recommendation may be obtained only by obtaining the new recommended video, or may be obtained by obtaining the new music, or may be obtained simultaneously, which is not limited herein. The specific obtaining manner may be to obtain the recommendation by sliding down as shown in fig. 3b to fig. 3d, may also be a touch manner (double-click blank area), and may also be an operation of shaking the terminal device, which is not limited herein. It is understood that during the operation, the user interface may be changed correspondingly with the user operation for better visual effect and user experience, as shown in fig. 3c and 3 d. After the video set is obtained, the video sets are displayed in a display interface according to the order of the video scores, as shown in fig. 3e, where video C and video D are displayed in the interface. Video C and video D are videos recommended after refresh. It should be noted that the layout manner of the current display interface is merely an example, and may include, but is not limited to, the manner of the current user interface.
In the embodiment of the invention, after responding to a video acquisition request for video recommendation, acquiring the history record of the user; the history record comprises play information of one or more videos, wherein the play information comprises a video label and play duration, and the video label is used for representing key information of the videos; acquiring a video set according to the video label; and recommending the videos in the video set to the user according to the playing time length. Through analyzing the playing time length and predicting the recommendation scores of the videos in the video recommendation set, and recommending from high to low according to the scores, the videos can be recommended individually, and the accuracy of video recommendation can be improved.
Referring to fig. 4, fig. 4 is a schematic flowchart of determining a video tag according to an embodiment of the present invention. The video recommendation method provided by the embodiment of the invention can comprise the following steps 401-:
401. one or more music keywords are extracted from the title and/or content of the first video.
Specifically, the first video is any one of a platform video or a video library obtained through preliminary screening, where the platform video may have a time length threshold and may be a short video lower than a preset time threshold. Each video in the video library corresponds to a video tag, and the video tag is a music keyword extracted from the title or content of the video. For example, words containing music knowledge may be extracted from the title or video content of the first video as tags for the first video. It is to be understood that the videos of music category can be divided into Music Video (MV) produced by record companies and videos related to music contents produced by non-record companies. In order to better understand the music knowledge contained in the video more deeply, corresponding singer information, song information and genre information can be extracted from the title of the video, and music information can be extracted from the video content. The content may be directly extracted or manually labeled, which is not limited herein.
402. And matching the one or more music keywords with the vocabularies in the word stock.
In a possible implementation manner, one or more music keywords extracted from the title and/or content of the first video are matched with words in a word bank, and the music keywords with the matching degree greater than a preset value are used as the tags of the video. The word bank is word bank information obtained by the user knowledge accumulation through the song bank knowledge in the platform. The matching may be to calculate the text similarity between the music keyword and the words in the word stock, or to search the music keyword in the word stock, and determine the music keyword as a matched tag in the case that the music keyword is searched. If the music keyword is not retrieved, the music keyword is added to an existing word stock. For example, if the title of the current video is "chua xun-only because of you too beautiful-MV", the music keywords that can be extracted from the title are "chua xun" and "only because of you too beautiful", wherein the contents in the current video can be manually labeled with "sing jump", "rap", "practise student", "basketball", etc., and then the extracted and manually labeled music keywords are matched with the words in the word stock.
403. And determining the corresponding music keyword of the one or more music keywords, the matching degree of which is greater than a fourth threshold value, as the video tag of the first video.
In a possible implementation manner, the music keywords with similarity greater than a fourth threshold to the word texts in the platform thesaurus may be determined as matching music keywords, and the matching music keywords are used as the video tags of the first video. And the music keywords can be directly searched in a word bank, and the searched words are used as video tags. Further, the videos may be stored in a database in correspondence with their tags. Wherein the manner of storage may include, but is not limited to, the examples of table 1.
TABLE 1
The video id is a serial number of the video, one or more singers in the current video may be in the singer list, one or more songs played in the video may be in the song list, the genre list may be a genre corresponding to one or more songs in the current video, and the keyword list may be singer information extracted from the singer list, song information extracted from the song list, or genre information extracted from the genre list, which is not limited herein. Taking the video "chua xun-only because you are too beautiful-MV" as an example, storing the video in correspondence with the tag may be as in the example of table 2.
TABLE 2
In the embodiment of the invention, after responding to a video acquisition request for video recommendation, acquiring the history record of the user; the history record comprises play information of one or more videos, wherein the play information comprises a video label and play duration, and the video label is used for representing key information of the videos; acquiring a video set according to the video label; and recommending the videos in the video set to the user according to the playing time length. Through analyzing the playing time length and predicting the recommendation scores of the videos in the video recommendation set, and recommending from high to low according to the scores, the videos can be recommended individually, and the accuracy of video recommendation can be improved.
Referring to fig. 5, fig. 5 is a schematic structural diagram of a video recommendation apparatus according to an embodiment of the present invention. The video recommendation apparatus 5000 provided in the embodiment of the present invention includes:
a first obtaining unit 501, configured to obtain a history record of a user after detecting a video obtaining request of the user; the history record comprises play information of one or more videos, wherein the play information comprises a video label and play duration, and the video label is used for representing key information of the videos;
a second obtaining unit 502, configured to obtain a video set according to the video tag;
a recommending unit 503, configured to recommend a video in the video set to the user according to the playing time length.
In a possible implementation manner, the second obtaining unit 502 is specifically configured to:
calculating an interest score of the video tag, wherein the interest score is used for representing the interest degree of the user in a preset time period;
selecting a label with the interest score of the video label higher than a first threshold value;
and acquiring videos from the video library according to the selected tags to obtain a video set.
In a possible implementation manner, the second obtaining unit 502 is specifically configured to:
and acquiring videos of which the similarity between the video tags corresponding to the videos in the video library and the video tags of the users is greater than a second threshold value to obtain a video set.
In a possible implementation manner, the second obtaining unit 502 is specifically configured to:
calculating the association degree between the user and each user in the user set;
selecting the user with the maximum association degree with the user in the user set;
extracting the video tags in the historical records of the users with the maximum relevance;
and acquiring a video from a video library according to the video tag of the user and the video tag of the user with the maximum association degree to obtain a video set.
In a possible implementation manner, the obtaining, by the second obtaining unit 502, a video from a video library according to the video tag of the user and the video tag of the user with the largest association degree, to obtain a video set includes:
acquiring videos from a video library according to the video tags of the users and the video tags of the users with the maximum association degree to obtain a plurality of videos;
and removing the videos of which the playing quantity is lower than a third threshold value from the plurality of videos to obtain a video set.
In a possible implementation manner, the recommending unit 503 is specifically configured to:
calculating the score of the video in the video set according to the playing time length;
sorting the videos in the video set from high to low according to corresponding scores to obtain a video list;
and displaying the videos to the user according to the sequence of the videos in the video list.
In a possible implementation manner, the calculating, by the recommending unit 503, the score of the video in the video set according to the playing time length includes:
calculating the watching proportion of the video in the history record of the user according to the playing time length and the total time length of the video corresponding to the playing time length, and determining the score of the video in the history record according to the watching proportion;
determining a target video corresponding to the video label of the video in the history record in the video set;
and determining the score of the video in the user record as the score of the target video in the video set.
In a possible implementation manner, the second obtaining unit 502 is specifically configured to:
acquiring a preset video library, wherein videos in the video library are acquired according to any one or more of video parameters, video playing amount and an uploading account number of the videos;
and acquiring a video set from a preset video library according to the video label.
In a possible implementation manner, the apparatus 5000 further includes:
a third obtaining unit 504, configured to extract one or more music keywords from a title and/or content of a first video, where the first video is any video in the video library;
a matching unit 505, configured to match the one or more music keywords with vocabularies in a lexicon;
a determining unit 506, configured to determine a music keyword, of the one or more music keywords, whose corresponding matching degree is greater than a fourth threshold value, as a video tag of the first video.
In a possible implementation manner, the history further includes one or more pieces of audio playing information, where the audio playing information includes an audio tag; the second obtaining unit 502 is specifically configured to:
and acquiring a video set from a preset video library according to the video label and/or the audio label.
In a specific implementation, the video recommendation apparatus may execute the implementation manners provided in the steps in fig. 1 to fig. 4 through the built-in modules and/or units of the video recommendation apparatus, which are not described herein again.
In the device provided by the embodiment of the invention, after responding to a video acquisition request for video recommendation, the history record of the user is acquired; the history record comprises play information of one or more videos, wherein the play information comprises a video label and play duration, and the video label is used for representing key information of the videos; acquiring a video set according to the video label; and recommending the videos in the video set to the user according to the playing time length. Through analyzing the playing time length and predicting the recommendation scores of the videos in the video recommendation set, and recommending from high to low according to the scores, the videos can be recommended individually, and the accuracy of video recommendation can be improved.
Referring to fig. 6, fig. 6 is a schematic structural diagram of a terminal device according to an embodiment of the present invention. As shown in fig. 6, the terminal device 6000 may include:
one or more processors 601, input devices 602, output devices 603, memory 604, transceivers 605. The processor 601, the input device 602, the output device 603, the memory 604, and the transceiver 605 are connected via a bus. Wherein the input device 602 may include a touch screen, a keyboard, a microphone, etc., the output device 603 may include a display screen, a speaker, etc., and the transceiver 605 is used for receiving and transmitting data. The memory 604 is used for storing a computer program comprising program instructions, and the processor 601 is used for executing the program instructions stored in the memory 604, wherein the processor 601 is configured for calling the program instructions to execute the following steps:
after a video acquisition request of a user is detected, acquiring a history record of the user; the history record comprises play information of one or more videos, wherein the play information comprises a video label and play duration, and the video label is used for representing key information of the videos;
acquiring a video set according to the video label;
and recommending the videos in the video set to the user according to the playing time length.
In a possible implementation manner, the obtaining, by the processor 601, a video set according to the video tag includes:
calculating an interest score of the video tag, wherein the interest score is used for representing the interest degree of the user in a preset time period;
selecting a label with the interest score of the video label higher than a first threshold value;
and acquiring videos from the video library according to the selected tags to obtain a video set.
In a possible implementation manner, the obtaining, by the processor 601, a video set according to the video tag includes:
and acquiring videos of which the similarity between the video tags corresponding to the videos in the video library and the video tags of the users is greater than a second threshold value to obtain a video set.
In a possible implementation manner, the obtaining, by the processor 601, a video set according to the video tag includes:
calculating the association degree between the user and each user in the user set;
selecting the user with the maximum association degree with the user in the user set;
extracting the video tags in the historical records of the users with the maximum relevance;
and acquiring a video from a video library according to the video tag of the user and the video tag of the user with the maximum association degree to obtain a video set.
In a possible implementation manner, the obtaining, by the processor 601, a video from a video library according to the video tag of the user and the video tag of the user with the largest association degree to obtain a video set includes:
acquiring videos from a video library according to the video tags of the users and the video tags of the users with the maximum association degree to obtain a plurality of videos;
and removing the videos of which the playing quantity is lower than a third threshold value from the plurality of videos to obtain a video set.
In a possible implementation manner, the recommending, by the processor 601, a video in the video set to the user according to the play time includes:
calculating the score of the video in the video set according to the playing time length;
sorting the videos in the video set from high to low according to corresponding scores to obtain a video list;
and displaying the videos to the user according to the sequence of the videos in the video list.
In a possible implementation manner, the calculating, by the processor 601, scores of videos in the video set according to the playing time duration includes:
calculating the watching proportion of the video in the history record of the user according to the playing time length and the total time length of the video corresponding to the playing time length, and determining the score of the video in the history record according to the watching proportion;
determining a target video corresponding to the video label of the video in the history record in the video set;
and determining the score of the video in the user record as the score of the target video in the video set.
In a possible implementation manner, the obtaining, by the processor 601, a video set according to the video tag includes:
acquiring a preset video library, wherein videos in the video library are acquired according to any one or more of video parameters, video playing amount and an uploading account number of the videos;
and acquiring a video set from a preset video library according to the video label.
In a possible implementation manner, the processor 601 is further configured to invoke program instructions to perform the following steps:
extracting one or more music keywords from the title and/or content of a first video, wherein the first video is any video in the video library;
matching the one or more music keywords with vocabularies in a word bank;
and determining the corresponding music keyword of the one or more music keywords, the matching degree of which is greater than a fourth threshold value, as the video tag of the first video.
In a possible implementation manner, the history further includes one or more pieces of audio playing information, where the audio playing information includes an audio tag; then the processor 601 obtains a video set according to the video tag, including:
and acquiring a video set from a preset video library according to the video label and/or the audio label.
In the electronic device provided by the embodiment of the invention, after responding to a video acquisition request for video recommendation, the history of the user is acquired; the history record comprises play information of one or more videos, wherein the play information comprises a video label and play duration, and the video label is used for representing key information of the videos; acquiring a video set according to the video label; and recommending the videos in the video set to the user according to the playing time length. Through analyzing the playing time length and predicting the recommendation scores of the videos in the video recommendation set, and recommending from high to low according to the scores, the videos can be recommended individually, and the accuracy of video recommendation can be improved.
It should be understood that in some possible embodiments, the processor 601 may be a Central Processing Unit (CPU), and the processor 601 may also be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and so on. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory 604 may include both read-only memory and random access memory, and provides instructions and data to the processor. A portion of the memory 604 may also include non-volatile random access memory.
In a specific implementation, the terminal device 6000 may execute the implementation manners provided in the steps in fig. 1 to fig. 4 through each built-in functional module thereof, which may specifically refer to the implementation manners provided in the steps, and no further description is given here.
The terms "first", "second", and the like in the claims, in the description and in the drawings of the present invention are used for distinguishing between different objects and not for describing a particular order. Furthermore, the terms "include" and "have," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or apparatus. Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the invention. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments. The term "and/or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.
Those of ordinary skill in the art will appreciate that the elements and algorithm steps of the examples described in connection with the embodiments disclosed herein may be embodied in electronic hardware, computer software, or combinations of both, and that the components and steps of the examples have been described in a functional general in the foregoing description for the purpose of illustrating clearly the interchangeability of hardware and software. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
The above disclosure is only for the purpose of illustrating the preferred embodiments of the present invention, and it is therefore to be understood that the invention is not limited by the scope of the appended claims.
Claims (13)
1. A method for video recommendation, comprising:
after a video acquisition request of a user is detected, acquiring a historical record of the user; the history record comprises play information of one or more videos, wherein the play information comprises video tags and play duration, and the video tags are used for representing key information of the videos;
acquiring a video set according to the video label;
and recommending the videos in the video set to the user according to the playing duration.
2. The method of claim 1, wherein obtaining a video set according to the video tag comprises:
calculating an interest score of the video tag, wherein the interest score is used for representing the interest degree of the user in a preset time period;
selecting a label with the interest score of the video label higher than a first threshold value;
and acquiring videos from the video library according to the selected tags to obtain a video set.
3. The method of claim 1, wherein obtaining a video set according to the video tag comprises:
and acquiring videos of which the similarity between the video tags corresponding to the videos in the video library and the video tags of the users is greater than a second threshold value to obtain a video set.
4. The method of claim 1, wherein obtaining a video set according to the video tag comprises:
calculating the association degree of the user and each user in the user set;
selecting the user with the maximum association degree with the user in the user set;
extracting the video tags in the historical records of the users with the maximum relevance;
and acquiring a video from a video library according to the video tag of the user and the video tag of the user with the maximum association degree to obtain a video set.
5. The method according to claim 4, wherein the obtaining a video from a video library according to the video tag of the user and the video tag of the user with the largest association degree to obtain a video set comprises:
acquiring videos from a video library according to the video tags of the users and the video tags of the users with the maximum relevance degree to obtain a plurality of videos;
and eliminating the videos of which the playing quantity is lower than a third threshold value from the plurality of videos to obtain a video set.
6. The method according to any one of claims 1-5, wherein the recommending videos in the video set to the user according to the playing duration comprises:
calculating the score of the video in the video set according to the playing time length;
sequencing the videos in the video set from high to low according to the corresponding scores to obtain a video list;
and displaying the videos to the user according to the sequence of the videos in the video list.
7. The method of claim 6, wherein said calculating scores for videos in the video set according to the playback time duration comprises:
calculating the watching proportion of the videos in the historical record of the user according to the playing time length and the total time length of the videos corresponding to the playing time length, and determining the score of the videos in the historical record according to the watching proportion;
determining a target video corresponding to a video tag of a video in the historical record in the video set;
and determining the score of the video in the historical record as the score of the target video.
8. The method of claim 1, wherein obtaining a video set according to the video tag comprises:
the method comprises the steps of obtaining a preset video library, wherein videos in the video library are obtained according to any one or more of video parameters, video playing amount and an uploading account number of the videos;
and acquiring a video set from a preset video library according to the video label.
9. The method of claim 8, further comprising:
extracting one or more music keywords from the title and/or content of a first video, wherein the first video is any video in the video library;
matching the one or more music keywords with vocabularies in a word bank;
and determining the corresponding music keywords of which the matching degree is greater than a fourth threshold value in the one or more music keywords as the video tags of the first video.
10. The method of claim 1, wherein the history record further comprises one or more audio playback information, the audio playback information comprising an audio tag; then, the obtaining a video set according to the video tag includes:
and acquiring a video set from a preset video library according to the video label and/or the audio label.
11. A video recommendation apparatus, comprising:
the device comprises a first acquisition unit, a second acquisition unit and a third acquisition unit, wherein the first acquisition unit is used for acquiring the history of a user after detecting a video acquisition request of the user; the history record comprises play information of one or more videos, wherein the play information comprises video tags and play duration, and the video tags are used for representing key information of the videos;
the second acquisition unit is used for acquiring a video set according to the video label;
and the recommending unit is used for recommending the videos in the video set to the user according to the playing time length.
12. A terminal device, comprising: a processor and a memory; the processor is coupled to a memory, wherein the memory is configured to store computer program code and the processor is configured to invoke the program code to perform the method of any of claims 1-10.
13. A computer storage medium, characterized in that the computer storage medium stores a computer program comprising program instructions which, when executed by a processor, perform the method according to any one of claims 1-10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010130419.4A CN111353068A (en) | 2020-02-28 | 2020-02-28 | Video recommendation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010130419.4A CN111353068A (en) | 2020-02-28 | 2020-02-28 | Video recommendation method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111353068A true CN111353068A (en) | 2020-06-30 |
Family
ID=71195918
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010130419.4A Pending CN111353068A (en) | 2020-02-28 | 2020-02-28 | Video recommendation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111353068A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111949807A (en) * | 2020-08-18 | 2020-11-17 | 腾讯科技（深圳）有限公司 | Data processing method, data processing device, computer equipment and storage medium |
| CN111966909A (en) * | 2020-08-26 | 2020-11-20 | 腾讯科技（深圳）有限公司 | Video recommendation method and device, electronic equipment and computer-readable storage medium |
| CN112153423A (en) * | 2020-09-15 | 2020-12-29 | 济南雪景网络技术有限公司 | User-self-driven short video intelligent recommendation method, electronic equipment and storage medium |
| CN112468851A (en) * | 2020-11-25 | 2021-03-09 | 深圳市易平方网络科技有限公司 | Video recommendation method and computer equipment |
| CN112507163A (en) * | 2020-12-02 | 2021-03-16 | 北京奇艺世纪科技有限公司 | Duration prediction model training method, recommendation method, device, equipment and 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 |
| CN113423014A (en) * | 2021-06-08 | 2021-09-21 | 深圳康佳电子科技有限公司 | Playing information pushing method and device, terminal equipment and storage medium |
| CN113507624A (en) * | 2021-09-10 | 2021-10-15 | 明品云(北京)数据科技有限公司 | Video information recommendation method and system |
| CN113873330A (en) * | 2021-08-31 | 2021-12-31 | 武汉卓尔数字传媒科技有限公司 | Video recommendation method and device, computer equipment and storage medium |
| CN114143612A (en) * | 2021-12-06 | 2022-03-04 | 北京达佳互联信息技术有限公司 | Video display method, video display device, electronic equipment, storage medium and program product |
| CN114419501A (en) * | 2022-01-11 | 2022-04-29 | 平安普惠企业管理有限公司 | Video recommendation method and device, computer equipment and storage medium |
| CN114528435A (en) * | 2020-11-23 | 2022-05-24 | 北京达佳互联信息技术有限公司 | Video sequencing method and device in search scene, electronic equipment and storage medium |
| CN117459798A (en) * | 2023-12-22 | 2024-01-26 | 厦门众联世纪股份有限公司 | Big data-based information display method, device, equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106407241A (en) * | 2016-03-21 | 2017-02-15 | 传线网络科技（上海）有限公司 | Video recommendation method and system |
| CN107846629A (en) * | 2017-10-11 | 2018-03-27 | 五八有限公司 | Recommend the method, apparatus and server of video to user |
| CN110059221A (en) * | 2019-03-11 | 2019-07-26 | 咪咕视讯科技有限公司 | Video recommendation method, electronic device and computer-readable storage medium |
| CN110493654A (en) * | 2019-08-20 | 2019-11-22 | 安徽抖范视频科技有限公司 | The recommendation of video and playback method and device in a kind of list of videos |
| CN110781391A (en) * | 2019-10-22 | 2020-02-11 | 腾讯科技（深圳）有限公司 | Information recommendation method, device, equipment and storage medium |
-
2020
- 2020-02-28 CN CN202010130419.4A patent/CN111353068A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106407241A (en) * | 2016-03-21 | 2017-02-15 | 传线网络科技（上海）有限公司 | Video recommendation method and system |
| CN107846629A (en) * | 2017-10-11 | 2018-03-27 | 五八有限公司 | Recommend the method, apparatus and server of video to user |
| CN110059221A (en) * | 2019-03-11 | 2019-07-26 | 咪咕视讯科技有限公司 | Video recommendation method, electronic device and computer-readable storage medium |
| CN110493654A (en) * | 2019-08-20 | 2019-11-22 | 安徽抖范视频科技有限公司 | The recommendation of video and playback method and device in a kind of list of videos |
| CN110781391A (en) * | 2019-10-22 | 2020-02-11 | 腾讯科技（深圳）有限公司 | Information recommendation method, device, equipment and storage medium |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111949807B (en) * | 2020-08-18 | 2024-06-25 | 腾讯科技（深圳）有限公司 | Data processing method, device, computer equipment and storage medium |
| CN111949807A (en) * | 2020-08-18 | 2020-11-17 | 腾讯科技（深圳）有限公司 | Data processing method, data processing device, computer equipment and storage medium |
| CN111966909A (en) * | 2020-08-26 | 2020-11-20 | 腾讯科技（深圳）有限公司 | Video recommendation method and device, electronic equipment and computer-readable storage medium |
| CN111966909B (en) * | 2020-08-26 | 2023-07-21 | 腾讯科技（深圳）有限公司 | Video recommendation method, device, electronic equipment and computer readable storage medium |
| CN112153423A (en) * | 2020-09-15 | 2020-12-29 | 济南雪景网络技术有限公司 | User-self-driven short video intelligent recommendation method, electronic equipment and storage medium |
| CN114528435A (en) * | 2020-11-23 | 2022-05-24 | 北京达佳互联信息技术有限公司 | Video sequencing method and device in search scene, electronic equipment and storage medium |
| CN112468851A (en) * | 2020-11-25 | 2021-03-09 | 深圳市易平方网络科技有限公司 | Video recommendation method and computer equipment |
| CN112507163A (en) * | 2020-12-02 | 2021-03-16 | 北京奇艺世纪科技有限公司 | Duration prediction model training method, recommendation method, device, equipment and 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 |
| CN113423014A (en) * | 2021-06-08 | 2021-09-21 | 深圳康佳电子科技有限公司 | Playing information pushing method and device, terminal equipment and storage medium |
| CN113873330A (en) * | 2021-08-31 | 2021-12-31 | 武汉卓尔数字传媒科技有限公司 | Video recommendation method and device, computer equipment and storage medium |
| CN113873330B (en) * | 2021-08-31 | 2023-03-10 | 武汉卓尔数字传媒科技有限公司 | Video recommendation method and device, computer equipment and storage medium |
| CN113507624A (en) * | 2021-09-10 | 2021-10-15 | 明品云(北京)数据科技有限公司 | Video information recommendation method and system |
| CN114143612A (en) * | 2021-12-06 | 2022-03-04 | 北京达佳互联信息技术有限公司 | Video display method, video display device, electronic equipment, storage medium and program product |
| CN114143612B (en) * | 2021-12-06 | 2024-03-15 | 北京达佳互联信息技术有限公司 | Video display method, device, electronic equipment, storage medium and program product |
| CN114419501A (en) * | 2022-01-11 | 2022-04-29 | 平安普惠企业管理有限公司 | Video recommendation method and device, computer equipment and storage medium |
| CN117459798A (en) * | 2023-12-22 | 2024-01-26 | 厦门众联世纪股份有限公司 | Big data-based information display method, device, equipment and storage medium |
| CN117459798B (en) * | 2023-12-22 | 2024-03-08 | 厦门众联世纪股份有限公司 | Big data-based information display method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111353068A (en) | Video recommendation method and device | |
| US9824150B2 (en) | Systems and methods for providing information discovery and retrieval | |
| WO2017096877A1 (en) | Recommendation method and device | |
| CN110134931B (en) | Medium title generation method, medium title generation device, electronic equipment and readable medium | |
| KR102249436B1 (en) | Contextualizing knowledge panels | |
| CN107209905A (en) | For personalized and task completion service, correspondence spends theme and sorted out | |
| CN101673186B (en) | Intelligent operating system and method based on keyword input | |
| CN109165302A (en) | Multimedia file recommendation method and device | |
| CN105335414B (en) | Music recommendation method and device and terminal | |
| US20200394213A1 (en) | Systems, Methods and Computer Program Products for Associating Media Content Having Different Modalities | |
| CN111708943B (en) | Search result display method and device for displaying search result | |
| US10083232B1 (en) | Weighting user feedback events based on device context | |
| CN110347866B (en) | Information processing method, information processing device, storage medium and electronic equipment | |
| CN109857901B (en) | Information display method and device, and method and device for information search | |
| CN105373580A (en) | Method and device for displaying subjects | |
| CN105550217B (en) | Scene music searching method and scene music searching device | |
| CN101763211A (en) | System for analyzing semanteme in real time and controlling related operation | |
| CN101984395A (en) | Intelligent operation system and method based on personal computer (PC) | |
| CN113626638A (en) | Short video recommendation processing method and device, intelligent terminal and storage medium | |
| CN113407775A (en) | Video searching method and device and electronic equipment | |
| JP5805134B2 (en) | Terminal device and device program | |
| CN110647504A (en) | Method and device for searching judicial documents | |
| CN108140034B (en) | Selecting content items based on received terms using a topic model | |
| CN112860921A (en) | Information searching method and device | |
| CN115834959A (en) | Video recommendation information determination method and device, electronic equipment and 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 |