CN108304441A - Network resource recommended method, device, electronic equipment, server and storage medium - Google Patents

Network resource recommended method, device, electronic equipment, server and storage medium Download PDF

Info

Publication number
CN108304441A
CN108304441A CN201711120052.2A CN201711120052A CN108304441A CN 108304441 A CN108304441 A CN 108304441A CN 201711120052 A CN201711120052 A CN 201711120052A CN 108304441 A CN108304441 A CN 108304441A
Authority
CN
China
Prior art keywords
information
internet resources
sample
target user
user
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.)
Granted
Application number
CN201711120052.2A
Other languages
Chinese (zh)
Other versions
CN108304441B (en
Inventor
邓亚平
连凤宗
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tencent Technology Shenzhen Co Ltd
Original Assignee
Tencent Technology Shenzhen Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Tencent Technology Shenzhen Co Ltd filed Critical Tencent Technology Shenzhen Co Ltd
Priority to CN201711120052.2A priority Critical patent/CN108304441B/en
Publication of CN108304441A publication Critical patent/CN108304441A/en
Application granted granted Critical
Publication of CN108304441B publication Critical patent/CN108304441B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
    • G06F16/43—Querying
    • G06F16/435—Filtering based on additional data, e.g. user or group profiles
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
    • G06F16/63—Querying
    • G06F16/635—Filtering based on additional data, e.g. user or group profiles
    • 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
    • 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

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Computational Linguistics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention discloses a kind of network resource recommended method, device, electronic equipment, server and storage mediums, belong to network technique field.The present invention is by using by network resource information, the resource recommendation model that the function usage behavior information of user and the history use information of user are trained, to carry out hobby prediction to user to be recommended and recommend, since the source of the sample data acquired in model training is more diversified, therefore, the resource recommendation model can describe the relationship between user and liked Internet resources from different perspectives, to accurately indicate that user likes degree to Internet resources, so that recommending the accuracy rate of its Internet resources liked higher to user, strong foundation is provided for commercial operation etc..

Description

Network resource recommended method, device, electronic equipment, server and storage medium
Technical field
The present invention relates to network technique field, more particularly to a kind of network resource recommended method, device, electronic equipment, clothes Business device and storage medium.
Background technology
With the development of network technology, the form of expression that networking products are capable of providing is more and more abundant.Networking products exist When providing service to the user, the online or offline clothes of such as Internet resources of picture, song, video diversified forms etc. can be provided Business.
For a user, the data volume that network resource server is provided is excessively huge, to from this mass data Oneself interested Internet resources is found, many times and flow can be expended, therefore, currently, network resource server can provide Network resource recommended function, by taking Internet resources are song, network resource server is song server as an example, song server can To play the song that played in song list according to the history of user, to find song similar with the song of history broadcasting song list Song, and recommend user.
However, the information of this recommendation method reference is single, the actual interest of user may not be able to be represented so that push away The accuracy rate recommended is very low.
Invention content
In order to solve problems in the prior art, an embodiment of the present invention provides a kind of network resource recommended method, device, electricity Sub- equipment, server and storage medium.The technical solution is as follows：
In a first aspect, a kind of network resource recommended method is provided, the method includes：
The function usage behavior information of target user to be recommended is obtained, the function usage behavior information is for indicating institute State usage behavior of the target user at least one function of network resource server；
According to the function usage behavior information of the target user, the preference profiles of the target user are obtained；
According to the preference profiles of the target user, Internet resources to be recommended and resource recommendation model, determine described in Preference of the target user to the Internet resources；
When the target user, which meets the preference of the Internet resources, presets recommendation condition, used to the target Recommend the Internet resources in family；
Wherein, the resource recommendation model is provided by the network resource server network resource information, multiple samples The function usage behavior information of this user and the history use information of the multiple sample of users train to obtain.
Second aspect, provides a kind of network resource recommended device, and described device includes：
First acquisition module, the function usage behavior information for obtaining target user to be recommended, the function use The usage behavior that behavioural information is used to indicate the target user at least one function of network resource server；
Second acquisition module obtains the target user for the function usage behavior information according to the target user Preference profiles；
Determining module, for preference profiles, the Internet resources and resource recommendation to be recommended according to the target user Model determines preference of the target user to the Internet resources；
Recommending module, for presetting recommendation condition when the target user meets the preference of the Internet resources When, recommend the Internet resources to the target user；
Wherein, the resource recommendation model is provided by the network resource server network resource information, multiple samples The function usage behavior information of this user and the history use information of the multiple sample of users train to obtain.
The third aspect, provides a kind of electronic equipment, and the electronic equipment includes processor and memory, the memory In be stored at least one instruction, at least one section of program, code set or instruction set, at least one instruction, described at least one Duan Chengxu, the code set or instruction set are loaded by the processor and are executed to realize the network money described in above-mentioned first aspect Recommend method in source.
Fourth aspect provides a kind of server, and the server includes processor and memory, is deposited in the memory Contain at least one instruction, at least one section of program, code set or instruction set, at least one instruction, at least one section of journey Sequence, the code set or instruction set are loaded by the processor and are executed to realize that the Internet resources described in above-mentioned first aspect push away Recommend method.
5th aspect, a kind of computer readable storage medium are stored at least one instruction, at least in the storage medium One section of program, code set or instruction set, at least one instruction, at least one section of program, the code set or the instruction set It is loaded by the processor and is executed to realize the network resource recommended method as described in above-mentioned first aspect.
The advantageous effect that technical solution provided in an embodiment of the present invention is brought is：
Make by using by the function usage behavior information of network resource information, target user and the history of target user The resource recommendation model trained with information, to carry out hobby prediction to target user to be recommended and recommend, due to model The source of the correlated characteristic used is more diversified, and therefore, which can describe user and institute from different perspectives The relationship between Internet resources is liked, to accurately indicate that user likes degree to Internet resources so that recommend to user The accuracy rate of its Internet resources liked is higher, and strong foundation is provided for commercial operation etc..
Description of the drawings
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for For those of ordinary skill in the art, without creative efforts, other are can also be obtained according to these attached drawings Attached drawing.
Fig. 1 is a kind of implementation environment schematic diagram of network resource recommended method provided in an embodiment of the present invention；
Fig. 2 is a kind of flow chart of network resource recommended method provided in an embodiment of the present invention；
Fig. 3 is a kind of network resource recommended method provided in an embodiment of the present invention by taking multimedia resource is song as an example Flow diagram；
Fig. 4 is a kind of network resource recommended interface schematic diagram provided in an embodiment of the present invention；
Fig. 5 is a kind of structural schematic diagram of network resource recommended device provided in an embodiment of the present invention；
Fig. 6 is the structure diagram of a kind of electronic equipment provided in an embodiment of the present invention；
Fig. 7 is a kind of structure diagram of server provided in an embodiment of the present invention.
Specific implementation mode
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached drawing to embodiment party of the present invention Formula is described in further detail.
Fig. 1 is a kind of implementation environment schematic diagram of network resource recommended method provided in an embodiment of the present invention.In the implementation Environment includes at least one terminal 101, network resource server 102.
The user of terminal 101 can access network resource server 102 by client, be taken thereby using Internet resources The network service that business device 102 is provided.For example, the terminal 101 can access Internet resources clothes by Video Applications client Business device 102, can also access the portal website of network resource server 102 by browser client.When target user is logical When crossing above method access client, these work(can be used by client use to functions such as Internet resources publication, search The information that can be generated, such as be exactly function usage behavior information to the record information that Internet resources search generates.And target user After Internet resources, the information of these Internet resources can be recorded, and form the history use information of the target user, example Such as, after certain target user plays certain song, the information of the song will be recorded to the history use letter of the target user In breath.
Wherein, Internet resources can be multimedia resource, and multimedia resource can be video, can also be audio.When more When media resource is video, video can be film, TV play, self-timer small video etc., based on multimedia resource information extraction Characteristics of the multimedia information can be video temperature, the languages of video and video type etc..Wherein, video temperature can be video Broadcasting time, video by like time, video by comment number etc., video type can be type of action, comedy type, Terrified type etc..
When multimedia resource is audio, audio can be song etc., the multimedia based on multimedia resource information extraction Characteristic information can be audio temperature, the language of audio and audio types etc..Wherein, audio temperature can be the broadcasting time of audio Number, audio by like time, audio are shared number etc., and audio types can be folk rhyme, electronic music, pop music etc..
For network resource server 102 for providing network service, which can refer to Video service, audio clothes Business, photo services, reading service and question and answer service etc..By taking network resource server is video server as an example, provided Video service may include the services such as net cast, video play online, video is downloaded, also, for net source service For device, the service provided can not be single service, for example, for video server, can be not limited only to Video service, but the other kinds of network service such as audio service is also provided, and for audio server, it can To be not limited only to audio service, the further types of network service such as Video service can also be provided, such as the service of K songs, audio Download etc..
Certainly, which can also provide the functions such as forwarding, comment, and the embodiment of the present invention, which does not do this, to be had Body limits.It can refer to converting certain film to video data stream, and video data stream is passed through that Online Video, which plays service, Videoconference client or portal website are supplied to terminal 101 operations such as to be played or downloaded offline online.
For network resource server 102, which can also have at least one data Library, to store the history use information of Internet resources characteristic information, sample of users and the function usage behavior of sample of users Information etc..
Wherein, Internet resources characteristic information may include the temperature of Internet resources, Internet resources language and Internet resources Type etc..For example, when Internet resources are multimedia resources, and when multimedia resource is video, characteristics of the multimedia information can be Video temperature, the languages of video and video type etc., when multimedia resource is audio, characteristics of the multimedia information can be audio Temperature, the language of audio and audio types etc..
Sample of users can refer to that the user of service, the history use information of sample of users are provided by network resource server May include that sample of users uses the information such as the number of Internet resources.
The function usage behavior information of sample of users may include Internet resources list information, Internet resources release news, Internet resources search for click information, the social sharing information of Internet resources, sample of users to the interactive informations etc. of Internet resources.
For different network resource types, the function usage behavior information of sample of users can be different, for example, working as net Network resource is multimedia resource, and when multimedia resource is video, under the function usage behavior information of multiple sample of users includes State at least one information：
1, user's registration information, such as user name, age, gender, the video type liked.
2, list of videos information, the list of videos created such as the video collection list information or sample of users of sample of users Information etc..
3, video upload information, such as sample of users upload oneself shooting small video.
4, video search click information passes through Video Applications client for indicating that sample of users is wanted to search for certain video Or browser client, when other relevant informations such as input video title or video author's name scan in search box, The information of the video recorded in search database of record.
5, the social networks sharing information of video, can be the number shared in Video Applications client, can also be Share other clients, such as the number of social networking application client.
6, to the interactive information of any video, for example, to the comment of certain video, thumb up.
It is more by taking multimedia resource is song as an example when Internet resources are multimedia resources, and multimedia resource is audio The function usage behavior information of a sample of users includes at least one of following information：
1, user's registration information, such as user name, age, gender, the music type liked.
2, song list information, for example, the song collection list information of sample of users, sample of users oneself create song list Information or the song list information etc. of sample of users collection.
3, K sings information, when indicating that sample of users sings function by music client end using K, to multimedia resource server The song information etc. of upload.
4, song search click information, for indicating that sample of users is wanted to search for certain song, by music client end or clear Device client of looking at searches for record when input other relevant informations such as song title or artist name scan in search box The information of the song recorded in database.
5, the social networks sharing information of song, can be the number for sharing certain song in music client end, also may be used To be to share other clients, such as the number of social networking application client.
6, to the interactive information of any song, for example, to the comment of certain song, thumb up.
It should be noted that the function usage behavior information of above-mentioned sample of users can needed by network resource server It is obtained from function usage behavior information database when carrying out model training.
Fig. 2 is a kind of flow chart of network resource recommended method provided in an embodiment of the present invention, referring to Fig. 2, shown in the Fig. 2 Flow specifically include two parts, first part is the history of function usage behavior information and sample of users based on sample of users The model training process of use information, second part are the prediction process based on recommended models.Passing through step 201 to 202 instructions After getting resource recommendation model, you can when being predicted directly use the resource recommendation model, as step 203 to 206, without being trained again or complicated calculations.In the present embodiment, by taking Internet resources are multimedia resources as an example, then method Executive agent is multimedia resource server, is described in detail to each step in conjunction with Fig. 2：
201, screening and feature extraction are carried out to sample initial data, obtains the characteristic information of multiple positive samples and multiple negative The characteristic information of sample, sample initial data include the history of the function usage behavior information and multiple sample of users of sample of users Use information.
Wherein, positive sample refers to the collection of multimedia resource and sample of users that stay time is greater than or equal to preset duration It closes, negative sample data refer to the set of multimedia resource and sample of users that stay time is less than the preset duration.For example, right For audio or video, which can refer to playing duration, and for web page resources etc., which can To be browsing duration.
It was recognized by the inventor that in conjunction with the function usage behavior information of user and the characteristics of the multimedia information of multimedia resource, When predicting the multimedia resource that user likes, due to reference information more comprehensively, and from multi-angular analysis user to more The hobby of media resource so as to the forecasting accuracy higher for liking degree of the multimedia resource of user.
Based on multimedia resource multimedia resource information provided by the server, multimedia resource server can extract more The history use information of media characteristic information, the function usage behavior information of user and user, and the above- mentioned information that will be extracted It is stored in the correspondence database of multimedia resource server, for subsequently carrying out feature extraction use.It is with multimedia resource For song, multimedia resource server based on the song information in Qu Ku, can extract song features information, user work( Can usage behavior information and user listen song historical information, and by the information storage extracted multimedia resource server pair It answers in database.
It before training pattern, needs to handle sample information, pretreated process includes：Screening sample process is commented Valence is with reference to preference process, positive and negative sample classification process, feature source screening process, characteristic extraction procedure, below to each Process describes in detail.
Screening sample process
Multimedia resource server, which gets the function usage behavior information of multiple sample of users to be screened and history, to be made After information, sample of users to be screened can be screened, after determining sample of users, then be directed to each sample and use Suitable multimedia resource characteristic information is chosen at family.
It, can be according to the function usage behavior of sample of users to be screened when being screened to sample of users to be screened Social networks sharing information in information, to the interactive information of multimedia resource, play multimedia resource and accumulate the progress such as duration Screening, pre-sets function usage behavior information threshold, in the social networks sharing information of sample of users to be screened, shares When the number of multimedia resource is more than function usage behavior information threshold or sample of users to be screened is to the mutual of multimedia resource In dynamic information, when being more than function usage behavior information threshold to the interactive number of multimedia resource, which is elected, And determine that it is sample of users.Multimedia resource threshold value is played alternatively, pre-setting, when sample of users to be screened plays more matchmakers Body resource time duration, which is more than, plays multimedia resource threshold value, which is elected, and determine that it is sample of users.This Sample Screening Samples user, can make the sample of users obtained after screening more representative so that the model that training obtains has more It is representative.
Evaluation reference preference process
After sample of users after being screened, since sample includes sample of users and multimedia resource, it is therefore desirable to more The multimedia resource evaluation reference preference that Media Resource Server provides.Multimedia resource server can be used according to sample The history stay time of multimedia resource is evaluated in the history use information at family, obtains the reference preference journey of multimedia resource Degree, this can be indicated with reference to preference with the form of score value.
By taking multimedia resource is song as an example, if setting the range of score value to [0,1], the song for needing to score first is determined Song, when then determining that the sample of users corresponds to the history stay time of the song by the history stay time divided by the whole song Long, obtained result is the score value that the song corresponds to the sample of users.Alternatively, it is also possible to be used according to the history of sample of users Information and function usage behavior information score jointly, such as according to above-mentioned processing mode, are obtained according to history use information One son scoring sings duration according to the K of the song and obtains a son scoring, calculate then according to above-mentioned similar processing mode The average value of two son scorings, the score value of the sample of users is corresponded to as the song.
During above-mentioned evaluation reference preference, the multimedia resource of evaluation reference preference can be original more Media resource, can also be the multimedia resource after screening, and the present embodiment does not limit herein.
Positive and negative sample classification process
After obtaining the reference preference of multimedia resource by above-mentioned evaluation reference preference process, according to multimedia Sample is divided into positive sample and negative sample by the reference preference of resource.When classifying to sample, multimedia resource Server can obtain the reference preference of multimedia resource in sample data, then according to pre-set sorting technique, Sample is divided into two classes, is positive sample and negative sample respectively.
Wherein, pre-set sorting technique can be indicated with reference to the form of preference score value, and score value is more than pre- If the sample of threshold value is positive sample, the sample that score value is less than or equal to predetermined threshold value is negative sample.For example, score value is more than 0.5 Sample becomes positive sample, and sample of the score value less than or equal to 0.5 becomes negative sample.
Feature source screening process
Because a data source of multimedia resource characteristic information is that song is single, thus can to the song list of sample of users into Row screening, can ensure that the sample of users elected and multimedia resource characteristic information are more representative, make training in this way Obtained model has more general applicability.When being screened to multimedia resource characteristic information, when multimedia resource feature When the source of information is that song is single, it can be screened according to the single temperature of song, it can also according to the singer of song in song list or specially It collects and is screened.
It, can when multimedia resource server is screened according to the single temperature of song by taking multimedia resource is song as an example With according to the single broadcasting time of song, by like time, shared number and screened, heat degree threshold is pre-set, when song is single When temperature is more than the heat degree threshold, then the song is singly screened, determine that it is a feature source.To be by like time Example, preset heat degree threshold are 100, when certain song it is single when being more than 100 by like time, which is singly screened, determines it For a feature source, other temperature types can be handled similarly, and this will not be repeated here.Screening multimedia resource feature letter in this way The source of breath can leave more representational data and provide feature, can prevent noise.
When multimedia resource server according to song list in song singer or album screen when, can according to song list in The singer of song or the same degree of album are screened, and same degree threshold value is pre-set.Singer to sing song in list is Example, preset same degree threshold value are 0.6, have 30 songs in list when singing, and wherein 20 songs are all by the same singer When performance, the numerical value for calculating same degree is 0.66, which is less than preset same degree threshold value, then can be by the song It singly screens, determines that it is a feature source.Screening song in this way is single, can be too single to avoid selected song list so that from When singing extraction song features information in list, the song features information extracted is inaccurate.
The detailed process of above-mentioned screening is only illustrated by taking song as an example, can also for other kinds of multimedia resource It similarly handles, this will not be repeated here.
Characteristic extraction procedure
It, can be according to the history use information and function usage behavior of sample of users after feature source data after being screened Information obtains the preference profiles of sample of users.Detailed process can be such as following step 1 to step 3：
Step 1：Multiple multimedias included by history use information and function usage behavior information to sample of users Resource information carry out vector characteristics extraction, obtain the user vector based on multimedia resource, the user vector based on label and The characteristic information of multimedia resource.
In force, by taking multimedia resource is song as an example, all song list information obtained after above-mentioned screening are determined, and All song identities in the single information of song, these song identities are extracted, using each song identity as a word, Then all song identities one sung in list form a sentence, and multiple song simple forms make these sentences at multiple sentences For language material, vectorial extraction algorithm, such as word2vector are then used, extracts its song vector.Then, according to sample of users Listen song historical information and song vector, the user vector based on song can be obtained.
Then, according to song vector and user vector based on song, the distance between above-mentioned vector can be calculated, it should be away from Degree is liked to the song from the sample of users can be used to refer to.
All label informations in all song list information obtained after above-mentioned screening, and the single information of song are determined, by these Label information extracts, and using each label as a word, all labels for then singing one in list form one Sentence, multiple song simple forms are at multiple sentences, using these sentences as language material, then use vectorial extraction algorithm, such as Word2vector etc. extracts its label vector.Then, song historical information and label vector are listened according to sample of users, can obtained To the user vector based on label.
Then, according to label vector and based on the user vector of label, the distance between vector is calculated, which can use To indicate that the sample of users likes degree to the songs of different labels.
Step 2：Collaborative filtering feature extraction is carried out to the history use information of sample of users, obtain sample of users based on The collaborative filtering feature of multimedia resource and the collaborative filtering feature based on sample of users.
In force, the process for obtaining the collaborative filtering feature based on song can be as follows：
According to the history use information of all sample of users, the similarity between each song is calculated, the similarity is bigger, and two The possibility that song is liked by the same sample of users is bigger.Acquisition is all in the history of a sample of users listens song information Then song is directed to a wherein song, in bent library in the range of all songs, the similar song with the song of generation one Similarity list, listens song list according to the song similarity list and sample of users, and a base is generated to the sample of users Each in the collaborative filtering feature vector of song, the collaborative filtering feature vector can represent the corresponding sample of the digit User likes the possibility of the song.
The process for obtaining the collaborative filtering feature based on sample of users can be as follows：Made according to the history of all sample of users With information, the similarity between each sample of users is calculated, two bigger sample of users of similarity, like same song can Energy property is bigger.Then it is directed to each sample of users, the list of one and the similarity of other sample of users are generated, according to this The song that may like with the similarity of other sample of users and other sample of users in similarity list, to the sample User generates a collaborative filtering feature vector based on sample of users, each in the collaborative filtering feature vector represents The sample of users likes the possibility of the corresponding song of the digit in list of songs.
Step 3：To carrying out sample of users feature according to the history use information and function usage behavior information of sample of users Extraction, obtains user's Figure Characteristics of sample of users.
When in the function usage behavior information of sample of users include user's registration information when, can to user's registration information into Row user characteristics extract, by the features such as the user name extracted, gender, age formed sample of users identity characteristic to Amount, the sample of users identity characteristic subvector can serve to indicate that the identity information of sample of users.
According to the history use information of sample of users and other function usage behavior information, in user's registration information The types of songs information etc. that sample of users is liked, the spies such as types of songs, singer, song language that extraction sample of users may like These features are formed a sample of users and like subvector by sign, and sample of users hobby subvector can serve to indicate that sample The song features that user likes.
Above-mentioned sample of users identity characteristic subvector and sample of users hobby subvector are synthesized a sample and used in order User's Figure Characteristics vector of user's Figure Characteristics vector at family, the sample of users can serve to indicate that the identity letter of sample of users The song features for ceasing and liking.
It should be noted that the information that can include specifically according to accessed function usage behavior information, to determine Include which step in above-mentioned steps 1 to step 3 on earth during the preference profiles of above-mentioned acquisition sample of users.For example, When the function usage behavior information got includes all information in function usage behavior information, that is, get user's registration letter Breath, song list information, K song information, song search click information, the social networks sharing information of song, to any song When interactive information, then above-mentioned work(step 1 is may be performed simultaneously to step 3.When the function usage behavior information got only includes Song list information and interactive information to any song, then can only execute above-mentioned steps 1.
202, the characteristic information of characteristic information, multiple negative samples based on multiple positive samples, the reference preference of positive negative sample Degree and the characteristics of the multimedia information of each multimedia resource carry out model training, obtain resource recommendation model.
By the reference preference of the characteristic information of multiple positive samples, the characteristic information of multiple negative samples, positive negative sample with And the characteristics of the multimedia information of each multimedia resource, it inputs together in resource recommendation model to be trained, step 202 can wrap Following steps 1 are included to step 2：
Step 1：By the reference preference of the characteristic information of multiple positive samples, the characteristic information of multiple negative samples, positive negative sample The current recommended models of characteristics of the multimedia information input of degree and each multimedia resource, output sample of users is to multiple samples The prediction preference of data.
In force, recommended models can be that FTRL (calculate by Follow the regularized Leader, logistic regression Method) on-line learning algorithm model, by the ginseng of the characteristic information of multiple positive samples, the characteristic information of multiple negative samples, positive negative sample The current recommended models of characteristics of the multimedia information input of preference and each multimedia resource are examined, in the characteristic information of sample Including the preference profiles of each sample of users and the characteristics of the multimedia information of multimedia resource and corresponding multimedia resource ginseng Examine preference, wherein the preference profiles of sample of users may include the calculated preference profiles of multiple vector forms.Pass through tax The initial weight of each preference profiles is given, prediction preference is then calculated, which can be expressed as the shape of score value The range of formula, the prediction score value can be [0,1], for indicating that current recommended models predict that the sample of users provides the multimedia Degree is liked in source, and prediction score value is higher, the sample of users to the multimedia resource of recommendation to like degree higher.
Step 2：According to the reference preference of prediction preference and sample data, adjust each in current recommended models The weight of feature obtains default recommended models.
In force, after the prediction preference for obtaining multiple model outputs, preference and input sample will each be predicted Multimedia resource is corresponding in this is compared with reference to preference, calculates error between the two.
According to obtained error, the weight of each feature in current recommended models is constantly adjusted, by finally obtained model Each feature weight of the weight as resource recommendation model, to obtain trained resource recommendation model.Trained money Source recommended models can accurately predict that user likes degree to multimedia resource, can improve and recommend to like more matchmakers to user The accuracy rate of body resource.
It should be noted that the above process is only an example of resource recommendation model training process, identical work is being played In the case of, the model and algorithm in the above process can be replaced by other models or algorithm, as matrix decomposition algorithm model, Deep learning algorithm model etc., this is not limited by the present invention.
It should be noted that above-mentioned model training process and prediction user are to the process for liking degree of multimedia resource Can be realized in the same equipment, for example, can be carried out by server, carry out model training in server side, and be based on The resource recommendation model that model training obtains is identified, alternatively, after can also carrying out model training by server side, will train Obtained resource recommendation model is issued to other electronic equipments, and is carried out based on the resource recommendation model by other electronic equipments Identification, the embodiment of the present invention are not limited specifically which electronic equipment to participate in the above process by.
Fig. 3 is a kind of network resource recommended method provided in an embodiment of the present invention by taking multimedia resource is song as an example Flow diagram.In conjunction with Fig. 3, by taking multimedia resource is song as an example, simplified summary is carried out to above-mentioned each step：
Multimedia resource server obtains bent library song information, the function of sample of users uses from corresponding database Behavioural information and sample of users listen song historical information, then, screened to sample of users.According to the sample of users after screening It is corresponding to listen song historical information, it scores song, obtains the score value of each song.Then, sample is divided into according to the score value Positive negative sample.Then, multimedia resource server listens song history to the function use information of sample of users and sample of users Information carries out feature extraction, obtains the preference profiles of sample of users.The preference profiles of sample of users include the user of feature based Vector, collaborative filtering feature and user's Figure Characteristics.The user vector of feature based includes user vector and base based on song In the user vector of label, collaborative filtering feature includes that the collaborative filtering feature based on song and the collaborative filtering based on user are special Sign, user's Figure Characteristics include user identity feature and user preferences feature.
Then, the characteristic information of song is obtained, the characteristic information of song includes song temperature, song language, types of songs. Finally, multimedia resource server is by the characteristic information of song obtained above, the preference profiles of sample of users and song Score value is input in model and is trained.After training model, so that it may to use model prediction user to like journey to song Degree.
Below to being illustrated to the process for liking degree of multimedia resource based on resource recommendation model prediction user.
203, the function usage behavior information of target user to be recommended is obtained.
In force, according to the function usage behavior information of target user to be recommended and multimedia resource information, so that it may To predict that favorable rating of the target user to multimedia resource, the history use information of target user are not essential information, this Sample can also pass through mesh if target user did not play multimedia resource on the client, or did not had history use information The function usage behavior information of mark user accurately predicts favorable rating of the target user to multimedia resource, and to target user The problem of being recommended, can not just be predicted to avoid no history use information and causing cold start-up.
204, according to the function usage behavior information of target user, the preference profiles of target user are obtained.
In force, after getting the function usage behavior information of target user, according to the processing side of above-mentioned steps 201 Formula carries out feature extraction to the function usage behavior information of target user and forms the characteristic information of vector form, obtain to The characteristic information of amount form constitutes the preference profiles of target user.
205, according to the preference profiles of target user, multimedia resource to be recommended and resource recommendation model, mesh is determined Mark preference of the user to multimedia resource.
It should be noted that if multimedia resource to be recommended is stored in advance in multimedia resource server, Then multimedia resource server can obtain multimedia resource characteristic information from characteristics of the multimedia information database；If waiting pushing away The multimedia resource recommended is not stored in advance in multimedia resource database, then multimedia resource server is to be recommended Multimedia resource carries out feature extraction, obtains multimedia resource characteristic information.
When needing to predict target user to multimedia resource when liking degree, can by multimedia resource characteristic information and The preference profiles of target user input resource recommendation model together.Can also be by multimedia resource to be recommended and target user Function usage behavior information handled together, multimedia resource to be recommended generate multimedia resource characteristic information, target The function usage behavior information of user generates the preference profiles of target user, then uses multimedia resource characteristic information and target The preference profiles at family input resource recommendation model.
By in multimedia resource characteristic information and the preference profiles of target user input resource recommendation model, target use is obtained To the preference of multimedia resource, which indicates that target user likes degree to song at family.The preference can Showed in the form of with score value, the score value is bigger, indicate target user to song to like degree bigger.
Wherein, resource recommendation model is used by multimedia resource multimedia resource information provided by the server, multiple samples The function usage behavior information at family and the history use information of multiple sample of users train to obtain.
206, when target user, which meets the preference of multimedia resource, presets recommendation condition, recommend to target user Multimedia resource.
The target user that above-mentioned steps 205 obtain compares the preference of multimedia resource with default recommendation condition Compared with if the default recommendation condition of preference satisfaction, the corresponding multimedia resource of the preference is determined as initially waiting for The multimedia resource of recommendation.Wherein, it presets the numerical value that recommendation condition can be preference and is more than default score value, default score value is The boundary value that technical staff chooses according to the score range of multimedia resource, for example, the preference of multimedia resource Numberical range is [0,1], then presets score value and could be provided as 0.5.
It should be noted that after determining initial multimedia resource to be recommended, need to judge initial multimedia to be recommended Whether resource meets default supplementary condition, presets supplementary condition if met, recommends to target user, otherwise not to mesh Mark user recommends.Default supplementary condition can be as follows：
Condition one：Multimedia resource is not included in the history use information of target user.Or, the history as target user makes When with including multimedia resource in information, and the residence time of multimedia resource and the time interval of current time are greater than or equal to Preset duration.
In force, it whether can inquire in the history use information of target user comprising initial multimedia money to be recommended The multimedia resource can be determined as multimedia resource to be recommended, recommend target user by source if not including；If including The residence time for then needing to inquire the multimedia resource, if the time interval of residence time and current time is greater than or equal to Preset duration, the i.e. multimedia resource were played in the recent period, then can the multimedia resource be recommended target user, Otherwise the multimedia resource is not recommended to target user.It can be pushed away in this way to avoid the multimedia resource for playing target user in the recent period It recommends to target user, avoids that success rate is recommended thus to be lowered, actually increase recommendation success rate.
Condition two：The scoring of multimedia resource is more than default scoring.
Wherein, the scoring of multimedia resource refers in relevant client, and all target users are to the multimedia resource Scoring, such as scoring etc. of the target user to certain film in applications client or browser client.
After multimedia resource server inquires the scoring of multimedia resource to be recommended, by multimedia resource to be recommended Scoring is compared with pre-set scoring, if the scoring of multimedia resource is more than default scoring, can be used to target The multimedia resource is recommended at family, does not otherwise recommend the multimedia resource to target user.It can be pushed away in this way to avoid to target user The multimedia resource that difference is commented is recommended, keeps the experience of target user more preferable.Wherein, it is that technical staff is tied based on multiple experiments to preset scoring The scoring that fruit is chosen.
Condition three：There is different multimedia resource titles from fixed multimedia resource to be recommended.
Multimedia resource to be recommended is subjected to duplicate removal, such as in song to be recommended, there are the three of a certain song A version, three versions are sung by three different singers, in this case, select song scoring highest in three versions Version, be determined as song to be recommended, other two version is not recommended then to target user, in this way can be to avoid to pushing away Chance is recommended to cause to waste.
It should be noted that above-mentioned condition can be used alone, can also two kinds of conditional combinations use, can also three kinds of items Part is applied in combination.For example, condition one can be applied in combination with condition two, i.e., multimedia is not included in the history use information of user Resource.Or, when in the history use information of target user include multimedia resource when, and the residence time of multimedia resource with work as The time interval of preceding time is greater than or equal to preset duration；And the scoring of multimedia resource is more than default scoring.In such case Under, the processing of this condition can be handled according to above-mentioned condition one and condition two, get rid of ineligible multimedia money Remaining multimedia resource is recommended target user by source.
It should be noted that after multimedia resource to be recommended is determined according to above-mentioned condition, can to target user into Row is recommended, and when recommendation can be handled as follows：The preference of multimedia resource is waited for fixed according to target user Recommend the sorting position in multimedia resource, includes in page location corresponding with sorting position by multimedia resource.
Multimedia resource to be recommended is ranked up its preference according to target user, it then will be to be recommended more Media resource is shown in the corresponding page location of clooating sequence of the multimedia resource.For example, as shown in figure 4, certain user opens When certain music client end, it can be seen that client recommends 10 songs recommended on the page, wherein singer's picture of 3 songs It can recycle and show on page top, the display in " song recommendations today " of the following page of 7 songs.Client is determined to this After the song that user recommends, song is ranked up its preference according to target user, the number of selection wherein preference It is worth highest preceding 10 song, highest preceding 3 song of the numerical value is recycled on customer terminal webpage top in order and is shown, will be remained Under 7 songs be shown in order in " song recommendations today ".The user can see at first cycle display 3 it is first he most may be used Can interested song, in browsing pages it can also be seen that shown in " song recommendations today " he may be very interested Song.For another example certain user open certain music client end want search for certain song when, open searched page, user is in search box When the keyword of search is wanted in middle input, search box can be shown and the relevant recommended song of keyword, Yi Jiyu below The relevant recommended song of the user.User can be made to easily find that the multimedia oneself liked provides according to recommendation in this way Source so that the accuracy rate of recommendation improves, and improves the Experience Degree of user.
The present embodiment can also have other processing modes, such as according to target user to more matchmakers by taking above-mentioned processing mode as an example The preference of body resource, the sorting position in fixed multimedia resource to be recommended, by multimedia resource include with The corresponding display batch of sorting position is medium, and this will not be repeated here.
For the present embodiment by taking above-mentioned multimedia resource as an example, Internet resources can also be other resources, such as news information, shopping Information, such as goods links.By taking news information as an example, resource recommendation model is trained based on news information, obtained money Source recommended models can be used for recommending news information, other Internet resources that can also carry out model training similarly to target user And recommendation, this is not limited by the present invention.
Method provided in an embodiment of the present invention, by using the function usage behavior information by network resource information, user And the resource recommendation model that the history use information of user is trained, to carry out hobby prediction to user to be recommended and push away It recommends, since the source of the sample data acquired in model training is more diversified, which can be from Different angle describes the relationship between user and liked Internet resources, to accurately indicate that user likes Internet resources Degree so that recommend the accuracy rate of its Internet resources liked higher to user, strong foundation is provided for commercial operation etc..
Fig. 5 is a kind of structural schematic diagram of network resource recommended device provided in an embodiment of the present invention.It is described referring to Fig. 5 Device includes：
First acquisition module 501, the function usage behavior information for obtaining target user to be recommended, the function make The usage behavior for being used to indicate the target user at least one function of network resource server with behavioural information；
Second acquisition module 502 obtains the target and uses for the function usage behavior information according to the target user The preference profiles at family；
Determining module 503, for being pushed away according to the preference profiles, Internet resources to be recommended and resource of the target user Model is recommended, determines preference of the target user to the Internet resources；
Recommending module 504, for meeting default recommendation item to the preference of the Internet resources as the target user When part, recommend the Internet resources to the target user；
Wherein, the default recommended models are provided by the network resource server network resource information, multiple samples The function usage behavior information of this user and the history use information of the multiple sample of users train to obtain.
In any possible realization method, the function usage behavior information of multiple sample of users includes at least one of following Information：It is user's registration information, network list information, network issued information, search click information, social networks sharing information, right The interactive information of any Internet resources.
In any possible realization method, described device further includes：
Processing module obtains the feature letter of multiple positive samples for carrying out screening and feature extraction to sample initial data The characteristic information of breath and multiple negative samples, the sample initial data include the function usage behavior information of sample of users and described The history use information of multiple sample of users；
Resource recommendation model training module, the feature letter for characteristic information, multiple negative samples based on multiple positive samples The Internet resources characteristic information of breath, the reference preference of positive negative sample and each Internet resources carries out model training, Obtain the resource recommendation model；
Wherein, positive sample refers to the set of Internet resources and sample of users that stay time is greater than or equal to preset duration, Negative sample data refer to the set of Internet resources and sample of users that stay time is less than the preset duration.
In any possible realization method, the resource recommendation model training module is used for：
The multiple sample data is inputted into current recommended models, exports the prediction preference journey of the multiple sample data Degree；
According to the reference preference of the prediction preference and the sample data, the current recommended models are adjusted In each feature weight, obtain the default recommended models.
In any possible realization method, second acquisition module is used for：
Multiple networks included by history use information and the function usage behavior information to the target user Resource information carries out vector characteristics extraction, obtains the user vector based on Internet resources, user vector and net based on label The characteristic information of network resource；
Collaborative filtering feature extraction is carried out to the history use information of the target user, obtains the base of the target user In the collaborative filtering feature of Internet resources and collaborative filtering feature based on user；
It is special that user is carried out to the history use information according to the target user and the function usage behavior information Sign extraction, obtains user's Figure Characteristics of the target user.
In any possible realization method, the recommending module is used for：
When the target user meets default recommendation condition to the preference of the Internet resources and meets multinomial default At least one of in supplementary condition, recommend the Internet resources to the target user；
Wherein, the multinomial default supplementary condition includes：
The Internet resources are not included in the history use information of the target user；Or, going through as the target user When in history use information including the Internet resources, and the residence time of the Internet resources and the time interval of current time are big In or equal to preset duration；
And/or
The scoring of the Internet resources is more than default scoring；
And/or
There is different Internet resources titles from fixed Internet resources to be recommended.
In any possible realization method, the recommending module is used for：
According to the target user to the preference of the Internet resources in fixed Internet resources to be recommended The Internet resources are included in page location corresponding with the sorting position by sorting position.
Method provided in an embodiment of the present invention, by using the function usage behavior information by network resource information, user And the resource recommendation model that the history use information of user is trained, to carry out hobby prediction to user to be recommended and push away It recommends, since the source of the sample data acquired in model training is more diversified, which can be from Different angle describes the relationship between user and liked Internet resources, to accurately indicate that user likes Internet resources Degree so that recommend the accuracy rate of its Internet resources liked higher to user, strong foundation is provided for commercial operation etc..
Fig. 6 is the structure diagram of a kind of electronic equipment provided in an embodiment of the present invention.The electronic equipment 600 can be portable Formula mobile electronic device, such as：Smart mobile phone, tablet computer, MP3 player (Moving Picture Experts Group Audio Layer III, dynamic image expert's compression standard audio level 3), MP4 (Moving Picture Experts Group Audio Layer IV, dynamic image expert's compression standard audio level 4) player.Electronic equipment 600 be also possible to by Referred to as other titles such as user equipment, portable electronic device.
In general, electronic equipment 600 includes：Processor 601 and memory 602.
Processor 601 may include one or more processing cores, such as 4 core processors, 8 core processors etc..Place DSP (Digital Signal Processing, Digital Signal Processing), FPGA (Field- may be used in reason device 601 Programmable Gate Array, field programmable gate array), PLA (Programmable Logic Array, may be programmed Logic array) at least one of example, in hardware realize.Processor 601 can also include primary processor and coprocessor, master Processor is the processor for being handled data in the awake state, also referred to as CPU (Central Processing Unit, central processing unit)；Coprocessor is the low power processor for being handled data in the standby state. In some embodiments, processor 601 can be integrated with GPU (Graphics Processing Unit, image processor), GPU is used to be responsible for the rendering and drafting of content to be shown needed for display screen.In some embodiments, processor 601 can also wrap AI (Artificial Intelligence, artificial intelligence) processor is included, the AI processors are for handling related machine learning Calculating operation.
Memory 602 may include one or more computer readable storage mediums, which can To be tangible and non-transient.Memory 602 may also include high-speed random access memory and nonvolatile memory, Such as one or more disk storage equipments, flash memory device.In some embodiments, non-transient in memory 602 Computer readable storage medium for storing at least one instruction, at least one instruction for by processor 601 it is performed with Realize network resource recommended method provided herein.
In some embodiments, electronic equipment 600 is also optional includes：Peripheral device interface 603 and at least one periphery Equipment.Specifically, peripheral equipment includes：Radio circuit 604, touch display screen 605, camera 606, voicefrequency circuit 607, positioning At least one of component 608 and power supply 609.
Peripheral device interface 603 can be used for I/O (Input/Output, input/output) is relevant at least one outer Peripheral equipment is connected to processor 601 and memory 602.In some embodiments, processor 601, memory 602 and peripheral equipment Interface 603 is integrated on same chip or circuit board；In some other embodiments, processor 601, memory 602 and outer Any one or two in peripheral equipment interface 603 can realize on individual chip or circuit board, the present embodiment to this not It is limited.
Radio circuit 604 is for receiving and emitting RF (Radio Frequency, radio frequency) signal, also referred to as electromagnetic signal.It penetrates Frequency circuit 604 is communicated by electromagnetic signal with communication network and other communication equipments.Radio circuit 604 turns electric signal It is changed to electromagnetic signal to be sent, alternatively, the electromagnetic signal received is converted to electric signal.Optionally, radio circuit 604 wraps It includes：Antenna system, RF transceivers, one or more amplifiers, tuner, oscillator, digital signal processor, codec chip Group, user identity module card etc..Radio circuit 604 can by least one wireless communication protocol come with other electronic equipments It is communicated.The wireless communication protocol includes but not limited to：WWW, Metropolitan Area Network (MAN), Intranet, each third generation mobile communication network (2G, 3G, 4G and 5G), WLAN and/or WiFi (Wireless Fidelity, Wireless Fidelity) network.In some embodiments In, radio circuit 604 can also include the related electricity of NFC (Near Field Communication, wireless near field communication) Road, the application are not limited this.
Touch display screen 605 is for showing UI (User Interface, user interface).The UI may include figure, text Sheet, icon, video and its their arbitrary combination.Touch display screen 605 also have acquisition on the surface of touch display screen 605 or The ability of the touch signal of surface.The touch signal can be used as control signal to be input to processor 601 and be handled.It touches Display screen 605 is touched for providing virtual push button and/or dummy keyboard, also referred to as soft button and/or soft keyboard.In some embodiments In, touch display screen 605 can be one, and the front panel of electronic equipment 600 is arranged；In further embodiments, display is touched Screen 605 can be at least two, be separately positioned on the different surfaces of electronic equipment 600 or in foldover design；In still other implementation Example in, touch display screen 605 can be flexible display screen, be arranged on the curved surface of electronic equipment 600 or fold plane on.Very Extremely, touch display screen 605 can also be arranged to non-rectangle irregular figure, namely abnormity screen.Touch display screen 605 can adopt With LCD (Liquid Crystal Display, liquid crystal display), (Organic Light-Emitting Diode, have OLED Machine light emitting diode) etc. materials prepare.
CCD camera assembly 606 is for acquiring image or video.Optionally, CCD camera assembly 606 include front camera and Rear camera.In general, front camera is for realizing video calling or self-timer, rear camera is for realizing photo or video Shooting.In some embodiments, rear camera at least two are main camera, depth of field camera, wide-angle imaging respectively Any one in head, to realize that main camera and the fusion of depth of field camera realize background blurring function, main camera and wide-angle Pan-shot and VR (Virtual Reality, virtual reality) shooting function are realized in camera fusion.In some embodiments In, CCD camera assembly 606 can also include flash lamp.Flash lamp can be monochromatic warm flash lamp, can also be double-colored temperature flash of light Lamp.Double-colored temperature flash lamp refers to the combination of warm light flash lamp and cold light flash lamp, can be used for the light compensation under different-colour.
Voicefrequency circuit 607 is used to provide the audio interface between user and electronic equipment 600.Voicefrequency circuit 607 can wrap Include microphone and loud speaker.Microphone is used to acquire the sound wave of user and environment, and converts sound waves into electric signal and be input to place Reason device 601 is handled, or is input to radio circuit 604 to realize voice communication.For stereo acquisition or the mesh of noise reduction , microphone can be multiple, be separately positioned on the different parts of electronic equipment 600.Microphone can also be array microphone Or omnidirectional's acquisition type microphone.Loud speaker is then used to the electric signal from processor 601 or radio circuit 604 being converted to sound Wave.Loud speaker can be traditional wafer speaker, can also be piezoelectric ceramic loudspeaker.When loud speaker is that piezoelectric ceramics is raised one's voice When device, the audible sound wave of the mankind can be not only converted electrical signals to, can also convert electrical signals to what the mankind did not heard Sound wave is to carry out the purposes such as ranging.In some embodiments, voicefrequency circuit 607 can also include earphone jack.
Positioning component 608 is used for the current geographic position of Positioning Electronic Devices 600, to realize navigation or LBS (Location Based Service, location based service).Positioning component 608 can be the GPS (Global based on the U.S. Positioning System, global positioning system), China dipper system or Russia Galileo system positioning group Part.
Power supply 609 is used to be powered for the various components in electronic equipment 600.Power supply 609 can be alternating current, direct current Electricity, disposable battery or rechargeable battery.When power supply 609 includes rechargeable battery, which can have micro USB Battery or wireless charging battery.Wired charging battery is the battery to be charged by Wireline, and wireless charging battery is to pass through The battery of wireless coil charging.The rechargeable battery can be also used for supporting fast charge technology.
In some embodiments, electronic equipment 600 further include there are one or multiple sensors 610.The one or more passes Sensor 610 includes but not limited to：Acceleration transducer 611, gyro sensor 612, pressure sensor 613, fingerprint sensor 614, optical sensor 615 and proximity sensor 616.
Acceleration transducer 611 can detect the acceleration in three reference axis of the coordinate system established with electronic equipment 600 Spend size.For example, acceleration transducer 611 can be used for detecting component of the acceleration of gravity in three reference axis.Processor The 601 acceleration of gravity signals that can be acquired according to acceleration transducer 611, control touch display screen 605 with transverse views or Longitudinal view carries out the display of user interface.Acceleration transducer 611 can be also used for game or the exercise data of user Acquisition.
Gyro sensor 612 can detect body direction and the rotational angle of electronic equipment 600, gyro sensor 612 can cooperate with acquisition user to act the 3D of electronic equipment 600 with acceleration transducer 611.Processor 601 is according to gyroscope The data that sensor 612 acquires, may be implemented following function：Action induction (for example changed according to the tilt operation of user Image stabilization, game control when UI), shooting and inertial navigation.
The lower layer of side frame and/or touch display screen 605 in electronic equipment 600 can be arranged in pressure sensor 613.When The gripping signal that user can be detected in the side frame of electronic equipment 600 to electronic equipment 600 is arranged in pressure sensor 613, Right-hand man's identification or prompt operation are carried out according to the gripping signal.When pressure sensor 613 is arranged under touch display screen 605 , can be according to user to the pressure operation of touch display screen 605 when layer, the operability control on the interfaces UI is controlled in realization System.Operability control includes at least one of button control, scroll bar control, icon control, menu control.
Fingerprint sensor 614 is used to acquire the fingerprint of user, with according to the identity of collected fingerprint recognition user.Knowing When the identity for not going out user is trusted identity, the user is authorized to execute relevant sensitive operation, the sensitive operation by processor 601 Including solution lock screen, check encryption information, download software, payment and change setting etc..Electricity can be set in fingerprint sensor 614 Front, the back side or the side of sub- equipment 600.When being provided with physical button or manufacturer Logo on electronic equipment 600, fingerprint sensing Device 614 can be integrated with physical button or manufacturer Logo.
Optical sensor 615 is for acquiring ambient light intensity.In one embodiment, processor 601 can be according to optics The ambient light intensity that sensor 615 acquires controls the display brightness of touch display screen 605.Specifically, when ambient light intensity is higher When, the display brightness of touch display screen 605 is turned up；When ambient light intensity is relatively low, the display for turning down touch display screen 605 is bright Degree.In another embodiment, the ambient light intensity that processor 601 can also be acquired according to optical sensor 615, dynamic adjust The acquisition parameters of CCD camera assembly 606.
Proximity sensor 616, also referred to as range sensor are generally arranged at the front of electronic equipment 600.Proximity sensor 616 the distance between the front for acquiring user and electronic equipment 600.In one embodiment, when proximity sensor 616 is examined When measuring the distance between the front of user and electronic equipment 600 and tapering into, touch display screen 605 is controlled by processor 601 It is switched to breath screen state from bright screen state；When proximity sensor 616 detect between user and the front of electronic equipment 600 away from When from becoming larger, touch display screen 605 being controlled by processor 601 and is switched to bright screen state from breath screen state.
It will be understood by those skilled in the art that structure shown in Fig. 6 does not constitute the restriction to electronic equipment 600, it can To include either combining certain components than illustrating more or fewer components or being arranged using different components.
Fig. 7 is a kind of structure diagram of server provided in an embodiment of the present invention.With reference to Fig. 7, server 700 includes processing Component 722 further comprises one or more processors, and by the memory resource representated by memory 732, for depositing Storage can be by the instruction of the execution of processing component 722, such as application program.The application program stored in memory 732 may include It is one or more each correspond to one group of instruction module.In addition, processing component 722 is configured as executing instruction, With the network resource recommended method for executing above-mentioned Fig. 2,3 illustrated embodiments provide.
Server 700 can also include that a power supply module 726 be configured as the power management of execute server 700, and one A wired or wireless network interface 750 is configured as server 700 being connected to network and input and output (I/O) interface 758.Server 700 can be operated based on the operating system for being stored in memory 732, such as Windows ServerTM, Mac OS XTM, UnixTM,LinuxTM, FreeBSDTMOr it is similar.
In the exemplary embodiment, a kind of computer readable storage medium is additionally provided, is stored at least in storage medium One instruction, at least one section of program, code set or instruction set, at least one instruction, at least one section of program, code set or instruction set It is loaded by processor and is executed to realize the network resource recommended method in above-described embodiment.It computer-readable is deposited for example, described Storage media can be ROM, random access memory (RAM), CD-ROM, tape, floppy disk and optical data storage devices etc..
One of ordinary skill in the art will appreciate that realizing that all or part of step of above-described embodiment can pass through hardware It completes, relevant hardware can also be instructed to complete by program, the program can be stored in a kind of computer-readable In storage medium, storage medium mentioned above can be read-only memory, disk or CD etc..
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all the present invention spirit and Within principle, any modification, equivalent replacement, improvement and so on should all be included in the protection scope of the present invention.

Claims (15)

1. a kind of network resource recommended method, which is characterized in that the method includes：
The function usage behavior information of target user is obtained, the function usage behavior information is for indicating the target user couple The usage behavior of at least one function of network resource server；
According to the function usage behavior information of the target user, the preference profiles of the target user are obtained；
According to the preference profiles of the target user, Internet resources to be recommended and resource recommendation model, the target is determined Preference of the user to the Internet resources；
When the target user, which meets the preference of the Internet resources, presets recommendation condition, pushed away to the target user Recommend the Internet resources；
Wherein, the resource recommendation model is provided by the network resource server network resource information, multiple samples are used The function usage behavior information at family and the history use information of the multiple sample of users train to obtain.
2. according to the method described in claim 1, it is characterized in that, under the function usage behavior information of multiple sample of users includes State at least one information：
User's registration information, Internet resources list information, Internet resources release news, search for click information, social networks is shared Information, to the interactive information of any Internet resources.
3. according to the method described in claim 1, it is characterized in that, for example following mistakes of the training method of the resource recommendation model Journey：
Screening and feature extraction are carried out to sample initial data, obtain the spy of the characteristic information and multiple negative samples of multiple positive samples Reference ceases, and the sample initial data includes the history of the function usage behavior information and the multiple sample of users of sample of users Use information；
The characteristic information of characteristic information, multiple negative samples, the reference preference of positive negative sample based on multiple positive samples and The Internet resources characteristic information of each Internet resources carries out model training, obtains the resource recommendation model；
Wherein, positive sample refers to the set of Internet resources and sample of users that stay time is greater than or equal to preset duration, bears sample Notebook data refers to the set of Internet resources and sample of users that stay time is less than the preset duration.
4. according to the method described in claim 3, it is characterized in that, described based on the multiple positive sample data, the multiple The Internet resources characteristic information of negative sample data and each Internet resources carries out model training, obtains the resource recommendation Model includes：
The multiple sample data is inputted into current recommended models, exports the sample of users to the pre- of the multiple sample data Survey preference；
According to the reference preference of the prediction preference and the sample data, adjust in the current recommended models The weight of each feature obtains the resource recommendation model.
5. according to the method described in claim 1, it is characterized in that, the history use information according to the target user and The function usage behavior information, the preference profiles for obtaining the target user include：
Multiple Internet resources included by history use information and the function usage behavior information to the target user Information carries out vector characteristics extraction, obtains the user vector based on Internet resources, the user vector based on label and network money The characteristic information in source；
Collaborative filtering feature extraction is carried out to the history use information of the target user, obtain the target user based on net The collaborative filtering feature of network resource and the collaborative filtering feature based on user；
History use information and the function usage behavior information to the target user carry out user characteristics extraction, obtain institute State user's Figure Characteristics of target user.
6. according to the method described in claim 1, it is characterized in that, described when the target user is to the inclined of the Internet resources When good degree meets default recommendation condition, include to the target user recommendation Internet resources：
When the target user meets default recommendation condition to the preference of the Internet resources and meets multinomial default supplement At least one of in condition, recommend the Internet resources to the target user；
Wherein, the multinomial default supplementary condition includes：
The Internet resources are not included in the history use information of the target user；Or, the history as the target user makes With in information include the Internet resources when, and the residence time of the Internet resources and the time interval of current time be more than or Equal to preset duration；
And/or
The scoring of the Internet resources is more than default scoring；
And/or
There is different Internet resources titles from fixed Internet resources to be recommended.
7. according to the method described in claim 1, it is characterized in that, described recommend the Internet resources packet to the target user It includes：
Sequence according to the target user to the preference of the Internet resources in fixed Internet resources to be recommended The Internet resources are included in page location corresponding with the sorting position by position.
8. a kind of network resource recommended device, which is characterized in that described device includes：
First acquisition module, the function usage behavior information for obtaining target user to be recommended, the function usage behavior The usage behavior that information is used to indicate the target user at least one function of network resource server；
Second acquisition module obtains the inclined of the target user for the function usage behavior information according to the target user Good feature；
Determining module is used for preference profiles, Internet resources to be recommended and resource recommendation model according to the target user, Determine preference of the target user to the Internet resources；
Recommending module is used for when the target user meets the preference of the Internet resources and presets recommendation condition, to The target user recommends the Internet resources；
Wherein, the resource recommendation model is provided by the network resource server network resource information, multiple samples are used The function usage behavior information at family and the history use information of the multiple sample of users train to obtain.
9. device according to claim 8, which is characterized in that described device further includes：
Processing module, for sample initial data carry out screening and feature extraction, obtain multiple positive samples characteristic information and The characteristic information of multiple negative samples, the sample initial data include the function usage behavior information of sample of users and the multiple The history use information of sample of users；
Resource recommendation model training module, the characteristic information, just for characteristic information, multiple negative samples based on multiple positive samples The Internet resources characteristic information of the reference preference of negative sample and each Internet resources carries out model training, obtains institute State resource recommendation model；
Wherein, positive sample refers to the set of Internet resources and sample of users that stay time is greater than or equal to preset duration, bears sample Notebook data refers to the set of Internet resources and sample of users that stay time is less than the preset duration.
10. device according to claim 9, which is characterized in that the resource recommendation model training module is used for：
The multiple sample data is inputted into current recommended models, exports the sample of users to the pre- of the multiple sample data Survey preference；
According to the reference preference of the prediction preference and the sample data, adjust in the current recommended models The weight of each feature obtains the resource recommendation model.
11. device according to claim 8, which is characterized in that second acquisition module is used for：
Multiple Internet resources included by history use information and the function usage behavior information to the target user Information carries out vector characteristics extraction, obtains the user vector based on Internet resources, the user vector based on label and network money The characteristic information in source；
Collaborative filtering feature extraction is carried out to the history use information of the target user, obtain the target user based on net The collaborative filtering feature of network resource and the collaborative filtering feature based on user；
User characteristics are carried out to the history use information according to the target user and the function usage behavior information to carry It takes, obtains user's Figure Characteristics of the target user.
12. device according to claim 8, which is characterized in that the recommending module is used for：
When the target user meets default recommendation condition to the preference of the Internet resources and meets multinomial default supplement At least one of in condition, recommend the Internet resources to the target user；
Wherein, the multinomial default supplementary condition includes：
The Internet resources are not included in the history use information of the target user；Or, the history as the target user makes With in information include the Internet resources when, and the residence time of the Internet resources and the time interval of current time be more than or Equal to preset duration；
And/or
The scoring of the Internet resources is more than default scoring；
And/or
There is different Internet resources titles from fixed Internet resources to be recommended.
13. a kind of electronic equipment, which is characterized in that the electronic equipment includes processor and memory, is deposited in the memory Contain at least one instruction, at least one section of program, code set or instruction set, at least one instruction, at least one section of journey Sequence, the code set or instruction set are loaded by the processor and are executed to realize the network as described in claim 1 to 7 is any Resource recommendation method.
14. a kind of server, which is characterized in that the server includes processor and memory, is stored in the memory At least one instruction, at least one section of program, code set or instruction set, at least one instruction, at least one section of program, the institute Code set or instruction set is stated to be loaded by the processor and executed to realize the Internet resources as described in claim 1 to 7 is any Recommendation method.
15. a kind of computer readable storage medium, which is characterized in that be stored at least one instruction, extremely in the storage medium Few one section of program, code set or instruction set, at least one instruction, at least one section of program, the code set or the instruction Collection is loaded by the processor and is executed to realize the network resource recommended method as described in claim 1 to 7 is any.
CN201711120052.2A 2017-11-14 2017-11-14 Network resource recommendation method and device, electronic equipment, server and storage medium Active CN108304441B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711120052.2A CN108304441B (en) 2017-11-14 2017-11-14 Network resource recommendation method and device, electronic equipment, server and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711120052.2A CN108304441B (en) 2017-11-14 2017-11-14 Network resource recommendation method and device, electronic equipment, server and storage medium

Publications (2)

Publication Number Publication Date
CN108304441A true CN108304441A (en) 2018-07-20
CN108304441B CN108304441B (en) 2022-04-26

Family

ID=62869590

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711120052.2A Active CN108304441B (en) 2017-11-14 2017-11-14 Network resource recommendation method and device, electronic equipment, server and storage medium

Country Status (1)

Country Link
CN (1) CN108304441B (en)

Cited By (65)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109271550A (en) * 2018-07-27 2019-01-25 华南理工大学 A kind of music personalization classification recommended method based on deep learning
CN109284445A (en) * 2018-12-11 2019-01-29 北京达佳互联信息技术有限公司 Recommended method, device, server and the storage medium of Internet resources
CN109299314A (en) * 2018-11-13 2019-02-01 百度在线网络技术（北京）有限公司 Music retrieval and the method, apparatus of recommendation, storage medium and terminal device
CN109344314A (en) * 2018-08-20 2019-02-15 腾讯科技（深圳）有限公司 A kind of data processing method, device and server
CN109408714A (en) * 2018-10-15 2019-03-01 上海基分文化传播有限公司 A kind of recommender system and method for multi-model fusion
CN109408724A (en) * 2018-11-06 2019-03-01 北京达佳互联信息技术有限公司 Multimedia resource estimates the determination method, apparatus and server of clicking rate
CN109447622A (en) * 2018-09-30 2019-03-08 中国银行股份有限公司 Type of transaction recommended method and system, intelligent Trade terminal
CN109508426A (en) * 2018-12-21 2019-03-22 深圳市智搜信息技术有限公司 A kind of intelligent recommendation method and its system and storage medium based on physical environment
CN109547814A (en) * 2018-12-13 2019-03-29 北京达佳互联信息技术有限公司 Video recommendation method, device, server and storage medium
CN109697258A (en) * 2018-12-27 2019-04-30 丹翰智能科技(上海)有限公司 It is a kind of for determining the method and apparatus of the customization financial information of target user
CN109740056A (en) * 2018-12-28 2019-05-10 丹翰智能科技(上海)有限公司 It is a kind of for provide a user customization financial information method and apparatus
CN109885719A (en) * 2019-01-24 2019-06-14 广州小鹏汽车科技有限公司 A kind of song recommendations method, system, terminal and storage medium
CN109994000A (en) * 2019-03-28 2019-07-09 掌阅科技股份有限公司 A kind of reading partner method, electronic equipment and computer storage medium
CN110134820A (en) * 2019-04-26 2019-08-16 湖南大学 A kind of mixed type individualized music recommended method being incremented by based on feature
CN110147882A (en) * 2018-09-03 2019-08-20 腾讯科技（深圳）有限公司 Training method, crowd's method of diffusion, device and the equipment of neural network model
CN110287418A (en) * 2019-06-26 2019-09-27 拉扎斯网络科技（上海）有限公司 Data processing and tendency characteristic value recognition methods, device, electronic equipment
CN110297939A (en) * 2019-06-21 2019-10-01 山东科技大学 A kind of music personalization system of fusion user behavior and cultural metadata
CN110297848A (en) * 2019-07-09 2019-10-01 深圳前海微众银行股份有限公司 Recommended models training method, terminal and storage medium based on federation's study
CN110413837A (en) * 2019-05-30 2019-11-05 腾讯科技（深圳）有限公司 Video recommendation method and device
CN110442790A (en) * 2019-08-07 2019-11-12 腾讯科技（深圳）有限公司 Recommend method, apparatus, server and the storage medium of multi-medium data
CN110704739A (en) * 2019-09-30 2020-01-17 汉海信息技术(上海)有限公司 Resource recommendation method and device and computer storage medium
CN110796505A (en) * 2018-08-03 2020-02-14 阿里巴巴集团控股有限公司 Service object recommendation method and device
CN110830812A (en) * 2019-10-31 2020-02-21 广州市网星信息技术有限公司 Similar anchor classification model training method, anchor recommendation method and related device
CN110851706A (en) * 2019-10-10 2020-02-28 百度在线网络技术（北京）有限公司 Training method and device for user click model, electronic equipment and storage medium
CN110866180A (en) * 2019-10-12 2020-03-06 平安国际智慧城市科技股份有限公司 Resource recommendation method, server and storage medium
CN110874639A (en) * 2018-08-31 2020-03-10 珠海格力电器股份有限公司 Method and device for acquiring operation information
CN110874767A (en) * 2019-11-19 2020-03-10 山东爱城市网信息技术有限公司 Business hall accurate marketing method based on collaborative filtering
CN110889020A (en) * 2019-11-22 2020-03-17 百度在线网络技术（北京）有限公司 Site resource mining method and device and electronic equipment
CN110910209A (en) * 2019-11-12 2020-03-24 腾讯科技（深圳）有限公司 Data processing method and device and computer readable storage medium
CN111159435A (en) * 2019-12-27 2020-05-15 北大方正集团有限公司 Multimedia resource processing method, system, terminal and computer readable storage medium
CN111274472A (en) * 2018-12-04 2020-06-12 北京嘀嘀无限科技发展有限公司 Information recommendation method and device, server and readable storage medium
CN111556366A (en) * 2020-04-02 2020-08-18 北京达佳互联信息技术有限公司 Multimedia resource display method, device, terminal, server and system
CN111625713A (en) * 2020-04-30 2020-09-04 平安国际智慧城市科技股份有限公司 Resource recommendation method and device based on big data, electronic equipment and medium
CN111737573A (en) * 2020-06-17 2020-10-02 北京三快在线科技有限公司 Resource recommendation method, device, equipment and storage medium
CN111753182A (en) * 2019-03-28 2020-10-09 北京达佳互联信息技术有限公司 Multimedia information recommendation method and device, electronic equipment and readable storage medium
CN111753124A (en) * 2019-03-29 2020-10-09 Tcl集团股份有限公司 Music recommendation method and server
CN111814051A (en) * 2020-07-08 2020-10-23 北京奇艺世纪科技有限公司 Resource type determination method and device
CN111858970A (en) * 2019-04-30 2020-10-30 北京达佳互联信息技术有限公司 Multimedia content recommendation method and device, electronic equipment and readable storage medium
CN111898018A (en) * 2019-05-06 2020-11-06 北京达佳互联信息技术有限公司 Virtual resource sending method and device, electronic equipment and storage medium
CN112000819A (en) * 2019-05-27 2020-11-27 北京达佳互联信息技术有限公司 Multimedia resource recommendation method and device, electronic equipment and storage medium
CN112148964A (en) * 2019-06-29 2020-12-29 阿里巴巴集团控股有限公司 Information processing and recommending method, system and equipment
CN112182460A (en) * 2019-07-04 2021-01-05 腾讯科技（北京）有限公司 Resource pushing method and device, storage medium and electronic device
CN112199022A (en) * 2020-10-23 2021-01-08 Oppo广东移动通信有限公司 Content display method and device, electronic equipment and storage medium
CN112200639A (en) * 2020-10-30 2021-01-08 杭州时趣信息技术有限公司 Information flow model construction method, device and medium
CN112241327A (en) * 2019-07-17 2021-01-19 腾讯科技（深圳）有限公司 Shared information processing method and device, storage medium and electronic 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
CN112765470A (en) * 2021-01-25 2021-05-07 腾讯科技（深圳）有限公司 Training method of content recommendation model, content recommendation method, device and equipment
CN112804566A (en) * 2019-11-14 2021-05-14 中兴通讯股份有限公司 Program recommendation method, device and computer readable storage medium
CN113205183A (en) * 2021-04-23 2021-08-03 北京达佳互联信息技术有限公司 Article recommendation network training method and device, electronic equipment and storage medium
CN113344647A (en) * 2021-07-14 2021-09-03 杭州网易云音乐科技有限公司 Information recommendation method and device
CN113365115A (en) * 2020-03-03 2021-09-07 北京达佳互联信息技术有限公司 Characteristic code determining method, device, server and storage medium
CN113382289A (en) * 2021-08-11 2021-09-10 北京达佳互联信息技术有限公司 Live broadcast room delivery method and device, electronic equipment and storage medium
CN113495985A (en) * 2020-03-18 2021-10-12 北京达佳互联信息技术有限公司 Article recommendation method and device, short video recommendation method and device and server
CN113535991A (en) * 2020-04-14 2021-10-22 北京达佳互联信息技术有限公司 Multimedia resource recommendation method and device, electronic equipment and storage medium
CN113539457A (en) * 2021-07-16 2021-10-22 挂号网（杭州）科技有限公司 Medical resource recommendation method and device, electronic equipment and storage medium
CN113722745A (en) * 2021-07-13 2021-11-30 苏州柯里特信息科技有限公司 eCTD universal technology document submission management method and system
CN113760521A (en) * 2020-09-22 2021-12-07 北京沃东天骏信息技术有限公司 Virtual resource allocation method and device
CN113836390A (en) * 2020-06-24 2021-12-24 北京达佳互联信息技术有限公司 Resource recommendation method and device, computer equipment and storage medium
CN114071119A (en) * 2020-07-31 2022-02-18 北京达佳互联信息技术有限公司 Resource testing method and device, electronic equipment and storage medium
CN114077701A (en) * 2020-08-13 2022-02-22 北京达佳互联信息技术有限公司 Method and device for determining resource information, computer equipment and storage medium
CN114779999A (en) * 2022-04-27 2022-07-22 北京达佳互联信息技术有限公司 Information display method and device, electronic equipment and storage medium
CN115079908A (en) * 2021-03-12 2022-09-20 北京字节跳动网络技术有限公司 Media content collection method and device, electronic equipment and computer readable storage medium
WO2023121631A1 (en) * 2021-12-22 2023-06-29 Turkcell Teknoloji Arastirma Ve Gelistirme Anonim Sirketi A system for creating personalized song
CN113722588B (en) * 2021-08-12 2023-09-05 北京达佳互联信息技术有限公司 Resource recommendation method and device and electronic equipment
CN110910209B (en) * 2019-11-12 2024-05-31 腾讯科技（深圳）有限公司 Data processing method, device and computer readable storage medium

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102402625A (en) * 2011-12-28 2012-04-04 深圳市五巨科技有限公司 Method and system for recommending music
CN102523511A (en) * 2011-11-09 2012-06-27 中国传媒大学 Network program aggregation and recommendation system and network program aggregation and recommendation method
CN104021233A (en) * 2014-06-30 2014-09-03 电子科技大学 Social network friend recommendation method based on community discovery
CN105868254A (en) * 2015-12-25 2016-08-17 乐视网信息技术（北京）股份有限公司 Information recommendation method and apparatus
CN106874522A (en) * 2017-03-29 2017-06-20 珠海习悦信息技术有限公司 Information recommendation method, device, storage medium and processor

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102523511A (en) * 2011-11-09 2012-06-27 中国传媒大学 Network program aggregation and recommendation system and network program aggregation and recommendation method
CN102402625A (en) * 2011-12-28 2012-04-04 深圳市五巨科技有限公司 Method and system for recommending music
CN104021233A (en) * 2014-06-30 2014-09-03 电子科技大学 Social network friend recommendation method based on community discovery
CN105868254A (en) * 2015-12-25 2016-08-17 乐视网信息技术（北京）股份有限公司 Information recommendation method and apparatus
CN106874522A (en) * 2017-03-29 2017-06-20 珠海习悦信息技术有限公司 Information recommendation method, device, storage medium and processor

Cited By (97)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109271550B (en) * 2018-07-27 2022-05-24 华南理工大学 Music personalized recommendation method based on deep learning
CN109271550A (en) * 2018-07-27 2019-01-25 华南理工大学 A kind of music personalization classification recommended method based on deep learning
CN110796505A (en) * 2018-08-03 2020-02-14 阿里巴巴集团控股有限公司 Service object recommendation method and device
CN110796505B (en) * 2018-08-03 2023-07-04 淘宝（中国）软件有限公司 Business object recommendation method and device
CN109344314B (en) * 2018-08-20 2021-11-16 腾讯科技（深圳）有限公司 Data processing method and device and server
CN109344314A (en) * 2018-08-20 2019-02-15 腾讯科技（深圳）有限公司 A kind of data processing method, device and server
CN110874639A (en) * 2018-08-31 2020-03-10 珠海格力电器股份有限公司 Method and device for acquiring operation information
CN110147882A (en) * 2018-09-03 2019-08-20 腾讯科技（深圳）有限公司 Training method, crowd's method of diffusion, device and the equipment of neural network model
CN110147882B (en) * 2018-09-03 2023-02-10 腾讯科技（深圳）有限公司 Neural network model training method, crowd diffusion method, device and equipment
CN109447622A (en) * 2018-09-30 2019-03-08 中国银行股份有限公司 Type of transaction recommended method and system, intelligent Trade terminal
CN109447622B (en) * 2018-09-30 2022-02-08 中国银行股份有限公司 Transaction type recommendation method and system and intelligent transaction terminal
CN109408714B (en) * 2018-10-15 2021-02-05 上海趣蕴网络科技有限公司 Multi-model fusion recommendation system and method
CN109408714A (en) * 2018-10-15 2019-03-01 上海基分文化传播有限公司 A kind of recommender system and method for multi-model fusion
CN109408724A (en) * 2018-11-06 2019-03-01 北京达佳互联信息技术有限公司 Multimedia resource estimates the determination method, apparatus and server of clicking rate
CN109299314A (en) * 2018-11-13 2019-02-01 百度在线网络技术（北京）有限公司 Music retrieval and the method, apparatus of recommendation, storage medium and terminal device
CN109299314B (en) * 2018-11-13 2019-12-27 百度在线网络技术（北京）有限公司 Music retrieval and recommendation method, device, storage medium and terminal equipment
CN111274472A (en) * 2018-12-04 2020-06-12 北京嘀嘀无限科技发展有限公司 Information recommendation method and device, server and readable storage medium
CN109284445A (en) * 2018-12-11 2019-01-29 北京达佳互联信息技术有限公司 Recommended method, device, server and the storage medium of Internet resources
CN109547814A (en) * 2018-12-13 2019-03-29 北京达佳互联信息技术有限公司 Video recommendation method, device, server and storage medium
CN109508426A (en) * 2018-12-21 2019-03-22 深圳市智搜信息技术有限公司 A kind of intelligent recommendation method and its system and storage medium based on physical environment
CN109697258A (en) * 2018-12-27 2019-04-30 丹翰智能科技(上海)有限公司 It is a kind of for determining the method and apparatus of the customization financial information of target user
CN109740056A (en) * 2018-12-28 2019-05-10 丹翰智能科技(上海)有限公司 It is a kind of for provide a user customization financial information method and apparatus
CN109885719A (en) * 2019-01-24 2019-06-14 广州小鹏汽车科技有限公司 A kind of song recommendations method, system, terminal and storage medium
CN111753182B (en) * 2019-03-28 2024-05-17 北京达佳互联信息技术有限公司 Multimedia information recommendation method and device, electronic equipment and readable storage medium
CN111753182A (en) * 2019-03-28 2020-10-09 北京达佳互联信息技术有限公司 Multimedia information recommendation method and device, electronic equipment and readable storage medium
CN109994000B (en) * 2019-03-28 2021-10-19 掌阅科技股份有限公司 Reading accompanying method, electronic equipment and computer storage medium
CN109994000A (en) * 2019-03-28 2019-07-09 掌阅科技股份有限公司 A kind of reading partner method, electronic equipment and computer storage medium
CN111753124A (en) * 2019-03-29 2020-10-09 Tcl集团股份有限公司 Music recommendation method and server
CN110134820A (en) * 2019-04-26 2019-08-16 湖南大学 A kind of mixed type individualized music recommended method being incremented by based on feature
CN111858970B (en) * 2019-04-30 2024-01-02 北京达佳互联信息技术有限公司 Multimedia content recommendation method and device, electronic equipment and readable storage medium
CN111858970A (en) * 2019-04-30 2020-10-30 北京达佳互联信息技术有限公司 Multimedia content recommendation method and device, electronic equipment and readable storage medium
CN111898018B (en) * 2019-05-06 2024-05-14 北京达佳互联信息技术有限公司 Virtual resource sending method and device, electronic equipment and storage medium
CN111898018A (en) * 2019-05-06 2020-11-06 北京达佳互联信息技术有限公司 Virtual resource sending method and device, electronic equipment and storage medium
CN112000819B (en) * 2019-05-27 2023-07-11 北京达佳互联信息技术有限公司 Multimedia resource recommendation method and device, electronic equipment and storage medium
CN112000819A (en) * 2019-05-27 2020-11-27 北京达佳互联信息技术有限公司 Multimedia resource recommendation method and device, electronic equipment and storage medium
CN110413837B (en) * 2019-05-30 2023-07-25 腾讯科技（深圳）有限公司 Video recommendation method and device
CN110413837A (en) * 2019-05-30 2019-11-05 腾讯科技（深圳）有限公司 Video recommendation method and device
CN110297939A (en) * 2019-06-21 2019-10-01 山东科技大学 A kind of music personalization system of fusion user behavior and cultural metadata
CN110287418A (en) * 2019-06-26 2019-09-27 拉扎斯网络科技（上海）有限公司 Data processing and tendency characteristic value recognition methods, device, electronic equipment
CN110287418B (en) * 2019-06-26 2021-10-19 拉扎斯网络科技（上海）有限公司 Data processing and tendency characteristic value identification method and device, and electronic equipment
CN112148964A (en) * 2019-06-29 2020-12-29 阿里巴巴集团控股有限公司 Information processing and recommending method, system and equipment
CN112182460A (en) * 2019-07-04 2021-01-05 腾讯科技（北京）有限公司 Resource pushing method and device, storage medium and electronic device
CN112182460B (en) * 2019-07-04 2024-03-15 深圳市雅阅科技有限公司 Resource pushing method and device, storage medium and electronic device
CN110297848A (en) * 2019-07-09 2019-10-01 深圳前海微众银行股份有限公司 Recommended models training method, terminal and storage medium based on federation's study
CN110297848B (en) * 2019-07-09 2024-02-23 深圳前海微众银行股份有限公司 Recommendation model training method, terminal and storage medium based on federal learning
CN112241327A (en) * 2019-07-17 2021-01-19 腾讯科技（深圳）有限公司 Shared information processing method and device, storage medium and electronic equipment
CN110442790B (en) * 2019-08-07 2024-05-10 深圳市雅阅科技有限公司 Method, device, server and storage medium for recommending multimedia data
CN110442790A (en) * 2019-08-07 2019-11-12 腾讯科技（深圳）有限公司 Recommend method, apparatus, server and the storage medium of multi-medium data
CN110704739A (en) * 2019-09-30 2020-01-17 汉海信息技术(上海)有限公司 Resource recommendation method and device and computer storage medium
CN110851706A (en) * 2019-10-10 2020-02-28 百度在线网络技术（北京）有限公司 Training method and device for user click model, electronic equipment and storage medium
US11838377B2 (en) 2019-10-10 2023-12-05 Baidu Online Network Technology (Beijing) Co., Ltd. Method, apparatus, electronic device and storage medium for training user click model
CN110866180B (en) * 2019-10-12 2022-07-29 平安国际智慧城市科技股份有限公司 Resource recommendation method, server and storage medium
CN110866180A (en) * 2019-10-12 2020-03-06 平安国际智慧城市科技股份有限公司 Resource recommendation method, server and storage medium
CN110830812A (en) * 2019-10-31 2020-02-21 广州市网星信息技术有限公司 Similar anchor classification model training method, anchor recommendation method and related device
CN110830812B (en) * 2019-10-31 2021-11-30 广州市网星信息技术有限公司 Similar anchor classification model training method, anchor recommendation method and related device
CN110910209B (en) * 2019-11-12 2024-05-31 腾讯科技（深圳）有限公司 Data processing method, device and computer readable storage medium
CN110910209A (en) * 2019-11-12 2020-03-24 腾讯科技（深圳）有限公司 Data processing method and device and computer readable storage medium
CN112804566A (en) * 2019-11-14 2021-05-14 中兴通讯股份有限公司 Program recommendation method, device and computer readable storage medium
CN110874767A (en) * 2019-11-19 2020-03-10 山东爱城市网信息技术有限公司 Business hall accurate marketing method based on collaborative filtering
CN110889020B (en) * 2019-11-22 2022-08-23 百度在线网络技术（北京）有限公司 Site resource mining method and device and electronic equipment
CN110889020A (en) * 2019-11-22 2020-03-17 百度在线网络技术（北京）有限公司 Site resource mining method and device and electronic equipment
CN111159435A (en) * 2019-12-27 2020-05-15 北大方正集团有限公司 Multimedia resource processing method, system, terminal and computer readable storage medium
CN111159435B (en) * 2019-12-27 2023-09-05 新方正控股发展有限责任公司 Multimedia resource processing method, system, terminal and computer readable storage medium
CN113365115A (en) * 2020-03-03 2021-09-07 北京达佳互联信息技术有限公司 Characteristic code determining method, device, server and storage medium
CN113365115B (en) * 2020-03-03 2022-11-04 北京达佳互联信息技术有限公司 Characteristic code determining method, device, server and storage medium
CN113495985B (en) * 2020-03-18 2024-03-19 北京达佳互联信息技术有限公司 Article recommendation method and device, short video recommendation method and device and server
CN113495985A (en) * 2020-03-18 2021-10-12 北京达佳互联信息技术有限公司 Article recommendation method and device, short video recommendation method and device and server
CN111556366A (en) * 2020-04-02 2020-08-18 北京达佳互联信息技术有限公司 Multimedia resource display method, device, terminal, server and system
CN113535991B (en) * 2020-04-14 2024-01-26 北京达佳互联信息技术有限公司 Multimedia resource recommendation method and device, electronic equipment and storage medium
CN113535991A (en) * 2020-04-14 2021-10-22 北京达佳互联信息技术有限公司 Multimedia resource recommendation method and device, electronic equipment and storage medium
CN111625713A (en) * 2020-04-30 2020-09-04 平安国际智慧城市科技股份有限公司 Resource recommendation method and device based on big data, electronic equipment and medium
CN111737573A (en) * 2020-06-17 2020-10-02 北京三快在线科技有限公司 Resource recommendation method, device, equipment and storage medium
CN113836390B (en) * 2020-06-24 2023-10-27 北京达佳互联信息技术有限公司 Resource recommendation method, device, computer equipment and storage medium
CN113836390A (en) * 2020-06-24 2021-12-24 北京达佳互联信息技术有限公司 Resource recommendation method and device, computer equipment and storage medium
CN111814051A (en) * 2020-07-08 2020-10-23 北京奇艺世纪科技有限公司 Resource type determination method and device
CN111814051B (en) * 2020-07-08 2023-07-28 北京奇艺世纪科技有限公司 Resource type determining method and device
CN114071119B (en) * 2020-07-31 2024-03-19 北京达佳互联信息技术有限公司 Resource testing method and device, electronic equipment and storage medium
CN114071119A (en) * 2020-07-31 2022-02-18 北京达佳互联信息技术有限公司 Resource testing method and device, electronic equipment and storage medium
CN114077701A (en) * 2020-08-13 2022-02-22 北京达佳互联信息技术有限公司 Method and device for determining resource information, computer equipment and storage medium
CN113760521A (en) * 2020-09-22 2021-12-07 北京沃东天骏信息技术有限公司 Virtual resource allocation method and device
CN112199022A (en) * 2020-10-23 2021-01-08 Oppo广东移动通信有限公司 Content display method and device, electronic equipment and storage medium
CN112200639A (en) * 2020-10-30 2021-01-08 杭州时趣信息技术有限公司 Information flow model construction method, device and medium
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
CN112765470A (en) * 2021-01-25 2021-05-07 腾讯科技（深圳）有限公司 Training method of content recommendation model, content recommendation method, device and equipment
CN112765470B (en) * 2021-01-25 2022-02-22 腾讯科技（深圳）有限公司 Training method of content recommendation model, content recommendation method, device and equipment
CN115079908A (en) * 2021-03-12 2022-09-20 北京字节跳动网络技术有限公司 Media content collection method and device, electronic equipment and computer readable storage medium
CN113205183B (en) * 2021-04-23 2024-05-14 北京达佳互联信息技术有限公司 Article recommendation network training method and device, electronic equipment and storage medium
CN113205183A (en) * 2021-04-23 2021-08-03 北京达佳互联信息技术有限公司 Article recommendation network training method and device, electronic equipment and storage medium
CN113722745B (en) * 2021-07-13 2024-02-06 苏州柯里特信息科技有限公司 eCTD general technical document submitting management method and system
CN113722745A (en) * 2021-07-13 2021-11-30 苏州柯里特信息科技有限公司 eCTD universal technology document submission management method and system
CN113344647A (en) * 2021-07-14 2021-09-03 杭州网易云音乐科技有限公司 Information recommendation method and device
CN113539457A (en) * 2021-07-16 2021-10-22 挂号网（杭州）科技有限公司 Medical resource recommendation method and device, electronic equipment and storage medium
CN113382289A (en) * 2021-08-11 2021-09-10 北京达佳互联信息技术有限公司 Live broadcast room delivery method and device, electronic equipment and storage medium
CN113382289B (en) * 2021-08-11 2021-11-02 北京达佳互联信息技术有限公司 Live broadcast room delivery method and device, electronic equipment and storage medium
CN113722588B (en) * 2021-08-12 2023-09-05 北京达佳互联信息技术有限公司 Resource recommendation method and device and electronic equipment
WO2023121631A1 (en) * 2021-12-22 2023-06-29 Turkcell Teknoloji Arastirma Ve Gelistirme Anonim Sirketi A system for creating personalized song
CN114779999A (en) * 2022-04-27 2022-07-22 北京达佳互联信息技术有限公司 Information display method and device, electronic equipment and storage medium

Also Published As

Publication number Publication date
CN108304441B (en) 2022-04-26

Similar Documents

Publication Publication Date Title
CN108304441A (en) Network resource recommended method, device, electronic equipment, server and storage medium
CN109740068B (en) Media data recommendation method, device and storage medium
CN110110203B (en) Resource information pushing method, server, resource information display method and terminal
WO2020215962A1 (en) Video recommendation method and device, computer device and storage medium
CN108197327B (en) Song recommendation method, device and storage medium
EP3757995A1 (en) Music recommendation method and apparatus, and computing device and medium
CN109918669B (en) Entity determining method, device and storage medium
CN110020140A (en) Recommendation display methods, apparatus and system
CN110199276A (en) With aspect tag query
CN109189879B (en) Electronic book display method and device
CN106355429A (en) Image material recommendation method and device
CN109657236B (en) Guidance information acquisition method, apparatus, electronic apparatus, and storage medium
CN110458360B (en) Method, device, equipment and storage medium for predicting hot resources
CN105701254A (en) Information processing method and device and device for processing information
CN111737573A (en) Resource recommendation method, device, equipment and storage medium
CN112163428A (en) Semantic tag acquisition method and device, node equipment and storage medium
CN107562939A (en) Vertical field news recommends method, apparatus and readable storage medium
CN109784351A (en) Data classification method, disaggregated model training method and device
CN111897996A (en) Topic label recommendation method, device, equipment and storage medium
CN111291200A (en) Multimedia resource display method and device, computer equipment and storage medium
CN110246110A (en) Image evaluation method, device and storage medium
CN111831917A (en) Content recommendation method, device, equipment and medium
CN112116391A (en) Multimedia resource delivery method and device, computer equipment and storage medium
CN111428522B (en) Translation corpus generation method, device, computer equipment and storage medium
CN107622074A (en) A kind of data processing method, device and computing device

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
GR01 Patent grant
GR01 Patent grant