CN105930429A - Music recommendation method and apparatus - Google Patents
Music recommendation method and apparatus Download PDFInfo
- Publication number
- CN105930429A CN105930429A CN201610245225.2A CN201610245225A CN105930429A CN 105930429 A CN105930429 A CN 105930429A CN 201610245225 A CN201610245225 A CN 201610245225A CN 105930429 A CN105930429 A CN 105930429A
- Authority
- CN
- China
- Prior art keywords
- music
- user
- model
- data
- candidate
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/60—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
- G06F16/637—Administration of user profiles, e.g. generation, initialization, adaptation or distribution
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Reverberation, Karaoke And Other Acoustics (AREA)
Abstract
Embodiments of the invention provide a music recommendation method and apparatus. The music recommendation method comprises the steps of detecting an event of initiating a search request by a user, and obtaining environment information according to a scene of initiating the request by the user; extracting request parameters of the user according to the search request, wherein the request parameters include a user model, the environment information and a tag of music content in the search request; screening out candidate music from a database according to the tag of the music content; and computing matching scores of the user and the candidate music in combination with the music model according to the user model, sorting the candidate music according to the matching scores, and recommending the sorted candidate music to the user. Correspondingly, the embodiment of the invention furthermore provides the music recommendation apparatus. The music recommendation apparatus comprises a detection module, an extraction module, a screening module and a recommendation module. According to the technical scheme provided by the embodiments of the invention, accurate recommendation services matched with user preferences are provided for the user, so that the personalized demands of the user can be met and the user experience is improved.
Description
Technical field
The present embodiments relate to technical field of mobile terminals, particularly relate to a kind of method that music is recommended
And device.
Background technology
Growing along with mobile terminal technology, the function that mobile terminal is had is more and more abundanter,
Including comprehensive search function, comprehensive search has very important ground for mobile devices such as mobile phones
Position, cell phone manufacturer all can provide for their mobile device and search for entrance easily, such as, pass through voice
The mode comprehensive search of input includes the polytypes such as the Internet, map, mail, contact person, App
Data, owing to listening music to be one of most important application of the mobile devices such as mobile phone, therefore music is also
One of most important search content, but way of search of the prior art is owing to being comprehensive search, that
Music result may possibly be mixed among other results many, and is inconvenient to find.
Attention rate currently for search convenience promotes, when user is desirable in scanning for day by day
The Search Results coming foremost is exactly just oneself content to be looked for.Due to the abundant in content of music and
Wide variety, and the grade of music user has diversified difference, recommends music time institute carrying out
The content recommended is not often that user is wanted, so current music way of recommendation Consumer's Experience
Spend the lowest.
To sum up, in view of technological deficiency of the prior art, in a kind of search entrance on the terminal device
Integrated music is recommended, it is possible to accurately recommend the technical scheme of the content that user is to be searched for urgently to propose.
Summary of the invention
The embodiment of the present invention provides the method and device that a kind of music is recommended, in order to solve in prior art
Search Search Results inconvenience, the coarse technological deficiency of the music way of recommendation.
The embodiment of the present invention provides a kind of method that music is recommended, including:
Detect that user initiates the event of searching request, initiate the scene of request according to user, obtain environment
Information;
Extract the required parameter of user according to described request, described required parameter includes user model, described
The label of music content in environmental information and described searching request;
Label according to described music content filters out candidate music from data base;
Mark is mated according to what described user model combining music model calculated user and described candidate music,
According to described coupling mark described candidate music it is ranked up and recommends user.
Correspondingly the embodiment of the present invention additionally provides the device that a kind of music is recommended, including:
Detection module, for detecting that user initiates the event of searching request, initiates request according to user
Scene, obtains environmental information;
Extraction module, for extracting the required parameter of user according to described request, described required parameter includes
The label of music content in user model, described environmental information and described searching request;
Screening module, for filtering out candidate music according to the label of described music content from data base;
Recommending module, for calculating user and described candidate's sound according to described user model combining music model
Happy coupling mark, is ranked up described candidate music according to described coupling mark and recommends user.
The technical scheme provided according to the present invention, it is possible to user model and music model according to foundation are for using
Accurate recommendation service that family provides and user preference matches, the most just i.e. comes the Search Results of foremost
It is the content that user oneself is to be looked for, meets the individual demand of user, improve user's experience.
Accompanying drawing explanation
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to reality
Execute the required accompanying drawing used in example or description of the prior art to be briefly described, it should be apparent that under,
Accompanying drawing during face describes is some embodiments of the present invention, for those of ordinary skill in the art,
On the premise of not paying creative work, it is also possible to obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is the schematic flow sheet of the method that music is recommended in the embodiment of the present invention;
Fig. 2 is the structural representation of the device that music is recommended in the embodiment of the present invention;
Fig. 3 is the structural representation of the device that a kind of preferably music is recommended in the embodiment of the present invention;
Fig. 4 is the structural representation of the device that another kind of preferably music is recommended in the embodiment of the present invention.
Detailed description of the invention
For making the purpose of the embodiment of the present invention, technical scheme and advantage clearer, below in conjunction with this
Accompanying drawing in bright embodiment, is clearly and completely described the technical scheme in the embodiment of the present invention,
Obviously, described embodiment is a part of embodiment of the present invention rather than whole embodiments.Based on
Embodiment in the present invention, those of ordinary skill in the art are obtained under not making creative work premise
The every other embodiment obtained, broadly falls into the scope of protection of the invention.
In currently available technology, user often uses following several ways searching for music when:
1, in App, find out certain a music App, open this App, find corresponding entrance (or
Person searches for) look for the music liked;
2, open the universal search page (such as Baidu's search), inside input the music name liked, so
After in Search Results find;
3, input, in the search entrance of embedded in mobile phone, the music name liked, then find in Search Results.
The mode of above search music is owing to being comprehensive search, then music result may possibly be mixed in many
Among other result, and be inconvenient to find.
When search for entrance music brought do the content recommended to user carry out commending contents to user time, because of
For the abundant in content of music and wide variety, and the grade of music user has diversified difference,
When carrying out recommending music, the content recommended is not often that user is wanted, it is recommended that content inaccuracy,
So current music way of recommendation user experience is the lowest.
For solving prior art is searched Search Results inconvenience, the coarse technology of the music way of recommendation
Defect, embodiments provides the method and device that a kind of music is recommended.
The embodiment of the embodiment of the present invention is described in detail below in conjunction with drawings and Examples, the most right
The embodiment of the present invention how application technology means solve technical problem and reach technology effect realize process
Can fully understand and implement according to this.
Embodiment one
Fig. 1 is the schematic flow sheet of the method that music is recommended in the embodiment of the present invention, as shown in Figure 1:
The embodiment of the present invention provides a kind of method that music is recommended, including:
Step S101: detect that user initiates the event of searching request, initiates the scene of request according to user,
Obtain environmental information;
When server detects the event that user initiates searching request, initiate the scene of request according to user,
Obtaining environmental information, wherein, environmental information includes: the ID of user, current device, date/period, no
Same environmental information can produce different impacts to the result recommended.
For the embodiment of the present invention preferably, detect that user initiates the event of searching request, wrap further
Include: detect that user enters search entrance and initiates the event of searching request, wherein, described search entrance bag
Include: search entrance, universal search entrance and the search entrance of APP of terminal built-in.Searching of terminal built-in
Rope entrance is the function of search entrance of the acquiescence that terminal carries, and may search for the various literary compositions included in terminal
Part;The search entrance that universal search entrance provides for search server, such as Baidu's search, may search for
The all resources being connected with search server;The search entrance of APP is that the search arranged in APP program enters
Mouthful, may search for the resource in APP and the related resource in the server being connected with APP.
The search entrance of terminal built-in can provide the user music searching entry mode easily, when with
When function of search is wanted to call in family, function of search can be initiated request and enter the event of search entrance.
Step S102: according to searching request extract user required parameter, required parameter include user model,
The label of music content in environmental information and searching request;
After getting environmental information, extract the required parameter of user according to Client-initiated searching request, please
Ask and parameter includes the label of music content in user model, environmental information and searching request, wherein,
User model is pre-build.
Step S103: filter out candidate music from data base according to the label of music content;
After extracting required parameter, screen from data base according to the label of the music content in required parameter
Going out the candidate music corresponding with music content label, wherein, candidate music may be recommended to user,
It is likely to be recommended to user.
Step S104: mate mark according to what user model combining music model calculated user and candidate music,
Candidate music is ranked up according to mating mark and recommends user.
According to the user model pre-build and music model, calculating user mates mark with candidate music,
Candidate music is ranked up according to mating mark and recommends user.
The method that the music provided according to the present embodiment is recommended, when user initiates searching request, obtains and uses
Scene during request is initiated at family, and obtains environmental information, obtains the required parameter of user according to searching request,
Label filtration according to the music content in required parameter goes out candidate music, then by the user pre-build
What model and music model calculated candidate music and user mates mark, is that user recommends according to mating mark
Individualized music, can provide the user inclined with user according to the user model set up and music model simultaneously
Match well accurate recommendation service, and the Search Results i.e. coming foremost is exactly just that user oneself to look for
Content, meet the individual demand of user, improve user's experience.
In the embodiment of the present invention, user model, music model all pre-build, are described in detail below one
Lower user model, music model set up process.
The embodiment of the present invention preferably, asks to extract required parameter according to user, the most also includes setting up using
Family model, including:
User is collected to the preference information of music and user data from various data sources, wherein, hobby letter
Breath includes song that user likes, singer, school, style, rhythm, age, and user data includes using
The music that family was listened, the webpage browsed, the video seen, search history, the App of use, shopping are gone through
History;
The scene occurred when combining collect preference information and user data according to preference information and user data,
Obtain the portrait of a labeling of user interest, set up user model according to the portrait of labeling.
Need when setting up user model user related data to be collected and models, set up user model,
It is collected user related data including that the hobby of music is believed by the user collected from various data sources
Breath, likes which song, which singer, where before including but not limited to song that user likes, user
A little schools, which style, what rhythm, what age etc., also include collecting user data, including
Collect the various data in the terminal allowed from user, include but not limited to music that user listened, clear
The webpage look at, the video seen, search history, the App of use, shopping history etc., based on these
Event, the scene occurred in conjunction with these events, obtain the portrait of a labeling of user interest, this
Portrait is exactly user model.
Pre-build music model, the embodiment of the present invention preferably, according to user model combining music model
Calculating user mates mark with candidate music, the most also includes setting up music model, including:
Gathering the related data of music content, related data includes that all labels of music content, music are special
Volume relevant picture, broadcast address and tendency data, wherein, label includes: music name, singer,
Write words, wrirte music, the lyrics, school, the age, country origin, tendency data include broadcasting time, collection people
Number, searching times, broadcasting ranking tendency;
Obtain a datumization to music content according to related data to describe, describe according to datumization and set up
Music model.
Need when setting up music model to gather and music content related data, including from various dimensions to music
All labels that content is described, no matter user gives expression to the interest to this music from which dimension,
Can be by corresponding tag match to this music, music model is to music content modeling,
Gather all labels of music content, include but not limited to music name, singer, write words, wrirte music, the lyrics,
School, age, country origin etc., the data relevant to music content also include but not limited to music album phase
The information such as picture, broadcast address and the tendency data closed, tendency data include broadcasting time, receipts
The tendency data such as Tibetan's number, searching times, broadcasting ranking tendency, according to the related data of music content
Obtain to music content datumization to describe, describe further according to datumization and set up music model.
According to user model combining music model calculating user and when mating mark of candidate music, also need
Order models to be used, order models sets up the most in advance, the embodiment of the present invention preferably, root
Mate mark according to what user model combining music model calculated user and candidate music, the most also include foundation
Order models, including:
Gather user and use the historical data of music product, from historical data, collect training data;
Combine user model according to training data and music model is trained, set up row according to training result
Sequence model.
Need to gather user when setting up order models and use the historical data of music product, can be according to machine
Device learning algorithm, uses collection training data the historical data of music product from user, by training number
According to combining user model and music model is trained, obtain order models according to training result, based on row
Sequence model, the model data of input user and the model data of music, it is possible to obtain user and candidate music
Coupling mark, for a user, the music that score is high is exactly the recommendation set of candidate, such as, defeated
Enter model data and the model data of arbitrary music of any user, it is possible to obtain a user and music
Coupling mark, the candidate music composition filtering out score high recommends set.
For step S103, the embodiment of the present invention preferably, according to the label of music content from data
Storehouse filters out candidate music, farther includes:
From data base, the music list corresponding with label is obtained according to the label in required parameter;
The candidate music corresponding with label is filtered out from music list.
Containing inverted file in data base, inverted file is a Fast Lookup Table, it is possible to according to respectively
Plant label quick obtaining and there is the candidate music list of this label, in the required parameter of acquisition includes music
The label held, corresponding with label according to the label of the music content inverted file in data base obtains
Music list, then from music list, filter out the candidate music corresponding with the label of music content,
Wherein, the label of music content include but not limited to singer, write words, wrirte music, relevant musical, the lyrics,
Age, school, style, country origin etc., according to the label of the music content in required parameter, from the row of falling
Extract the music of candidate in file, be then passed to order models and carry out subsequent operation.
For step S104, the embodiment of the present invention preferably, according to user model combining music model
Calculating user mates mark with candidate music, is ranked up by candidate music according to mating mark and recommends
To user, farther include:
Data in data in user model, music model and candidate music are input in order models,
Mark is mated according to what order models obtained user and candidate music;
According to coupling mark order from big to small, candidate music is ranked up, mark will be mated more than dividing
The candidate music of number threshold value recommends user.
According to user model combining music model calculating user and when mating mark of candidate music, also need
Data in data in user model, music model and candidate music are inputted by order models to be used
In order models, order models can obtain the coupling mark of a user and candidate music through calculating,
Order models according to the coupling mark of candidate music to candidate music according to coupling the carrying out from big to small of mark
Sequence, pre-sets and score threshold simultaneously, is recommended more than the candidate music of score threshold by coupling mark
To user, so as to accomplish the purpose to user's accurate recommendation music, wherein, the number in user model
According to including but not limited to collect user from various data sources to the preference information of music and user data, use
Family includes song that user likes, singer, school, style, rhythm, age to the preference information of music
Etc., user data include the music that user listened, the webpage browsed, the video seen, search history,
The App of use, shopping history etc., the data in music model include but not limited to the phase of music content
Closing data, related data includes picture, the broadcasting ground that all labels of music content, music album are relevant
Location and tendency data etc., wherein, label includes: music name, singer, writes words, wrirte music, sing
Word, school, age, country origin etc., tendency data include broadcasting time, collection number, search time
Number, broadcasting ranking tendency etc..
Duplicate in order to avoid the candidate music recommended again occurs copyright is illegal, in recommendation results
Music and to user can not the situation of normal presentation, the embodiment of the present invention preferably, will according to coupling mark
Candidate music is ranked up and recommends user, the most also includes:
The candidate music recommended is carried out Business Processing, including: copyright filtration treatment, duplicate removal filtration treatment
And displaying logical process.
After calculating the coupling mark of candidate music, candidate music is ranked up, is recommending to user
Before music, legal by ensureing the music copyright recommended, need the copyright to candidate music to carry out copyright
Filtration treatment, filters out illegal for copyright music, in order to avoid user when listening to music due to
The trouble that copyright reasons can not normally be listened to.
There may come a time when to there will be the situation that multiple music is identical when screening candidate music, so needing time
Contrast between the music of choosing, carry out duplicate removal filtration treatment, the music of repetition is filtered out, it is to avoid use
The family music that uppick repeats when listening to music according to recommendation results.
To user show before, in addition it is also necessary to recommend music be shown logical process, including to
The field that family is shown processes, it is ensured that user can normally listen to recommended music.
To recommend candidate music carry out Business Processing, include but not limited to copyright filter, go heavy filtration,
Show logic etc., it is ensured that the result of recommendation is that copyright is legal, unduplicated, it is normal to show.
Embodiment two
Fig. 2 is the structural representation of the device that music is recommended in the embodiment of the present invention, as shown in Figure 2:
Correspondingly the present embodiment additionally provides the device that a kind of music is recommended, including:
Detection module 1, for detecting that user initiates the event of searching request, initiates request according to user
Scene, obtain environmental information;
Extraction module 2, for extracting the required parameter of user, described required parameter bag according to described request
Include the label of music content in user model, described environmental information and described searching request;
Extraction module 2 couples with detection module 1, and the environmental information got is sent extremely by detection module 1
Extraction module 2, extraction module 2 extracts the required parameter of user according to searching request.
Screening module 3, for filtering out candidate music according to the label of described music content from data base;
Screening module 3 couples with extraction module 2, according to the label of the music content that extraction module 2 extracts
The music corresponding with the label of music content is filtered out as candidate music from data base.
Recommending module 4, for according to user model combining music model calculate user and candidate music
Partition number, is ranked up candidate music according to mating mark and recommends user.
Recommending module 4 couples with screening module 3 phase, according to user model and music model calculate user with
The coupling mark of candidate music, is ranked up candidate music further according to mating mark and recommends user.
The device that the music provided according to the present embodiment is recommended, detection module 1 detects that user initiates search
During request, obtaining user and initiate environmental information when asking, extraction module 2 extracts according to searching request and uses
The required parameter at family, screening module 3 goes out candidate's sound according to the label filtration of the music content in required parameter
Happy, it is recommended that module 4 is that user recommends individualized music according to the mark that mates of candidate music with user, for
User provides music searching entry mode easily, simultaneously can be according to the user model set up and music
Model provides the user accurate recommendation service, i.e. come the Search Results of foremost be exactly just user from
The content that oneself is to be looked for, meets the individual demand of user, improves user's experience.Fig. 3 is this
The structural representation of the device that a kind of preferably music is recommended in inventive embodiments, as shown in Figure 3:
The present embodiment preferably, also includes: set up module 5, for collecting user from various data sources
Preference information and user data to music, wherein, preference information include song that user likes, singer,
School, style, rhythm, age, user data includes music, the webpage browsed that user listened, sees
The video crossed, search history, the App of use, shopping history;
The scene occurred when combining collect preference information and user data according to preference information and user data,
Obtain the portrait of a labeling of user interest, set up user model according to the portrait of labeling.
Preferably, setting up module 5, be additionally operable to gather the related data of music content, related data includes
Picture, broadcast address and the tendency data that all labels of music content, music album are relevant, its
In, label includes: music name, singer, write words, wrirte music, the lyrics, school, the age, country origin, become
Gesture data include broadcasting time, collection number, searching times, broadcasting ranking tendency;
Obtain a datumization to music content according to related data to describe, describe according to datumization and set up
Music model.
Preferably, set up module 5, be additionally operable to gather user and use the historical data of music product, from going through
History data are collected training data;
Combine user model according to training data and music model is trained, set up row according to training result
Sequence model.
Set up module 5 to couple with detection module 1, be used for pre-building user model, music model and
Order models.
Preferably, screen module 3, be further used for obtaining from data base according to the label in required parameter
Take the music list corresponding with label;
The candidate music corresponding with label is filtered out from music list.
Preferably, it is recommended that module 4, it is further used in the data in user model, music model
Data and candidate music are input in order models, according to order models obtain user and candidate music
Partition number;
According to coupling mark order from big to small, candidate music is ranked up, mark will be mated more than dividing
The candidate music of number threshold value recommends user.
Fig. 4 is the structural representation of the device that another kind of preferably music is recommended in the present embodiment, such as Fig. 4
Shown in:
The present embodiment preferably, also includes: Service Processing Module 6, for entering the candidate music recommended
Row Business Processing, including: copyright filtration treatment, duplicate removal filtration treatment and displaying logical process.
Service Processing Module 6 couples mutually with recommending module, for the candidate music recommended is carried out business phase
The various process closed, including the solution of the replication problem screened between the copyright legitimacy of music, recommendation results
Certainly, field process shown to user etc., in order to ensure the candidate music recommended be copyright legal,
Unduplicated, show normal.
It addition, the present embodiment is preferably, detection module 1, it is further used for: detect that user enters and search
Rope entrance initiates the event of searching request, and wherein, search entrance includes: the search entrance of terminal built-in,
The search entrance of universal search entrance and APP.
Terminal shown in Fig. 2-4 can perform the method for embodiment illustrated in fig. 1, it is achieved principle and technique effect
With reference to embodiment illustrated in fig. 1, method is corresponding with terminal feature, can be cross-referenced, the most another
One repeats.
Application scenarios
Below by application scenarios, in order to the embodiment of the present invention is done more detailed understanding, needs explanation
Be following application scenarios be not whole scenes of the embodiment of the present invention, simply in the embodiment of the present invention
Certain applications scene.
User enter into search entrance time, according to user initiate request time scene, obtain user ID,
Current device, date/period, extract the required parameter of user, according to the music content in required parameter
Label filtration goes out candidate music, further according to the data in the user model pre-build and music model, row
Sequence model calculates the coupling mark of candidate music, according to coupling mark, music is recommended user, it is recommended that
Music is that copyright is legal, unduplicated, it is normal to show.
In sum, the technical scheme provided according to the present invention, when user initiates searching request, obtain
User initiates scene during request, and obtains environmental information, obtains the request ginseng of user according to searching request
Number, go out candidate music according to the label filtration of the music content in required parameter, then by pre-building
What user model and music model calculated candidate music and user mates mark, is user according to mating mark
Recommend individualized music, provide the user music searching entry mode easily, simultaneously can be according to building
Vertical user model and music model provide the user accurate recommendation service, i.e. come the search of foremost
Result is exactly just the content that user oneself is to be looked for, and meets the individual demand of user, improves user
Experience.
Device embodiment described above is only schematically, wherein said illustrates as separating component
Unit can be or may not be physically separate, the parts shown as unit can be or
Person may not be physical location, i.e. may be located at a place, or can also be distributed to multiple network
On unit.Some or all of module therein can be selected according to the actual needs to realize the present embodiment
The purpose of scheme.Those of ordinary skill in the art are not in the case of paying performing creative labour, the most permissible
Understand and implement.
Last it is noted that above example is only in order to illustrate technical scheme, rather than to it
Limit;Although the present invention being described in detail with reference to previous embodiment, the ordinary skill of this area
Personnel it is understood that the technical scheme described in foregoing embodiments still can be modified by it, or
Person carries out equivalent to wherein portion of techniques feature;And these amendments or replacement, do not make corresponding skill
The essence of art scheme departs from the spirit and scope of various embodiments of the present invention technical scheme.
Claims (16)
1. the method that a music is recommended, it is characterised in that including:
Detect that user initiates the event of searching request, initiate the scene of request according to user, obtain environment
Information;
According to described searching request extract user required parameter, described required parameter include user model,
The label of music content in described environmental information and described searching request;
Label according to described music content filters out candidate music from data base;
Mark is mated according to what described user model combining music model calculated user and described candidate music,
According to described coupling mark described candidate music it is ranked up and recommends user.
Method the most according to claim 1, it is characterised in that ask to extract according to described user and ask
Seek parameter, the most also include setting up described user model, including:
User is collected to the preference information of music and user data, wherein, described happiness from various data sources
Good information includes song that user likes, singer, school, style, rhythm, age, described number of users
According to include the music that user listened, the webpage browsed, the video seen, search history, the App of use,
Shopping history;
Combine according to described preference information and described user data and collect described preference information and described number of users
According to time the scene that occurs, obtain the portrait of a labeling of user interest, according to the picture of described labeling
As setting up described user model.
Method the most according to claim 1, it is characterised in that combine sound according to described user model
Happy model calculating user mates mark with described candidate music, the most also includes setting up described music model,
Including:
Gather music content related data, described related data include described music content all labels,
Picture, broadcast address and the tendency data that music album is relevant, wherein, described label includes: sound
Happy name, singer, write words, wrirte music, the lyrics, school, the age, country origin, described tendency data include
Broadcasting time, collection number, searching times, broadcasting ranking tendency;
Obtain a datumization to described music content according to described related data to describe, according to described number
Described music model is set up according to changing description.
Method the most according to claim 1, it is characterised in that combine sound according to described user model
Happy model calculating user mates mark with described candidate music, the most also includes setting up order models, bag
Include:
Gather user and use the historical data of music product, from described historical data, collect training data;
Combine described user model according to described training data and described music model is trained, according to instruction
Practice result and set up described order models.
Method the most according to claim 1, it is characterised in that according to the label of described music content
From data base, filter out candidate music, farther include:
Obtain corresponding with described label from described data base according to the described label in described required parameter
Music list;
The described candidate music corresponding with described label is filtered out from described music list.
Method the most according to claim 4, it is characterised in that combine sound according to described user model
Happy model calculating user mates mark, according to described coupling mark by described candidate with described candidate music
Music is ranked up and recommends user, farther includes:
Data in data in described user model, described music model and described candidate music are inputted
In described order models, mate mark according to what described order models obtained user and described candidate music;
According to described coupling mark order from big to small, described candidate music is ranked up, by described
Partition number recommends user more than the described candidate music of score threshold.
Method the most according to claim 1, it is characterised in that according to described coupling mark by described
Candidate music is ranked up and recommends user, the most also includes:
The described candidate music recommended is carried out Business Processing, including: copyright filtration treatment, goes heavy filtration
Process and show logical process.
8. according to the arbitrary described method of claim 1-7, it is characterised in that detect that user initiates to search
The event of rope request, farther includes: detect that user enters search entrance and initiates the event of searching request,
Wherein, described search entrance includes: search entrance, universal search entrance and the APP of terminal built-in searches
Rope entrance.
9. the device that a music is recommended, it is characterised in that including:
Detection module, for detecting that user initiates the event of searching request, initiates request according to user
Scene, obtains environmental information;
Extraction module, for extracting the required parameter of user, described required parameter according to described searching request
Including the label of music content in user model, described environmental information and described searching request;
Screening module, for filtering out candidate music according to the label of described music content from data base;
Recommending module, for calculating user and described candidate's sound according to described user model combining music model
Happy coupling mark, is ranked up described candidate music according to described coupling mark and recommends user.
Device the most according to claim 9, it is characterised in that also include:
Set up module, for collecting user from various data sources to the preference information of music and user data,
Wherein, described preference information includes song that user likes, singer, school, style, rhythm, age,
Described user data include the music that user listened, the webpage browsed, the video seen, search history,
The App of use, shopping history;
Combine according to described preference information and described user data and collect described preference information and described number of users
According to time the scene that occurs, obtain the portrait of a labeling of user interest, according to the picture of described labeling
As setting up described user model.
11. devices according to claim 10, it is characterised in that
Described setting up module, be additionally operable to gather the related data of music content, described related data includes institute
State all labels of music content, picture, broadcast address and tendency data that music album is relevant,
Wherein, described label includes: music name, singer, write words, wrirte music, the lyrics, school, the age, state
Not, described tendency data include broadcasting time, collection number, searching times, broadcasting ranking tendency;
Obtain a datumization to described music content according to described related data to describe, according to described number
Described music model is set up according to changing description.
12. devices according to claim 10, it is characterised in that
Described set up module, be additionally operable to gather user and use the historical data of music product, from described history
Data are collected training data;
Combine described user model according to described training data and described music model is trained, according to instruction
Practice result and set up described order models.
13. devices according to claim 9, it is characterised in that
Described screening module, is further used for according to the described label in described required parameter from described data
Storehouse obtains the music list corresponding with described label;
The described candidate music corresponding with described label is filtered out from described music list.
14. devices according to claim 12, it is characterised in that
Described recommending module, is further used in the data in described user model, described music model
Data and described candidate music be input in described order models, according to described order models obtain user
With described candidate music mate mark;
According to described coupling mark order from big to small, described candidate music is ranked up, by described
Partition number recommends user more than the described candidate music of score threshold.
15. devices according to claim 9, it is characterised in that also include:
Service Processing Module, for the described candidate music recommended is carried out Business Processing, including: copyright
Filtration treatment, duplicate removal filtration treatment and displaying logical process.
16. according to the arbitrary described device of claim 9-15, it is characterised in that described detection module,
Be further used for: detect user enter search entrance initiate searching request event, wherein, described in search
Rope entrance includes: search entrance, universal search entrance and the search entrance of APP of terminal built-in.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610245225.2A CN105930429A (en) | 2016-04-19 | 2016-04-19 | Music recommendation method and apparatus |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610245225.2A CN105930429A (en) | 2016-04-19 | 2016-04-19 | Music recommendation method and apparatus |
Publications (1)
Publication Number | Publication Date |
---|---|
CN105930429A true CN105930429A (en) | 2016-09-07 |
Family
ID=56839514
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610245225.2A Pending CN105930429A (en) | 2016-04-19 | 2016-04-19 | Music recommendation method and apparatus |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105930429A (en) |
Cited By (33)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106446048A (en) * | 2016-08-31 | 2017-02-22 | 维沃移动通信有限公司 | Song recommendation method and mobile terminal |
CN106572390A (en) * | 2016-11-09 | 2017-04-19 | 咪咕音乐有限公司 | Audio and video recommending method and equipment |
CN106686431A (en) * | 2016-12-08 | 2017-05-17 | 杭州网易云音乐科技有限公司 | Synthesizing method and equipment of audio file |
CN106802943A (en) * | 2017-01-03 | 2017-06-06 | 海信集团有限公司 | Music based on message of film and TV recommends method and device |
CN106846191A (en) * | 2016-11-25 | 2017-06-13 | 北京粉笔蓝天科技有限公司 | A kind of method of combination of curriculums table, system and server |
CN106844660A (en) * | 2017-01-23 | 2017-06-13 | 腾讯科技(深圳)有限公司 | A kind of music recommends method and device |
CN108052515A (en) * | 2017-10-19 | 2018-05-18 | 福建中金在线信息科技有限公司 | The method and apparatus for storing local data |
CN108319635A (en) * | 2017-12-15 | 2018-07-24 | 海南智媒云图科技股份有限公司 | A kind of multi-platform music sources integrate method, electronic equipment and the storage medium of broadcasting |
CN108345419A (en) * | 2017-01-25 | 2018-07-31 | 华为技术有限公司 | A kind of generation method and device of information recommendation list |
CN108509534A (en) * | 2018-03-15 | 2018-09-07 | 华南理工大学 | Individualized music commending system based on deep learning and its implementation |
CN109063163A (en) * | 2018-08-14 | 2018-12-21 | 腾讯科技(深圳)有限公司 | A kind of method, apparatus, terminal device and medium that music is recommended |
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 |
CN109857901A (en) * | 2019-01-25 | 2019-06-07 | 杭州网易云音乐科技有限公司 | Information displaying method and device and method and apparatus for information search |
CN109992694A (en) * | 2019-04-17 | 2019-07-09 | 重庆誉存大数据科技有限公司 | A kind of music intelligent recommendation method and system |
CN110020192A (en) * | 2018-07-31 | 2019-07-16 | 北京微播视界科技有限公司 | A kind of information content method for pushing and device, server device |
CN110209844A (en) * | 2019-05-17 | 2019-09-06 | 腾讯音乐娱乐科技(深圳)有限公司 | Multi-medium data matching process, device and storage medium |
CN110402438A (en) * | 2016-12-29 | 2019-11-01 | 谷歌有限责任公司 | Music from focus inquiry is recommended |
CN110717062A (en) * | 2018-07-11 | 2020-01-21 | 阿里巴巴集团控股有限公司 | Music searching and vehicle-mounted music playing method, device, equipment and storage medium |
CN110809187A (en) * | 2019-10-31 | 2020-02-18 | Oppo广东移动通信有限公司 | Video selection method, video selection device, storage medium and electronic equipment |
CN110874788A (en) * | 2019-11-13 | 2020-03-10 | 合肥美的电冰箱有限公司 | Product recommendation method and device and computer storage medium |
CN111046221A (en) * | 2019-12-17 | 2020-04-21 | 腾讯科技(深圳)有限公司 | Song recommendation method and device, terminal equipment and storage medium |
CN111078998A (en) * | 2019-11-19 | 2020-04-28 | Oppo(重庆)智能科技有限公司 | Information retrieval method, information retrieval device, storage medium and server |
CN111259192A (en) * | 2020-01-15 | 2020-06-09 | 腾讯科技(深圳)有限公司 | Audio recommendation method and device |
CN111435369A (en) * | 2019-01-14 | 2020-07-21 | 腾讯科技(深圳)有限公司 | Music recommendation method, device, terminal and storage medium |
CN111723289A (en) * | 2020-06-08 | 2020-09-29 | 北京声智科技有限公司 | Information recommendation method and device |
CN111723226A (en) * | 2020-06-14 | 2020-09-29 | 黄雨勤 | Information management method based on big data and Internet and artificial intelligence cloud server |
CN112579837A (en) * | 2019-09-27 | 2021-03-30 | 阿里巴巴集团控股有限公司 | Data processing method and device, electronic equipment and computer storage medium |
CN112711702A (en) * | 2019-10-25 | 2021-04-27 | 华为技术有限公司 | Recommendation method based on user privacy data, and apparatus, medium, and system thereof |
CN113641902A (en) * | 2021-08-10 | 2021-11-12 | 广州艾美网络科技有限公司 | Music information pushing method and device, computer equipment and storage medium thereof |
CN113742514A (en) * | 2021-09-03 | 2021-12-03 | 林飞鹏 | Accurate music searching method and device |
CN114501059A (en) * | 2022-01-12 | 2022-05-13 | 阿里巴巴(中国)有限公司 | Video score processing method and device, electronic equipment and readable storage medium |
CN114722234A (en) * | 2022-04-07 | 2022-07-08 | 平安科技(深圳)有限公司 | Music recommendation method, device and storage medium based on artificial intelligence |
CN115760296A (en) * | 2022-11-24 | 2023-03-07 | 阿里巴巴(中国)有限公司 | Page data processing and browsing method, terminal device and storage medium |
Citations (4)
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 |
EP2763056A1 (en) * | 2007-03-31 | 2014-08-06 | Sony Deutschland Gmbh | Method for content recommendation |
CN103970873A (en) * | 2014-05-14 | 2014-08-06 | 中国联合网络通信集团有限公司 | Music recommending method and system |
US8838617B2 (en) * | 2011-11-28 | 2014-09-16 | Electronics And Telecommunications Research Institute | Method and apparatus for searching for recommended music using emotional information of music |
-
2016
- 2016-04-19 CN CN201610245225.2A patent/CN105930429A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2763056A1 (en) * | 2007-03-31 | 2014-08-06 | Sony Deutschland Gmbh | Method for content recommendation |
US8838617B2 (en) * | 2011-11-28 | 2014-09-16 | Electronics And Telecommunications Research Institute | Method and apparatus for searching for recommended music using emotional information of music |
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 |
CN103970873A (en) * | 2014-05-14 | 2014-08-06 | 中国联合网络通信集团有限公司 | Music recommending method and system |
Cited By (50)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106446048A (en) * | 2016-08-31 | 2017-02-22 | 维沃移动通信有限公司 | Song recommendation method and mobile terminal |
CN106572390A (en) * | 2016-11-09 | 2017-04-19 | 咪咕音乐有限公司 | Audio and video recommending method and equipment |
CN106572390B (en) * | 2016-11-09 | 2020-03-13 | 咪咕音乐有限公司 | Audio and video recommendation method and equipment |
CN106846191A (en) * | 2016-11-25 | 2017-06-13 | 北京粉笔蓝天科技有限公司 | A kind of method of combination of curriculums table, system and server |
CN106686431B (en) * | 2016-12-08 | 2019-12-10 | 杭州网易云音乐科技有限公司 | Audio file synthesis method and device |
CN106686431A (en) * | 2016-12-08 | 2017-05-17 | 杭州网易云音乐科技有限公司 | Synthesizing method and equipment of audio file |
CN110402438A (en) * | 2016-12-29 | 2019-11-01 | 谷歌有限责任公司 | Music from focus inquiry is recommended |
US11704359B2 (en) | 2016-12-29 | 2023-07-18 | Google Llc | Music recommendations from trending queries |
CN110402438B (en) * | 2016-12-29 | 2024-03-08 | 谷歌有限责任公司 | Music recommendation from popular queries |
CN106802943B (en) * | 2017-01-03 | 2020-06-09 | 海信集团有限公司 | Music recommendation method and device based on movie and television information |
CN106802943A (en) * | 2017-01-03 | 2017-06-06 | 海信集团有限公司 | Music based on message of film and TV recommends method and device |
CN106844660B (en) * | 2017-01-23 | 2022-09-13 | 腾讯科技(深圳)有限公司 | Music recommendation method and device |
CN106844660A (en) * | 2017-01-23 | 2017-06-13 | 腾讯科技(深圳)有限公司 | A kind of music recommends method and device |
CN108345419A (en) * | 2017-01-25 | 2018-07-31 | 华为技术有限公司 | A kind of generation method and device of information recommendation list |
CN108052515A (en) * | 2017-10-19 | 2018-05-18 | 福建中金在线信息科技有限公司 | The method and apparatus for storing local data |
CN108319635A (en) * | 2017-12-15 | 2018-07-24 | 海南智媒云图科技股份有限公司 | A kind of multi-platform music sources integrate method, electronic equipment and the storage medium of broadcasting |
CN108509534A (en) * | 2018-03-15 | 2018-09-07 | 华南理工大学 | Individualized music commending system based on deep learning and its implementation |
CN108509534B (en) * | 2018-03-15 | 2022-03-25 | 华南理工大学 | Personalized music recommendation system based on deep learning and implementation method thereof |
CN110717062A (en) * | 2018-07-11 | 2020-01-21 | 阿里巴巴集团控股有限公司 | Music searching and vehicle-mounted music playing method, device, equipment and storage medium |
CN110717062B (en) * | 2018-07-11 | 2024-03-22 | 斑马智行网络(香港)有限公司 | Music search and vehicle-mounted music playing method, device, equipment and storage medium |
CN110020192A (en) * | 2018-07-31 | 2019-07-16 | 北京微播视界科技有限公司 | A kind of information content method for pushing and device, server device |
CN109063163A (en) * | 2018-08-14 | 2018-12-21 | 腾讯科技(深圳)有限公司 | A kind of method, apparatus, terminal device and medium that music is recommended |
CN109063163B (en) * | 2018-08-14 | 2022-12-02 | 腾讯科技(深圳)有限公司 | Music recommendation method, device, terminal equipment and medium |
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 |
CN111435369B (en) * | 2019-01-14 | 2024-04-09 | 腾讯科技(深圳)有限公司 | Music recommendation method, device, terminal and storage medium |
CN111435369A (en) * | 2019-01-14 | 2020-07-21 | 腾讯科技(深圳)有限公司 | Music recommendation method, device, terminal and storage medium |
CN109857901A (en) * | 2019-01-25 | 2019-06-07 | 杭州网易云音乐科技有限公司 | Information displaying method and device and method and apparatus for information search |
CN109992694A (en) * | 2019-04-17 | 2019-07-09 | 重庆誉存大数据科技有限公司 | A kind of music intelligent recommendation method and system |
CN110209844A (en) * | 2019-05-17 | 2019-09-06 | 腾讯音乐娱乐科技(深圳)有限公司 | Multi-medium data matching process, device and storage medium |
CN112579837A (en) * | 2019-09-27 | 2021-03-30 | 阿里巴巴集团控股有限公司 | Data processing method and device, electronic equipment and computer storage medium |
CN112711702A (en) * | 2019-10-25 | 2021-04-27 | 华为技术有限公司 | Recommendation method based on user privacy data, and apparatus, medium, and system thereof |
CN110809187A (en) * | 2019-10-31 | 2020-02-18 | Oppo广东移动通信有限公司 | Video selection method, video selection device, storage medium and electronic equipment |
CN110874788A (en) * | 2019-11-13 | 2020-03-10 | 合肥美的电冰箱有限公司 | Product recommendation method and device and computer storage medium |
CN111078998A (en) * | 2019-11-19 | 2020-04-28 | Oppo(重庆)智能科技有限公司 | Information retrieval method, information retrieval device, storage medium and server |
CN111078998B (en) * | 2019-11-19 | 2024-03-12 | Oppo(重庆)智能科技有限公司 | Information retrieval method, device, storage medium and server |
CN111046221B (en) * | 2019-12-17 | 2024-06-07 | 腾讯科技(深圳)有限公司 | Song recommendation method, device, terminal equipment and storage medium |
CN111046221A (en) * | 2019-12-17 | 2020-04-21 | 腾讯科技(深圳)有限公司 | Song recommendation method and device, terminal equipment and storage medium |
CN111259192B (en) * | 2020-01-15 | 2023-12-01 | 腾讯科技(深圳)有限公司 | Audio recommendation method and device |
CN111259192A (en) * | 2020-01-15 | 2020-06-09 | 腾讯科技(深圳)有限公司 | Audio recommendation method and device |
CN111723289A (en) * | 2020-06-08 | 2020-09-29 | 北京声智科技有限公司 | Information recommendation method and device |
CN111723289B (en) * | 2020-06-08 | 2024-02-02 | 北京声智科技有限公司 | Information recommendation method and device |
CN111723226A (en) * | 2020-06-14 | 2020-09-29 | 黄雨勤 | Information management method based on big data and Internet and artificial intelligence cloud server |
CN113641902A (en) * | 2021-08-10 | 2021-11-12 | 广州艾美网络科技有限公司 | Music information pushing method and device, computer equipment and storage medium thereof |
CN113742514B (en) * | 2021-09-03 | 2023-11-24 | 林飞鹏 | Music accurate searching method and device |
CN113742514A (en) * | 2021-09-03 | 2021-12-03 | 林飞鹏 | Accurate music searching method and device |
CN114501059A (en) * | 2022-01-12 | 2022-05-13 | 阿里巴巴(中国)有限公司 | Video score processing method and device, electronic equipment and readable storage medium |
CN114722234A (en) * | 2022-04-07 | 2022-07-08 | 平安科技(深圳)有限公司 | Music recommendation method, device and storage medium based on artificial intelligence |
CN115760296B (en) * | 2022-11-24 | 2023-10-17 | 阿里巴巴(中国)网络技术有限公司 | Page data processing and browsing method, terminal equipment and storage medium |
CN115760296A (en) * | 2022-11-24 | 2023-03-07 | 阿里巴巴(中国)有限公司 | Page data processing and browsing method, terminal device and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105930429A (en) | Music recommendation method and apparatus | |
CN104573054B (en) | A kind of information-pushing method and equipment | |
KR101343609B1 (en) | Apparatus and Method for Automatically recommending Application using Augmented Reality Data | |
CN107562939B (en) | Vertical domain news recommendation method and device and readable storage medium | |
CN103927309B (en) | A kind of method and device to business object markup information label | |
CN107357787B (en) | Semantic interaction method and device and electronic equipment | |
CN111353068A (en) | Video recommendation method and device | |
CN105488154A (en) | Theme application recommendation method and device | |
CN106294787A (en) | Information pushing method and device and electronic equipment | |
JP6440650B2 (en) | User review providing method, apparatus therefor and computer program therefor | |
US10769196B2 (en) | Method and apparatus for displaying electronic photo, and mobile device | |
CN106233282A (en) | Use the application searches of capacity of equipment | |
CN104035927A (en) | User behavior-based search method and system | |
CN106250553A (en) | A kind of service recommendation method and terminal | |
CN103914545A (en) | Search display method and device | |
US20160154891A1 (en) | Intelligent-Predictable Input Method and System | |
CN103984741A (en) | Method and system for extracting user attribute information | |
CN111191133B (en) | Service search processing method, device and equipment | |
US20170046440A1 (en) | Information processing device, information processing method, and program | |
CN104008180A (en) | Association method of structural data with picture, association device thereof | |
CN103207917A (en) | Method for marking multimedia content and method and system for generating recommended content | |
CN112230838A (en) | Article processing method, article processing device, article processing equipment and computer readable storage medium | |
CN108470057B (en) | Generating and pushing method, device, terminal, server and medium of integrated information | |
KR20170107868A (en) | Method and system to recommend music contents by database composed of user's context, recommended music and use pattern | |
CN105574480B (en) | A kind of information processing method, device and terminal |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160907 |
|
WD01 | Invention patent application deemed withdrawn after publication |