CN107862085A - A kind of music supplying system based on big data - Google Patents
A kind of music supplying system based on big data Download PDFInfo
- Publication number
- CN107862085A CN107862085A CN201711239611.1A CN201711239611A CN107862085A CN 107862085 A CN107862085 A CN 107862085A CN 201711239611 A CN201711239611 A CN 201711239611A CN 107862085 A CN107862085 A CN 107862085A
- Authority
- CN
- China
- Prior art keywords
- music
- user
- supplying system
- song
- big data
- 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
-
- 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
- G06F16/68—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/686—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, title or artist information, time, location or usage information, user ratings
Abstract
A kind of music supplying system based on big data of the present application, the mass data provided based on big data platform, identify label, classification and the user crowd of different music according to many algorithms, and use habit according to user, listen to custom and it is expected listen to impression and carry out more accurately music push.
Description
Technical field
The present invention relates to big data to analyze and process field, and in particular to a kind of music supplying system based on big data.
Background technology
In internet, continuous development is present, and people often receive various message push.Each commercial podium
Largely analyzed also for different users, specific message is sent to specific crowd.But in music field, to user's
Also there are many problems in music push, not a good algorithm, divide after the various information of user is carried out into unified quantization
Analysis is handled, and is then targetedly pushed, the general preliminary classification basis remained in crowd of current music push
On.
For this problem, a kind of music supplying system based on big data of the present application is main to consider music push
Precision is high, subtracts the problems such as single process of cumbersome addition song, lifting Consumer's Experience, and different music are identified according to many algorithms
Label, classification and user crowd, and use habit according to user, listen to custom and it is expected listen to impression carry out it is more accurate
Music push.Music supplying system of the present invention can be merged, calculated using the big scale of construction data radix of big data, divided
Analysis, the features such as plus application software, hardware and software platform, provide the user and more comfortable easily service.
The content of the invention
Music supplying system of the present invention can identify label, classification and the user of different music according to many algorithms
Crowd, the use of foundation user, listen to and be accustomed to and be expected to listen to experience more accurately to push relevant musical.
The basic implementation of push algorithm has following three kinds at present：
I) content-based recommendation (content-based filtering)
Content-based recommendation refers to the retrieval of music information herein.Academicly general alleged content-based is
Refer in particular to audio content in itself, relate generally to feature extraction, special edition, singer and the lyrics etc. and be based on text or tags
Factor, commonly used to be combined with content to improve recall precision.
II) it is based on collaborative filtering recommending (collaboration filtering)
It is that the ranking list row based on broad sense is recommended with popular seniority among brothers and sisters based on collaborative filtering recommending.
III) (social recommendation) is recommended in socialization.
It is a kind of recommendation based on relation that socialization, which is recommended,.
Music supplying system of the present invention is concentrated mainly on to third-party application information to the improvement for pushing algorithm
In terms of acquisition, latency algorithm, the improvement of cosine-algorithm scheduling algorithm.Its structure design is primary concern is that improve music push essence
Accuracy, Consumer's Experience is lifted while subtracting cumbersome addition and singing single process.
Label, classification and the user crowd of the invention that different music are identified according to many algorithms, the use of foundation user,
Listen to custom and be expected to listen to experience and more accurately push relevant musical.Wherein algorithm includes obtaining for third-party application information
Take, potential element algorithm, Boundary algorithm and association option.Specific implementation is：
1st, user can be listened song type, listen to the information such as number using initial stage, system according to the inquiry input content of user
Related song is pushed for user, and user profile is saved in cloud database.
2nd, system transfers music sources according to the user profile preserved in database for user.
3rd, support that one-touch addition similar songs song is single, it is simple and efficient.
4th, music system management platform is built in data center, and data center provides and safeguarded for the unified of database
And data backup, improve music supplying system maintainability and reliability of the present invention.
Specifically, a kind of music supplying system based on big data is claimed in the application, it is characterised in that the music pushes away
Send system push song track method be：
User preferences are obtained by the database of third-party application, as the reference pushed first；
Listen to the element included in music when in use by user, using potential element algorithm, show that user likes
Types of songs included in element species, by it with Boundary algorithm, compared with the song that association option is recommended；
Preferential push has the song track of coincidence more afterwards, and secondly push meets the song track of individual event algorithm.
Music supplying system based on big data as described above, is further characterized in that, all sounds of music supplying system
Happy push be all after being calculated the relevant information of user, searched in database, analyzing preserve it is same to the user
Catalogue, and data message is constantly updated in use, ensure that push is accurate reachable.
Music supplying system based on big data as described above, is further characterized in that, the potential element algorithm is to sing
Qu Yuansuization, labeling, and be attached in song information；Then quantum chemical method goes out preference of the user to different elements, enters
And adjust push frequency.
Music supplying system based on big data as described above, is further characterized in that, the music supplying system can be with
Including song information input function.
Music supplying system based on big data as described above, is further characterized in that, the music supplying system supports one
Keyed addition similar songs song is single.
Music supplying system based on big data as described above, is further characterized in that, the music supplying system can be with
Registered including user's real name.
Brief description of the drawings
Fig. 1, music supplying system algorithm schematic diagram
Embodiment
The specific content of the invention is as follows：
1) acquisition of third-party application information：Interface can be set between third-party application and the system, you can by with
The connection software JDBC of database, the database information of third-party application is obtained, pushed as user in the initialization of the system
Information source；
2) potential element algorithm：
A) by song element, labeling, isolate similar to rock and roll, prevalence, march, piano music, it is sad, glad,
The elements such as absolute music, and be attached in song information；If user has element in song label, then counted when selecting song
This element, and occurrence number+1 can be recorded according to storehouse；
B) user-latent factor matrix Q, represents the preference of different user for different elements, and 1 represents and likes very much
Vigorously, 0 representative does not like.
| Elements A | Element B | Elements C | Element D | Element E | |
| User X | 1 | 0.9 | 0.6 | 0.7 | 0 |
Music matrix P, represent the composition that every kind of music contains various elements
| Elements A | Element B | Elements C | Element D | Element E | |
| Music Y | 0.8 | 0.9 | 0.6 | 0.7 | 0.5 |
It can show that user X likes degree to music Y by upper 2 table：
1*0.8+0.9*0.9+0.6*0.6+0.7*0.7+0*0.5=2.46
Each user, which is so calculated every song, can obtain rating matrix M of the different user to different songs, by upper
The algorithm in face can draw the music type that user likes, and be expressed in matrix as
M=QPT
C a quantitative criteria) is formulated：Single circulation=5, share=4, collection=3, actively play=2, hear out=1, jump
=-2 are crossed, draws black=- 5, obtainable actual rating matrix M in analysis, that is, input matrix is probably this appearance：
| Music 1 | Music 2 | Music 3 | |
| User 1 | 5 | 4 | -2 |
| User 2 | 3 | 5 | -5 |
Thus scoring Ms of the user X to music Y can be obtained, a point higher explanation user more likes this music, i.e. increase accordingly pushes away
Send frequency.
3) Boundary algorithm and association option：
A) Boundary algorithm：If user N likes music A, B, C, user M likes music E, F, G, if A is similar to M hobby
(identical element is more), then it can attempt E, F, G being pushed to N user, A, B, C are pushed to M user；
B) association option：If user N likes music A, B, C, user M likes music B, C, G, then can attempt to push G
To N user, A is pushed to M user, that is, likes X user also to like Y；
Embodiment：
1) premise for building this system must be based on possessing magnanimity music data, user data big data platform branch
Hold, all music, user push all must in database to all relevant informations carry out for several times calculate, search for (user listens to
Song is more, and push song type is more accurate), and data inputting data base directory is provided, bind, preserve extremely with user after analysis
Same catalogue, and data message is constantly updated in use, ensure that push is accurate reachable；
2) user can be obtained user preferences by the database of third-party application and (still catch music using initial stage
Element), as the reference recommended first；
3) user when in use, by listening to the element included in music, using potential element algorithm, show that user likes
Element species included in the types of songs of love, and preferentially pushed away compared with the song that association option is recommended with Boundary algorithm
Coincidence song is recommended, secondly recommends the song for meeting individual event algorithm；
4) music supplying system can also include friendly man-machine interface interaction design, song information input (title, music
Type) function, user can real name registration, include the information such as name, ID card No., photo；
5) music supplying system supports that one-touch addition similar songs song is single.
It should be evident that illustrated above is only the specific embodiment of the present invention, for the common skill in this area
For art personnel, on the premise of not paying creative work, other technical schemes can also be obtained according to above-described embodiment,
And the equivalent variations made in the scope of protection of the invention all should be fallen within the scope of protection of the present invention, and belong to the present invention
The scope of protection.
In summary, the mass data that music supplying system of the present invention is provided based on big data platform, according to more
The algorithm of kind identifies label, classification and the user crowd of different music, the use of foundation user, listens to custom and expected receipts
Impression is listened more accurately to push relevant musical.
Claims (6)
1. a kind of music supplying system based on big data, it is characterised in that the music supplying system pushes the side of song track
Method is：
User preferences are obtained by the database of third-party application, as the reference pushed first；
Listen to the element included in music when in use by user, using potential element algorithm, draw the song that user likes
Element species included in bent type, by it with Boundary algorithm, compared with the song that association option is recommended；
Preferential push has the song track of coincidence more afterwards, and secondly push meets the song track of individual event algorithm.
2. the music supplying system based on big data as claimed in claim 1, is further characterized in that, the music supplying system institute
The push for having music is all after being calculated the relevant information of user, searched in database, analyzing, to preserve to the user's
Same catalogue, and data message is constantly updated in use, ensure that push is accurate reachable.
3. the music supplying system based on big data as claimed in claim 2, is further characterized in that, the potential element algorithm is
By song element, labeling, and it is attached in song information；Then quantum chemical method goes out preference journey of the user to different elements
Degree, and then adjust push frequency.
4. the music supplying system based on big data as claimed in claim 3, is further characterized in that, the music supplying system is also
Song information input function can be included.
5. the music supplying system based on big data as claimed in claim 4, is further characterized in that, the music supplying system branch
It is single to hold one-touch addition similar songs song.
6. the music supplying system based on big data as claimed in claim 5, is further characterized in that, the music supplying system is also
User's real name can be included to register.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711239611.1A CN107862085A (en) | 2017-11-30 | 2017-11-30 | A kind of music supplying system based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711239611.1A CN107862085A (en) | 2017-11-30 | 2017-11-30 | A kind of music supplying system based on big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107862085A true CN107862085A (en) | 2018-03-30 |
Family
ID=61704239
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711239611.1A Pending CN107862085A (en) | 2017-11-30 | 2017-11-30 | A kind of music supplying system based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107862085A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108876517A (en) * | 2018-05-31 | 2018-11-23 | 周国俊 | User preferences analysis method and system based on artificial intelligence |
| CN109740016A (en) * | 2019-01-03 | 2019-05-10 | 百度在线网络技术（北京）有限公司 | Method, apparatus, server and the computer readable storage medium of music query |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104991900A (en) * | 2015-06-09 | 2015-10-21 | 腾讯科技（深圳）有限公司 | Method and apparatus for pushing music data |
| CN106227816A (en) * | 2016-07-22 | 2016-12-14 | 北京小米移动软件有限公司 | Push the method and device that song is single |
| CN106557560A (en) * | 2016-11-11 | 2017-04-05 | 天翼爱音乐文化科技有限公司 | Level music based on user interest recommends method |
-
2017
- 2017-11-30 CN CN201711239611.1A patent/CN107862085A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104991900A (en) * | 2015-06-09 | 2015-10-21 | 腾讯科技（深圳）有限公司 | Method and apparatus for pushing music data |
| CN106227816A (en) * | 2016-07-22 | 2016-12-14 | 北京小米移动软件有限公司 | Push the method and device that song is single |
| CN106557560A (en) * | 2016-11-11 | 2017-04-05 | 天翼爱音乐文化科技有限公司 | Level music based on user interest recommends method |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108876517A (en) * | 2018-05-31 | 2018-11-23 | 周国俊 | User preferences analysis method and system based on artificial intelligence |
| CN108876517B (en) * | 2018-05-31 | 2022-06-17 | 周国俊 | User preference analysis method and system based on artificial intelligence |
| CN109740016A (en) * | 2019-01-03 | 2019-05-10 | 百度在线网络技术（北京）有限公司 | Method, apparatus, server and the computer readable storage medium of music query |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103049440B (en) | A kind of recommendation process method of related article and disposal system | |
| CN104281622B (en) | Information recommendation method and device in a kind of social media | |
| CN104899302A (en) | Method and device for recommending music to user | |
| US20140245147A1 (en) | Active Playlist Having Dynamic Media Item Groups | |
| CN102654859A (en) | Method and system for recommending songs | |
| CN102566945B (en) | Method and system for realizing automatic acquisition and on-demand printing of book | |
| CN105868267B (en) | A kind of modeling method of mobile social networking user interest | |
| CN102654860A (en) | Personalized music recommendation method and system | |
| CN107977373A (en) | A kind of recommendation method of song | |
| CN102207948A (en) | Method for generating incident statement sentence material base | |
| CN103390044A (en) | Method and device for identifying linkage type POI (Point Of Interest) data | |
| US20180225305A1 (en) | Method for displaying landmark data | |
| CN108038506A (en) | A kind of library automatic classification method | |
| CN107862085A (en) | A kind of music supplying system based on big data | |
| TW202135094A (en) | A data collection method and system | |
| CN109829138A (en) | File comparison method, device, electronic equipment and computer readable storage medium | |
| CN105404681A (en) | Live broadcast sentiment classification method and apparatus | |
| CN108717445A (en) | A kind of online social platform user interest recommendation method based on historical data | |
| Knudsen et al. | Substance abuse treatment counselors and tobacco use: a comparison of comprehensive and indoor-only workplace smoking bans | |
| Oliveira et al. | Musical success in the United States and Brazil: Novel datasets and temporal analyses | |
| JP2013130921A (en) | Information processing device, information processing method, and information processing program | |
| KR101738771B1 (en) | Music classifying method based on melody and recommendation method of melody based on music database | |
| US11449901B1 (en) | System and method for identifying content relevant to a user based on gathering contextual information from music and music player environmental factors | |
| CN110417714A (en) | A kind of multimedia communication based on live scene and network real-time interactive system | |
| Lamprianidis et al. | Extraction, integration and exploration of crowdsourced geospatial content from multiple web sources |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180330 |