CN104424210B - A kind of information recommendation method, system and server - Google Patents
A kind of information recommendation method, system and server Download PDFInfo
- Publication number
- CN104424210B CN104424210B CN201310369757.3A CN201310369757A CN104424210B CN 104424210 B CN104424210 B CN 104424210B CN 201310369757 A CN201310369757 A CN 201310369757A CN 104424210 B CN104424210 B CN 104424210B
- Authority
- CN
- China
- Prior art keywords
- information
- recommendation
- list
- described information
- 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.)
- Active
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
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
-
- 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
Abstract
The invention discloses a kind of information recommendation system, the system includes：Service Processing Unit, relational database and KV databases；The Service Processing Unit, for the historical operating data of described information to be sent to relational database and KV databases in real time；When being additionally operable to receive information recommendation request, the status data of described information is read from the relational database, the historical operating data of described information is read from the KV databases, the recommendation list of described information is calculated according to the historical operating data of described information, the status data of described information and user profile, and returns to the recommendation list；The relational database, for calculating status data according to the historical operating data of described information；The KV databases, for storing the historical operating data of described information.The present invention also discloses a kind of information recommendation method.Using the solution of the present invention, the susceptibility of information recommendation is substantially increased, improves the experience of user.
Description
Technical field
The present invention relates to Internet technology, and in particular to a kind of information recommendation method, system and server.
Background technology
With the development of the Internet, applications, system can record recommended user according to the historical operation of user may be interested
Information, such as song, video or commodity that user may be interested etc..Information recommendation service at present uses off-line calculation to believe
Cease the mode of recommendation list：Commending system is daily in the set time according to user's trimestral full dose historical operation recording gauge recently
The information recommendation list that user may be interested is calculated, Fig. 1 is the composition structural representation of information recommendation system in the prior art,
As shown in figure 1, information recommendation system includes client 11, operation system 12 and relational database 13 in the prior art；Based on existing
Some information systems, its recommendation method comprise the following steps：
The information that client 11 asks to recommend to operation system 12, the recommendation request at the customer in response end 11 of operation system 12,
Information recommendation list is returned into client 11.
Accordingly, information history operation data real-time report is to operation system 12, and is directed into daily by operation system 12
Relational database 13 is once.Such as：Song is recommended in the request of client 11, and operation system 12 records the playback of songs of client 11
Data report relational database, and import the relevant information for playing song in record, such as the singer of song and the stream of song
The information such as group.
Relational database 13 calculates primary information recommendation list, by described in daily according to the historical operating data of the information
Information recommendation list imports operation system 12, recommends when being asked for client 11 to client 11.
The shortcomings that existing information recommendation method is present includes the following aspects：
In existing information recommendation process, information history operation data is imported relational database 13 by operation system 12 daily
Common practices once is the information history operation note for all users for reporting the previous day daily, is limited to relational database 13
Performance, data report process to generally require to expend 2 hours, and relational database 13 calculates user's information interested
Recommendation list, 15 hours are generally required, and described information recommendation list is imported into operation system 12, generally required 3 small
When time, in this way, the process of a recommendation information may need the time of more than one day, that is to say, that ask on the day of user
The information of recommendation information interested before being likely to two days, the susceptibility that can so cause to recommend reduce.If relation data
Storehouse is broken down, and also results in the result of calculation export failure of certain day, can so cause recommendation results that user sees today and
The striking resemblances seen yesterday, it can so influence the experience of user.And proposed algorithm is updated each time, the proving period of effect
Can be very long.
In addition, the number of current information recommendation list restricted information, can only at most recommend 100 information daily, if with
The number that request is recommended on the day of family is excessive, and all information in information recommendation list can then recommend again it is recommended that cross
Information in described information recommendation list, cause user it is seen that the information repeated, influences the experience of user.
The content of the invention
In view of this, it is a primary object of the present invention to provide a kind of information recommendation method, system and server, Neng Goushi
When calculate user's information interested and return to user's recommendation list, lift the experience of user.
To reach above-mentioned purpose, the technical proposal of the invention is realized in this way：
The invention provides a kind of information recommendation system, the system includes：Service Processing Unit, relational database and
Key-value（KV, Key-Value）Database；Wherein,
The Service Processing Unit, for the historical operating data of information to be sent to relational database and KV data in real time
Storehouse；When being additionally operable to receive information recommendation request, the status data of described information is read from the relational database, from described
The historical operating data of described information is read in KV databases, according to the historical operating data of described information, the shape of described information
State data and user profile calculate the recommendation list of described information, and return to the recommendation list；
The relational database, for calculating status data according to the historical operating data of described information；
The KV databases, for storing the historical operating data of described information.
In such scheme, the system also includes user information database, and for storing user profile, the user profile includes
At least one of：The age of user, the sex of user, user location, the occupation of user, the hobby of user.
In such scheme, the Service Processing Unit includes：Operation system module, operation processing module, status data pipe
Manage module and real-time recommendation module；Wherein,
The operation system module, for sending the historical operating data of described information to the operation processing module；Also
The recommendation list of described information for the real-time recommendation module to be sent returns to client；
The operation processing module, for the historical operating data point for the described information for sending the operation system module
Do not send to relational database and KV databases；
The status data management module, for obtaining the status data of described information from the relational database, and
The status data is sent to the real-time recommendation module；
The real-time recommendation module, when being asked for receiving information recommendation, the letter is read from the KV databases
The historical operating data of breath, and the status number sent according to the historical operating data, the status data management module
The recommendation list of described information is calculated according to and from the user profile that is read in the user information database, and by the recommendation of described information
List returns to operation system module.
In such scheme, the real-time recommendation module is according to the historical operating data, the status data management module
The status data sent and the user profile read from the user information database calculate the recommendation list of described information, bag
Include：
Information in the historical operating data is subjected to scoring operations according to the operation attenuation coefficient/weight table, obtained
Obtain the score list of the operation information；
The score list of the operation information is associated with described information similarity, obtain the first of the recommendation list
Procedure list；
The related information of the user is obtained according to the customer attribute information, by the related information added to described the
One procedure list, obtain the second procedure list of the recommendation list；
Second procedure list is associated with described information attribute information, obtain the recommendation list of described information.
In such scheme, described information commending system also includes effect display module；
The relational database, it is additionally operable to calculate the recommendation effect of described information according to the historical operating data of described information
Data, and the effect data is sent to effect display module；
The effect display module, the effect data sent for receiving the relational database, and receiving effect
When checking request, the effect data is shown.
Present invention also offers a kind of information recommendation method, methods described includes：
The historical operating data of information is sent to relational database and KV databases in real time；
When receiving information recommendation request, the status data of described information is read from the relational database, from described
The historical data of described information is read in KV databases；Wherein, the status data of described information by the relational database according to
The historical operating data of described information is calculated；
Described information is calculated according to the historical operating data of described information, the status data of described information and user profile
Recommendation list, and return to the recommendation list.
In such scheme, the status data includes：Operate attenuation coefficient/weight table, information similarity and information attribute
Information.
It is described to be believed according to the historical operating data of described information, the status data of described information and user in such scheme
Breath calculates the recommendation list of described information, including：
Information in the historical operating data is subjected to scoring operations according to the operation attenuation coefficient/weight table, obtained
Obtain the score list of the operation information；
The score list of the operation information is associated with described information similarity, obtain the first of the recommendation list
Procedure list；
The related information of the user is obtained according to the customer attribute information, by the related information added to described the
One procedure list, obtain the second procedure list of the recommendation list；
Second procedure list is associated with described information attribute information, obtain the recommendation list of described information.
In such scheme, methods described also includes：
The relational database calculates the recommendation effect data of described information according to the historical operating data of described information, and
When receiving effect and checking request, the effect data is shown.
In such scheme, the recommendation effect data include：Operation of the user to the information in the recommendation list is complete
Degree.
Information recommendation method, system and server provided by the invention, Service Processing Unit grasp the history of described information
Sent when counting factually to relational database and KV databases；When receiving information recommendation request, according to the relation data
Status data, the historical operating data of the described information read from the KV databases and the user profile that storehouse calculates calculate institute
The recommendation list of information is stated, and returns to the recommendation list；It so, it is possible to calculate recommendation list in real time, returning to user immediately can
Energy information interested, substantially increases the sensitivity of recommendation, improves the experience of user；By the way that historical operating data is real-time
Collect to relational database, effectively raise the efficiency of data summarization so that relational database can quickly calculate recommendation
Effect, whether immediate reaction proposed algorithm is effective, in order to improve proposed algorithm immediately, improves the efficiency of algorithm optimization.
Brief description of the drawings
Fig. 1 is the composition structural representation of information recommendation system in the prior art；
Fig. 2 is the composition structural representation of the information recommendation system of the embodiment of the present invention；
Fig. 3 is the schematic flow sheet of the information recommendation method of the embodiment of the present invention；
Fig. 4 is the schematic diagram of the computational methods of the recommendation list of the embodiment of the present invention.
Embodiment
The present invention basic thought be：Receive client information recommendation request after in real time calculate described information recommendation
List, and to historical operating data and status data separate management, pass through read operation more efficiently KV database purchases history
Operation data, by relational database storage state data, when needing to calculate recommendation list, read respectively from KV databases
Historical operating data, the reads status data from relational database, according to the historical operating data, the status data and
User profile calculates recommendation list.
Below in conjunction with the accompanying drawings and specific embodiment the present invention is further described in more detail.
Fig. 2 is the composition structural representation of the information recommendation system of the embodiment of the present invention, as shown in Fig. 2 described information pushes away
Recommending system mainly includes：Service Processing Unit 21, relational database 22 and KV databases 23；Wherein,
The Service Processing Unit 21, for the historical operating data of described information to be sent to relational database 22 in real time
With KV databases 23；When being additionally operable to receive information recommendation request, the shape of described information is read from the relational database 22
State data, the historical operating data of described information is read from the KV databases 23, according to the historical operation number of described information
The recommendation list of described information is calculated according to, the status data of described information and user profile, and returns to the recommendation list；
The relational database 22, for calculating status data according to the historical operating data of described information；
The KV databases 23, for storing the historical operating data of described information.
Preferably, the system also includes user information database 24, for storing user profile, the user profile include with
It is at least one lower：Province where age of user, sex, user, the occupation of user, the hobby etc. of user.
Preferably, the Service Processing Unit 21 includes：Operation system module 211, operation processing module 212, status number
According to management module 213 and real-time recommendation module 214；Wherein,
The operation system module 211, for sending the historical operation number of described information to the operation processing module 212
According to；The recommendation list for the described information for being additionally operable to send the real-time recommendation module 214 returns to client；
The operation processing module 212, for the historical operation for the described information for sending the operation system module 211
Data are respectively sent to relational database 22 and KV databases 23；
The status data management module 213, for obtaining the status number of described information from the relational database 22
According to, and the status data is sent to the real-time recommendation module 214；
The real-time recommendation module 214, when being asked for receiving information recommendation, institute is read from the KV databases 23
The historical operating data of information is stated, and according to being sent the historical operating data, the status data management module 213
Status data and the user profile that is read from the user information database 24 calculate the recommendation list of described information, and by the letter
The recommendation list of breath returns to operation system module 211.
Preferably, described information commending system also includes effect display module 25,
The relational database 22, the recommendation for being additionally operable to calculate described information according to the historical operating data of described information are imitated
Fruit data, and the effect data is sent to effect display module 25；
The effect display module 25, the effect data sent for receiving the relational database 22, and receiving
When effect checks request, the effect data is shown.
Wherein, the Service Processing Unit 21 of described information commending system in actual applications, can be by the centre in system
Manage device（Central Processing Unit, CPU）Or digital signal processor（Digital Signal Processor,
DSP）Or programmable gate array（Field-Programmable Gate Array, FPGA）Realize；Described information commending system
KV databases 23 and user information database 24 in actual applications, can by system memory realize；Described information is recommended
In actual applications, its data processing function can be real by the CPU in system or DSP or FPGA for the relational database 22 of system
Existing, its store function can be realized by memory；The effect display module 25 of described information commending system in actual applications, can be by
Display is realized.
The embodiment of the present invention additionally provides a kind of server, and the information that the server includes described in the embodiment of the present invention pushes away
Recommend system.
Based on above- mentioned information commending system, the embodiments of the invention provide a kind of information recommendation method, as shown in figure 3, Fig. 3
For the schematic flow sheet of the information recommendation method of the embodiment of the present invention, comprise the following steps：
Step 301：The historical operating data of information is sent to relational database in real time.
The status data of described information is calculated according to the historical operating data of described information for the relational database.This
In, the status data includes：Operate attenuation coefficient/weight table, information similarity and information attribute information.
Specifically, the relational database can according to data in the historical operating data generation time and user it is inclined
The attenuation coefficient/weight table is set well, and the time caused by data is more early, and coefficient/weighted value is lower, on the contrary then coefficient/weight
Value is higher；The number of information operation is more, and coefficient/weighted value is higher, on the contrary then coefficient/weighted value is lower；Described information is similar
Spend and be more than zero decimal for being less than one for one, more close closer to 1, the numerical value of described information similarity can be according to being institute in advance
The label for stating information setting obtains, and is that described information sets label according to the attribute information of described information, and identical label is more,
It is then more similar, it is that music sets label, such as label to be arranged to according to the attribute information of music by taking music as an example：85-95
Generation, piano, lyric etc.；Described information attribute information can obtain according to the historical operating data of user, such as going through according to user
History operation data obtains the information attribute liked and do not liked of user, and exemplified by recommending music, described information attribute information can
The attribute informations such as the languages of school, music including music, the singer of music.
Step 302~step 303：When receiving information recommendation request, described information is read from the relational database
Status data, the historical data of described information is read from the KV databases, according to the status data of described information, described
The historical operating data and user profile of information calculate the recommendation list of described information and return to recommendation list.
Preferably, it is described that institute is calculated according to the status data of information, the historical operating data of described information and user profile
The recommendation list of information is stated, including：
Information in the historical operating data is subjected to scoring operations according to the operation attenuation coefficient/weight table, obtained
Obtain the score list of the operation information；
The score list of the operation information is associated with described information similarity, obtain the first of the recommendation list
Procedure list；
The related information of the user is obtained according to the customer attribute information, by the related information added to described the
One procedure list, obtain the second procedure list of the recommendation list；
Second procedure list is associated with described information attribute information, obtain the recommendation list of described information.
The implementation process of information recommendation method of the embodiment of the present invention is made into one so that user asks to recommend song as an example below
Step detailed description.
User logs in client by computer or mobile phone, such as logs in QQ music clients end, the operation system module to
Operation processing module sends the historical operating data of the user in real time, including music, the music etc. of search listened to, at operation
The historical operating data is respectively sent to relational database and KV databases by reason module；
The relational database calculates status data according to the historical operating data by preset rules, specifically, described
Status data includes：Operate attenuation coefficient/weight table, information similarity and information attribute information etc..The relational database
Attenuation coefficient/weight the table, number can be set according to the generation time of data in the historical operating data and the preference of user
More early according to the caused time, coefficient/weighted value is lower, on the contrary then coefficient/weighted value is higher；What identical music was listened to or searched for
Number is more, and coefficient/weighted value is higher, on the contrary then coefficient/weighted value is lower；Described information similarity is one and is less than more than zero
One decimal, more close closer to 1, the numerical value of described information similarity can obtain according to the label set in advance for described information
Take, according to the attribute information of described information be described information set label, identical label is more, then more similar, using music as
Example, it is that music sets label, such as label to be arranged to according to the attribute information of music：85-95 ages, piano, lyric etc.；It is described
Information attribute information can obtain according to the historical operating data of user, for example obtain user's according to the historical operating data of user
The information attribute liked and do not liked, including the attribute information such as singer of the languages of the school of music, music, music.
The KV database purchases have had the historical operating data of user；
When user, which logs in client request, recommends music, the music that client sends the user to operation system module pushes away
Request is recommended, real-time recommendation module calculates the music recommendation list of the user；Status data management module is from the relational database
The middle status data for reading the user, the status data include operation attenuation coefficient/weight table, information similarity, information category
Property information etc., the real-time recommendation module is sent to by the status data；The real-time recommendation module is from the KV data
The historical operating data of the user is read in storehouse, according to the user of the historical operating data, the status data and the user
Information calculates recommendation list.Specifically, Fig. 4 is the schematic diagram of the computational methods of the recommendation list of the embodiment of the present invention, such as Fig. 4 institutes
Show, comprise the following steps：
The real-time recommendation module is beaten the song in historical operating data according to operation attenuation coefficient/weight table
Point, obtain the score list of operation song；The score list of the operation song is associated with information similarity, recommended
List 1；
If the recommendation list 1 is few for sky, or the number of songs in the recommendation list 1, then believed according to user property
Breath obtains the song that user may like, such as according to the information such as the age of user, the occupation of user, user location, retrieval
The other users similar with active user, the song that the other users are liked is added in the recommendation list of active user,
Obtain recommendation list 2；
The recommendation list 2 is associated with information attribute information, and described information attribute information is not liked including active user
Joyous song attributes information, including：The school of music, the languages of music, the attribute information such as singer of music, if active user
The music of a certain singer is not liked, then is deleted the music of the singer from the recommendation list 2；In addition, for working as
The music that preceding user had operated recently, while deleted, to recommend new music to user；Delete corresponding music
Afterwards, recommendation list 3 is obtained；
The recommendation list 3 is associated with information attribute information 2, and described information attribute information is the same as the letter in previous step
It is identical to cease attribute information, it is therefore intended that the music in the recommendation list 3 is adjusted operation, adjusts the recommendation list 3
In occur without the music of identical singer in continuous at least ten songs, corresponding to adjust after operation terminates, acquisition finally pushes away
Recommend list.
After the real-time recommendation module calculates the consequently recommended list of acquisition, the recommendation list is back to the business system
System module, and client is returned to by the operation system module in time.
Preferably, the step also includes：The relational database is according to calculating the historical operating data of described information
The recommendation effect data of information, and when receiving effect and checking request, show the effect data.
Preferably, the recommendation effect data include：Operation integrity degree of the user to the information in the recommendation list.
Here, the recommendation effect data may include a variety of data, exemplified by recommending music, the letter in the recommendation list
The operation integrity degree of breath can be user to be listened to ratio, averagely listens to duration, flat to the complete of the music in music recommendation list
Listen to number, 60% accounting for listening to music, 30% accounting for listening to music etc.；Exemplified by recommending video, then the recommendation
Information operation integrity degree in list can be user to be watched ratio, averagely watches to the complete of the video in video recommendations list
Duration, averagely watch the data such as number；It can be seen that whether the information recommended is liked by user from these data, so as to
Reflect whether current proposed algorithm is effective.
Specifically, within a period of time, the relational database calculates music according to the historical operating data is received
Recommendation effect data, the effect data is sent to effect display module, and the effect data is checked in attendant
When, show the effect data.The recommendation effect data are specifically reflected as the recommendation effect data to current proposed algorithm, than
Such as user completely listens to ratio on the music recommendation page, averagely listens to duration, averagely listens to number, from these data
It can be seen that whether the music recommended is liked by user, so as to reflect whether current proposed algorithm is effective.
Table 1 calculates the recommendation effect list obtained to be actual, and as shown in table 1, what player deep bid represented is all in QQ
The old song form of listening at music client end is；What real time correlation was recommended to represent is in the case where user has operation history, for recommending sound
Happy feedback；Real-time property recommend represent be user in the case of no historical operating data, for recommend music it is anti-
Feedback.From table 1 it follows that by completely listening song ratio, 30% accounting listened to 60% accounting of song, listen to song, being averaged
To listen song duration, averagely listen these indexs such as the first number of song, can react whether user is liked the music of recommendation, numeral is bigger,
Illustrate that user more likes.
Table 1
If the information recommendation method described in the embodiment of the present invention is realized in the form of software function module and as independently
Production marketing or in use, can also be stored in a computer read/write memory medium.Based on such understanding, this hair
The part that the technical scheme of bright embodiment substantially contributes to prior art in other words can in the form of software product body
Reveal and, the computer software product is stored in a storage medium, including some instructions are causing a computer to set
It is standby（Can be personal computer, server or network equipment etc.）Perform the whole of each embodiment methods described of the present invention
Or part.And foregoing storage medium includes：USB flash disk, mobile hard disk, read-only storage（ROM, Read-Only Memory）, with
Machine accesses memory（RAM, Random Access Memory）, magnetic disc or CD etc. are various can be with Jie of store program codes
Matter.So, the embodiment of the present invention is not restricted to any specific hardware and software combination.
Accordingly, the embodiment of the present invention also provides a kind of computer-readable storage medium, wherein computer program is stored with, the meter
Calculation machine program is used for the information recommendation method for performing the embodiment of the present invention.
The foregoing is only a preferred embodiment of the present invention, is not intended to limit the scope of the present invention.It is all
All any modification, equivalent and improvement made within the spirit and scope of the present invention etc., it is all contained in the protection model of the present invention
Within enclosing.
Claims (8)
1. a kind of information recommendation system, it is characterised in that the system includes：Service Processing Unit, relational database and key-value
(KV) database；Wherein,
The Service Processing Unit, for the historical operating data of information to be sent to relational database and KV databases in real time；
When being additionally operable to receive information recommendation request, the status data of described information is read from the relational database, from the KV
The historical operating data of described information is read in database, according to the historical operating data of described information, the state of described information
Data and user profile calculate the recommendation list of described information, and return to the recommendation list；
The relational database, for calculating status data according to the historical operating data of described information；
The KV databases, for storing the historical operating data of described information；
Wherein, the Service Processing Unit, when being asked for receiving information recommendation, the letter is read from the KV databases
The historical operating data of breath, the information in the historical operating data is beaten according to the operation attenuation coefficient/weight table
Divide operation, obtain the score list of the operation information；By the score list of the operation information and described information similarity phase
Association, obtain the first procedure list of the recommendation list；The association that the user is obtained according to the customer attribute information is believed
Breath, is added to first procedure list by the related information, obtains the second procedure list of the recommendation list；By described in
Second procedure list is associated with described information attribute information, obtains the recommendation list of described information, and returns to the recommendation row
Table.
2. information recommendation system according to claim 1, it is characterised in that the system also includes user information database, uses
In storage user profile, the user profile includes at least one of：The age of user, the sex of user, user location,
The occupation of user, the hobby of user.
3. information recommendation system according to claim 1, it is characterised in that the Service Processing Unit includes：Business system
System module, operation processing module, status data management module and real-time recommendation module；Wherein,
The operation system module, for sending the historical operating data of described information to the operation processing module；It is additionally operable to
The recommendation list for the described information that the real-time recommendation module is sent returns to client；
The operation processing module, the historical operating data of the described information for the operation system module to be sent are sent out respectively
Deliver to relational database and KV databases；
The status data management module, for obtaining the status data of described information from the relational database, and by institute
State status data and be sent to the real-time recommendation module；
The real-time recommendation module, when being asked for receiving information recommendation, described information is read from the KV databases
Historical operating data, the information in the historical operating data is subjected to marking behaviour according to the operation attenuation coefficient/weight table
Make, obtain the score list of the operation information；The score list of the operation information is associated with described information similarity,
Obtain the first procedure list of the recommendation list；The related information of the user is obtained according to the customer attribute information, will
The related information is added to first procedure list, obtains the second procedure list of the recommendation list；By described second
Procedure list is associated with described information attribute information, obtains the recommendation list of described information, and return the recommendation list to
The operation system module.
4. information recommendation system according to claim 1, it is characterised in that described information commending system also shows including effect
Show module；
The relational database, it is additionally operable to calculate the recommendation effect number of described information according to the historical operating data of described information
According to, and the effect data is sent to effect display module；
The effect display module, the effect data sent for receiving the relational database, and checked receiving effect
During request, the effect data is shown.
5. a kind of information recommendation method, it is characterised in that methods described includes：
The historical operating data of information is sent to relational database and key-value (KV) database in real time；
When receiving information recommendation request, the status data of described information is read from the relational database, from the KV numbers
According to the historical data that described information is read in storehouse；Wherein, the status data of described information as the relational database according to
The historical operating data of information is calculated；
The recommendation of described information is calculated according to the historical operating data of described information, the status data of described information and user profile
List, and return to the recommendation list；
Wherein, it is described according to calculating the historical operating data of described information, the status data of described information and user profile
The recommendation list of information, including：Information in the historical operating data is carried out according to the operation attenuation coefficient/weight table
Scoring operations, obtain the score list of the operation information；
The score list of the operation information is associated with described information similarity, obtain the first process of the recommendation list
List；
The related information of the user is obtained according to the customer attribute information, the related information is added to first mistake
Cheng Liebiao, obtain the second procedure list of the recommendation list；
Second procedure list is associated with described information attribute information, obtain the recommendation list of described information.
6. information recommendation method according to claim 5, it is characterised in that the status data includes：Operation decay system
Number/weight table, information similarity and information attribute information.
7. according to the method for claim 5, it is characterised in that methods described also includes：
According to the historical operating data of described information calculate described information recommendation effect data, and receive effect check please
When asking, the effect data is shown；Historical operation of the recommendation effect data by the relational database according to described information
Data, which calculate, to be obtained.
8. information recommendation method according to claim 7, it is characterised in that the recommendation effect data include：User couple
The operation integrity degree of information in the recommendation list.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310369757.3A CN104424210B (en) | 2013-08-22 | 2013-08-22 | A kind of information recommendation method, system and server |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310369757.3A CN104424210B (en) | 2013-08-22 | 2013-08-22 | A kind of information recommendation method, system and server |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104424210A CN104424210A (en) | 2015-03-18 |
| CN104424210B true CN104424210B (en) | 2017-11-14 |
Family
ID=52973209
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310369757.3A Active CN104424210B (en) | 2013-08-22 | 2013-08-22 | A kind of information recommendation method, system and server |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104424210B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106708829B (en) * | 2015-07-31 | 2022-05-10 | 腾讯科技（深圳）有限公司 | Data recommendation method and recommendation system |
| CN105512934A (en) * | 2015-10-22 | 2016-04-20 | 深圳怡化电脑股份有限公司 | Method and device for inputting business parameters |
| CN106803178B (en) * | 2015-11-26 | 2020-09-18 | 阿里巴巴集团控股有限公司 | Method and equipment for processing entity |
| CN106844359A (en) * | 2015-12-04 | 2017-06-13 | 深圳富泰宏精密工业有限公司 | Server and its music service system and method |
| CN105959374B (en) | 2016-05-12 | 2019-05-03 | 腾讯科技（深圳）有限公司 | A kind of data recommendation method and its equipment |
| CN107451141B (en) * | 2016-05-30 | 2021-01-29 | 阿里巴巴集团控股有限公司 | Data recommendation processing interaction method, device and system |
| CN106210127B (en) * | 2016-08-15 | 2019-07-16 | 腾讯科技（深圳）有限公司 | A kind of information processing method, server and client |
| CN107180086A (en) * | 2017-05-09 | 2017-09-19 | 中国石油集团川庆钻探工程有限公司 | A kind of drilling well real time data quick storage and dissemination method |
| CN108076353A (en) * | 2017-05-18 | 2018-05-25 | 北京市商汤科技开发有限公司 | Business object recommends method, apparatus, storage medium and electronic equipment |
| CN107547740A (en) * | 2017-08-28 | 2018-01-05 | 江西博瑞彤芸科技有限公司 | The management method and system of station list |
| CN110019925A (en) * | 2017-12-07 | 2019-07-16 | 北京雷客天地科技有限公司 | Batch deletes the method and device of song |
| CN111291018B (en) * | 2018-12-07 | 2023-06-23 | 北京沃东天骏信息技术有限公司 | Data management method, device, equipment and storage medium |
| CN116828535B (en) * | 2023-08-30 | 2023-11-14 | 太一云境技术有限公司 | Audio sharing method and system based on wireless transmission |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102332006A (en) * | 2011-08-03 | 2012-01-25 | 百度在线网络技术（北京）有限公司 | Information push control method and device |
| CN102654860A (en) * | 2011-03-01 | 2012-09-05 | 北京彩云在线技术开发有限公司 | Personalized music recommendation method and system |
| CN103049865A (en) * | 2012-12-17 | 2013-04-17 | 中国农业大学 | Method and system for initiatively recommending product information service |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110125753A1 (en) * | 2009-11-20 | 2011-05-26 | Rovi Technologies Corporation | Data delivery for a content system |
-
2013
- 2013-08-22 CN CN201310369757.3A patent/CN104424210B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102654860A (en) * | 2011-03-01 | 2012-09-05 | 北京彩云在线技术开发有限公司 | Personalized music recommendation method and system |
| CN102332006A (en) * | 2011-08-03 | 2012-01-25 | 百度在线网络技术（北京）有限公司 | Information push control method and device |
| CN103049865A (en) * | 2012-12-17 | 2013-04-17 | 中国农业大学 | Method and system for initiatively recommending product information service |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104424210A (en) | 2015-03-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104424210B (en) | A kind of information recommendation method, system and server | |
| US9483730B2 (en) | Hybrid review synthesis | |
| US10176170B2 (en) | Systems for dynamically generating and presenting narrative content | |
| US10445809B2 (en) | Relationship discovery engine | |
| US10275782B2 (en) | Variation of minimum advertisement relevance quality threshold based on search query attributes | |
| US7542929B2 (en) | Increases in sales rank as a measure of interest | |
| CN105765573B (en) | Improvements in website traffic optimization | |
| US9589277B2 (en) | Search service advertisement selection | |
| US10025807B2 (en) | Dynamic data acquisition method and system | |
| US9679018B1 (en) | Document ranking based on entity frequency | |
| CN108256119A (en) | A kind of construction method of resource recommendation model and the resource recommendation method based on the model | |
| CN109408665A (en) | A kind of information recommendation method and device, storage medium | |
| US20090313227A1 (en) | Searching Using Patterns of Usage | |
| US10152478B2 (en) | Apparatus, system and method for string disambiguation and entity ranking | |
| CN107885745A (en) | A kind of song recommendations method and device | |
| JP2013168186A (en) | Review processing method and system | |
| CN104756504A (en) | Methods and apparatus to estimate demographics of users employing social media | |
| EP2513852A2 (en) | Method and system for automatically identifying related content to an electronic text | |
| CN102298750B (en) | The method and device of playback is clicked on for advertisement search | |
| CN110795613A (en) | Commodity searching method, device and system and electronic equipment | |
| CN112990986B (en) | Advertisement service information publishing system based on block chain | |
| CN106940723A (en) | A kind of news search method and device | |
| US11487835B2 (en) | Information processing system, information processing method, and program | |
| CN108416610B (en) | User history feedback information forming method and advertisement putting frequency control method | |
| US20180349372A1 (en) | Media item recommendations based on social relationships |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20150318 Assignee: Ocean interactive (Beijing) Information Technology Co., Ltd. Assignor: Tencent Technology (Shenzhen) Co., Ltd. Contract record no.: 2016990000422 Denomination of invention: Information recommendation method, information recommendation system and server License type: Common License Record date: 20161009 |
|
| LICC | Enforcement, change and cancellation of record of contracts on the licence for exploitation of a patent or utility model | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |