CN107609198A - One kind recommends method, apparatus and computer-readable recording medium - Google Patents
One kind recommends method, apparatus and computer-readable recording medium Download PDFInfo
- Publication number
- CN107609198A CN107609198A CN201710986686.XA CN201710986686A CN107609198A CN 107609198 A CN107609198 A CN 107609198A CN 201710986686 A CN201710986686 A CN 201710986686A CN 107609198 A CN107609198 A CN 107609198A
- Authority
- CN
- China
- Prior art keywords
- scene
- recommendation
- generalization bounds
- digital content
- recommended
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of recommendation method, including：Obtain the relevant information that can determine scene；Using the relevant information, determine that target recommends scene；Default corresponding relation based on each recommendation scene and Generalization bounds, determine that the target recommends Generalization bounds corresponding to scene；Wherein, the single Generalization bounds include：Correspond respectively to the results set to be recommended of the digital content of at least two types；In the Generalization bounds that the default corresponding relation includes, the information recommendation algorithm that results set institute to be recommended foundation corresponding at least two Generalization bounds difference be present is different；The results set to be recommended that Generalization bounds according to determining include, information recommendation is carried out to user.The present invention further simultaneously discloses a kind of recommendation apparatus and computer-readable recording medium.
Description
Technical field
The present invention relates to intelligent recommendation field, more particularly to a kind of recommendation method, apparatus and computer-readable recording medium.
Background technology
At present, with the development of computer technology, information content is increasing, by recommending to determine effective information increasingly
It is common.During actual recommendation, in intelligent recommendation field, especially digital content recommending field, it can be disposed for each scene a set of
Proposed algorithm, proposed algorithm iteration is carried out according to the effect of recommendation, it is logical to continue to optimize the purpose that proposed algorithm reaches recommendation.
But this way of recommendation does not consider that same user can face different scenes, and same user exists
The problem of preference under different scenes may shift.
The content of the invention
In view of this, the embodiment of the present invention it is expected to provide a kind of recommendation method, apparatus and computer-readable recording medium, energy
Information recommendation is enough carried out to user based on scene.
What the technical scheme of the embodiment of the present invention was realized in：
The embodiment of the present invention provides a kind of recommendation method, and methods described includes：
Obtain the relevant information that can determine scene；
Using the relevant information, determine that target recommends scene；
Default corresponding relation based on each recommendation scene and Generalization bounds, determine that the target recommends to recommend corresponding to scene
Strategy；Wherein, the single Generalization bounds include：Correspond respectively to the result set to be recommended of the digital content of at least two types
Close；In the Generalization bounds that the default corresponding relation includes, result to be recommended corresponding at least two Generalization bounds difference be present
The information recommendation algorithm for gathering institute's foundation is different；
The results set to be recommended that Generalization bounds according to determining include, information recommendation is carried out to user.
In such scheme, the acquisition can determine the relevant information of scene, including：
Judge whether default migration efficiency；
It is determined that when default migration efficiency is not present, the relevant information that can determine scene is obtained.
In such scheme, methods described also includes：
When it is determined that default migration efficiency be present, based on default migration efficiency, it is determined that corresponding with the default migration efficiency
Results set to be recommended.
It is described to utilize the relevant information in such scheme, determine that target recommends scene, including：
Using the relevant information, the type of user is determined；
The type of the user of the determination information related to user is combined, determines that target recommends scene.
In such scheme, after determining that the target recommends Generalization bounds corresponding to scene, methods described also includes：
Recommend from the target of determination in the recommendation results set that Generalization bounds corresponding to scene include, determine each type number
The information recommendation ratio of word content；
Information recommendation ratio based on determination is recommended.
In such scheme, the default corresponding relation based on each recommendation scene and Generalization bounds, determine that the target recommends field
Generalization bounds corresponding to scape, including：
Whether the target for judging to determine recommends scene and historic scenery identical；
It is corresponding with history Generalization bounds from historic scenery when it is determined that identical and determination has recommendation for historic scenery
History Generalization bounds corresponding to searching historic scenery in relation, the history Generalization bounds of lookup are pushed away as the target of the determination
Recommend the Generalization bounds of scene；When it is determined that when different, the default corresponding relation based on each recommendation scene and Generalization bounds, it is determined that described
Target recommends Generalization bounds corresponding to scene.
In such scheme, methods described also includes：
Obtain the historical feedback information of user；
Using the historical feedback information, judge whether that being preferentially based on scene determines Generalization bounds；
When it is determined that determining Generalization bounds based on scene, the default corresponding relation based on each recommendation scene and Generalization bounds is true
Determine Generalization bounds；
When determining the use of the historical feedback information and determining Generalization bounds, based on the historical feedback information, it is determined that pushing away
Algorithm is recommended, recommendation results is obtained using the proposed algorithm of determination, Generalization bounds is obtained based on recommendation results.
The embodiment of the present invention provides a kind of recommendation method, and methods described includes：
Obtain the results set to be recommended that target recommends Generalization bounds corresponding to scene to include；The target recommends scene pair
The Generalization bounds answered are that default corresponding relation of the first server based on each recommendation scene and Generalization bounds determines；
Using the corresponding relation of recommendation results and digital content, the knot to be recommended that the results set to be recommended includes is determined
Digital content corresponding to fruit；
Judge whether the digital content of the determination can use；
When it is determined that can use, the digital content of the determination is sent to client and is shown.
In such scheme, methods described also includes：
Obtain digital content data or user data；
Using the digital content data or user data of acquisition, recommending digital content is generated according to the type of digital content
List；
The recommending digital contents list of generation is stored in the first database；The recommending digital contents list is used for for the
One server determines the recommendation results of corresponding scene.
In such scheme, when generating recommending digital contents list, methods described also includes：
Using the digital content data or user data, recommendation results are generated；
The recommendation results of generation are carried out with key assignments corresponding, generate the recommending digital content of the affiliated type of the digital content
List.
The embodiment of the present invention provides a kind of recommendation apparatus, and described device includes：
First acquisition module, the relevant information of scene is can determine for obtaining；
First determining module, for utilizing the relevant information, determine that target recommends scene；Based on each recommendation scene with pushing away
The default corresponding relation of strategy is recommended, determines that the target recommends Generalization bounds corresponding to scene；Wherein, the single Generalization bounds
Including：Correspond respectively to the results set to be recommended of the digital content of at least two types；What the default corresponding relation included
In Generalization bounds, the information recommendation algorithm of results set institute to be recommended foundation corresponding at least two Generalization bounds difference be present not
Together；
Recommending module, for the results set to be recommended included according to the Generalization bounds determined, enter row information to user
Recommend.
In such scheme, first acquisition module, specifically for judging whether default migration efficiency；
It is determined that when default migration efficiency is not present, the relevant information that can determine scene is obtained.
In such scheme, first acquisition module, it is additionally operable to when it is determined that default migration efficiency be present, based on default fortune
Battalion's strategy, it is determined that results set to be recommended corresponding with the default migration efficiency.
In such scheme, first determining module, specifically for utilizing the relevant information, the type of user is determined；
The type of the user of the determination information related to user is combined, determines that target recommends scene.
In such scheme, the recommending module, it is additionally operable to recommend Generalization bounds corresponding to scene to include from the target of determination
Recommendation results set in, determine the information recommendation ratio of each type digital content；Information recommendation ratio based on determination is entered
Row is recommended.
In such scheme, first determining module, recommend scene and historic scenery specifically for the target for judging to determine
It is whether identical；When it is determined that identical and determination has recommendation for historic scenery, from historic scenery and pair of history Generalization bounds
History Generalization bounds corresponding to middle lookup historic scenery, the target using the history Generalization bounds of lookup as the determination should be related to
Recommend the Generalization bounds of scene；When it is determined that when different, the default corresponding relation based on each recommendation scene and Generalization bounds, determining institute
State target and recommend Generalization bounds corresponding to scene.
In such scheme, described device also includes the first judge module,
First judge module, for obtaining the historical feedback information of user；Using the historical feedback information, judge
Whether Generalization bounds are preferentially determined based on scene；When it is determined that determining Generalization bounds based on scene, based on each recommendation scene with pushing away
The default corresponding relation for recommending strategy determines Generalization bounds；When determining the use of the historical feedback information and determining Generalization bounds, base
In the historical feedback information, proposed algorithm is determined, recommendation results is obtained using the proposed algorithm of determination, is obtained based on recommendation results
To Generalization bounds.
The embodiment of the present invention provides a kind of recommendation apparatus, and described device includes：
Second acquisition module, the results set to be recommended for recommending Generalization bounds corresponding to scene to include for obtaining target；
It is default corresponding based on each recommendation scene and Generalization bounds that the target recommends Generalization bounds corresponding to scene to be first server
What relation determined；
Second determining module, for the corresponding relation using recommendation results and digital content, determine the result to be recommended
Gather digital content corresponding to the result to be recommended included；
Second judge module, for judging whether the digital content of the determination can use；, will be described true when it is determined that can use
Fixed digital content is sent to client and is shown.
The embodiment of the present invention provides a kind of computer-readable recording medium, is stored thereon with computer program, the calculating
Machine program realizes the step of any recommendation method described above when being executed by processor.
The embodiment of the present invention provides a kind of recommendation apparatus, including：Processor and for store can run on a processor
Computer program memory,
Wherein, when the processor is used to run the computer program, any recommendation method described above is performed
Step.
Recommendation method, apparatus provided in an embodiment of the present invention and computer-readable recording medium, acquisition can determine scene
Relevant information；Using the relevant information, determine that target recommends scene；Default corresponding pass based on scene with recommendation results
System, scene, and each default corresponding relation for recommending scene and Generalization bounds are recommended according to the target, determine that the target pushes away
Recommend Generalization bounds corresponding to scene；Wherein, the single Generalization bounds include：Correspond respectively in the numeral of at least two types
The results set to be recommended held；In the Generalization bounds that the default corresponding relation includes, at least two Generalization bounds difference be present
The information recommendation algorithm of corresponding results set institute to be recommended foundation is different；Generalization bounds according to determining include to be recommended
Results set, information recommendation is carried out to user.In embodiments of the present invention, using the relevant information, determine that target recommends field
Scape；Default corresponding relation based on each recommendation scene and Generalization bounds, determine that the target recommends Generalization bounds corresponding to scene,
The results set to be recommended that Generalization bounds according to determining include, information recommendation is carried out to user.It is based on therefore, it is possible to realize
Scene carries out information recommendation to user.
Brief description of the drawings
Fig. 1 is the implementation process schematic diagram one that the embodiment of the present invention recommends method；
Fig. 2 is the implementation process schematic diagram two that the embodiment of the present invention recommends method；
Fig. 3 is the composition structural representation one of recommendation apparatus of the embodiment of the present invention；
Fig. 4 is the composition structural representation two of recommendation apparatus of the embodiment of the present invention；
Fig. 5 is the composition structural representation three of recommendation apparatus of the embodiment of the present invention；
Fig. 6 is the specific implementation schematic flow sheet that the embodiment of the present invention is recommended；
Fig. 7 is that first server of the embodiment of the present invention interacts the composition for realizing digital content recommending with second server
Structural representation.
Embodiment
In the embodiment of the present invention, the relevant information that can determine scene is obtained；Using the relevant information, determine that target pushes away
Recommend scene；Scene, and each default corresponding relation for recommending scene and Generalization bounds are recommended according to the target, determine the mesh
Mark recommends Generalization bounds corresponding to scene；Wherein, the single Generalization bounds include：Correspond respectively to the number of at least two types
The results set to be recommended of word content；In the Generalization bounds that the default corresponding relation includes, at least two Generalization bounds be present
The information recommendation algorithm of results set institute to be recommended foundation is different corresponding to respectively；What the Generalization bounds according to determining included treats
Recommendation results set, information recommendation is carried out to user.
The characteristics of in order to more fully hereinafter understand the embodiment of the present invention and technology contents, below in conjunction with the accompanying drawings to this hair
The realization of bright embodiment is described in detail, appended accompanying drawing purposes of discussion only for reference, is not used for limiting the present invention.
As shown in figure 1, describing the embodiment of the present invention in detail recommends method, the recommendation method of the present embodiment is applied to the first clothes
It is engaged in device side, comprising the following steps：
Step 101：Obtain the relevant information that can determine scene.
In one embodiment, the acquisition can determine including for scene：
Judge whether default migration efficiency；It is determined that when default migration efficiency is not present, acquisition can determine scene
Relevant information.
In one embodiment, methods described also includes：
When it is determined that default migration efficiency be present, based on default migration efficiency, it is determined that corresponding with the default migration efficiency
Results set to be recommended.
Specifically, can be according to major holiday (such as National Day special topic, mid-autumn special topic) or major event (as locality is drilled
The meeting of singing, modern drama performance etc.) determine results set to be recommended.
Step 102：Using the relevant information, determine that target recommends scene.
In one embodiment, it is described to utilize the relevant information, determine that target recommends scene, including：Utilize the correlation
Information, determine the type of user；The type of the user of the determination information related to user is combined, determines that target recommends field
Scape.
Here, the relevant information can include：The natural quality (such as sex, age etc.) and access record of user
Deng.During practical application, first server is recorded according to the natural quality and access of user, and user is classified as into a certain classification (as deeply
Degree search type user browses type user)；Record, user can be drawn if the natural quality of user can not be obtained and accessed
It is divided into default category (such as browsing type user).
Table 1 recommends scene for the target determined, as shown in table 1, if user type is deep search type user, it is determined that
It is scenario A that target, which recommends scene,；If user type to browse type user, it is determined that target to recommend scene be scenario B；If with
To browse type user, user related information includes temporal information (as at night), positional information (such as user's family), will used family type
Family type and user related information are combined, it is determined that target recommend scene be scene C.
| Deep search type user | Scenario A |
| Browse type user | Scenario B |
| Browse type user+evening+user's family | Scene C |
Table 1
Step 103：Default corresponding relation based on each recommendation scene and Generalization bounds, determine that the target recommends scene pair
The Generalization bounds answered.
Wherein, the single Generalization bounds include：Correspond respectively to the knot to be recommended of the digital content of at least two types
Fruit set；In the Generalization bounds that the default corresponding relation includes, exist to be recommended corresponding at least two Generalization bounds difference
The information recommendation algorithm of results set institute foundation is different.
During practical application, determine that the target is recommended in the recommending digital contents list that can also be obtained from the first database
Results set to be recommended corresponding to scene.
Table 2 is each recommendation scene and the default corresponding relation of Generalization bounds, as shown in table 2, recommends scene to include scene
A, scenario B and scene C, Generalization bounds include Generalization bounds A, Generalization bounds B and recommendation results C.By taking Generalization bounds A as an example, push away
Recommend tactful A including video class 3 recommendation results (i.e. recommendation results 1, recommendation results 2 and recommendation results 3), 2 of game class
Recommendation results (i.e. recommendation results 1, recommendation results 2), 3 recommendation results of animation class (recommendation results 1, recommendation results 2 and push away
Recommend result 3), 2 recommendation results (i.e. recommendation results 1, recommendation results 2) of music class, 1 recommendation results of books class (push away
Recommend result 1).
Table 2
For example, scenario A can be：User watches paid video at night；The recommendation of video class corresponding with scenario A
As a result 1 can be：The video bought recently.Scenario B can be：Deep search type user；Recommendation results bag corresponding with scenario B
Include：The recommendation results 1 (video of pleasantly surprised recommendation) of recommendation results 9 (guessing the video that you like)+video class of video class.Scene C
Can be：Browse type user；Recommendation results corresponding with scenario B include：The recommendation results of music class (guess the sound that you like with 6
It is happy)+4 (music of pleasantly surprised recommendation).
Here, the information recommendation algorithm of results set institute to be recommended foundation corresponding at least two Generalization bounds difference be present
It is different.For example the information recommendation algorithm of Generalization bounds A results set institute to be recommended foundations corresponding with Generalization bounds B difference is not
Together.
During practical application, key assignments corresponding to the recommendation results in results set to be recommended and the recommendation can also be tied
Fruit is correspondingly stored in the second database.Wherein, the second database can be key-value databases, i.e., one kind is with key-value pair
The database of data storage.
In one embodiment, after determining that the target recommends Generalization bounds corresponding to scene, methods described also includes：From
The target of determination is recommended in the recommendation results set that Generalization bounds corresponding to scene include, and determines the letter of each type digital content
Cease recommendation ratio；Information recommendation ratio based on determination is recommended.
During practical application, the end message obtained can be combined, recommends Generalization bounds corresponding to scene from the target of determination
Including recommendation results set in, determine the information recommendation ratio of each type digital content.Wherein, end message can wrap
Include：The installation lists of various applications in client, caching various data types accounting, (such as mobile terminal is network scenarios
It is no in the WLAN or mobile terminal network speed), request time (the i.e. response time of mobile terminal, for reflecting
The processing speed of mobile terminal processor) etc..
Specifically, when user opens client, client can obtain the installation list of various applications, can also obtain
Video file and music file of the caching in user's SD card etc. are gathered in the case of obtaining user's permission, can also be obtained in real time
Network condition.The client by message interface by end message real-time report to first server, first server according to
The end message of acquisition, recommend from the target of determination in the recommendation results set that Generalization bounds corresponding to scene include, it is determined that often
The information recommendation ratio of type digital content.
For example, if according to end message parse it is local it is data cached in 90% be video, network 4G,
It can determine that the ratio that video class accounts for type of recommendation (game, animation, music, video, books) is 80%, game class accounts for recommendation class
The ratio of type (game, animation, music, video, books) is 20%.When carrying out information recommendation, can recommend in video type
4 short-sighted frequencies, 1 trivial games in type of play.
In one embodiment, the default corresponding relation based on each recommendation scene and Generalization bounds, determine that the target is recommended
Generalization bounds corresponding to scene, including：Whether the target for judging to determine recommends scene and historic scenery identical；When it is determined that it is identical and
It is determined that when recommendation be present for historic scenery, historic scenery pair is searched from the corresponding relation of historic scenery and history Generalization bounds
The history Generalization bounds answered, recommend the Generalization bounds of scene using the history Generalization bounds of lookup as the target of the determination；When
When determining different, the default corresponding relation based on each recommendation scene and Generalization bounds, determine that the target is recommended corresponding to scene
Generalization bounds.
In one embodiment, methods described also includes：Obtain the historical feedback information of user；Believed using the historical feedback
Breath, judges whether that being preferentially based on scene determines Generalization bounds；When it is determined that determining Generalization bounds based on scene, based on each recommendation field
The default corresponding relation of scape and Generalization bounds determines Generalization bounds；Generalization bounds are determined when determining the use of the historical feedback information
When, based on the historical feedback information, proposed algorithm is determined, recommendation results are obtained using the proposed algorithm of determination, based on recommendation
As a result Generalization bounds are obtained.
During practical application, the feedback information that user is directed to the digital content of recommendation (is such as liked, not liked, thumbing up, collecting
Deng etc.) pass through KAFKA message interfaces real-time synchronization to first server.Because the digital content of each type of recommendation is taken
Band batch number, therefore first server can inquire about recommendation results and recommendation corresponding to the digital content recommended according to batch number
Algorithm.
For example, if the digital content of a certain type of the user for recommending is operated (as clicked on screen
For the liking of the digital content, do not like, thumb up, the button such as collect), then first server can according to the operation of user,
Determine recommendation results and proposed algorithm that user prefers under scene.According to hit rate generating algorithm tilting value, if life
Middle rate reaches 20%, then 100% tilt to the recommendation results corresponding to digital content recommended.
Step 104：The results set to be recommended that Generalization bounds according to determining include, information recommendation is carried out to user.
During practical application, first server can also be by key assignments corresponding to the recommendation results in results set to be recommended to
Two servers are sent；The key assignments is used for second server and obtains recommendation results corresponding to scene from the second database and utilize institute
State recommendation results and determine digital content corresponding to scene.
Recommendation method provided in an embodiment of the present invention, obtain the relevant information that can determine scene；Utilize the related letter
Breath, determine that target recommends scene；Default corresponding relation based on each recommendation scene and Generalization bounds, determine that the target recommends field
Generalization bounds corresponding to scape；Wherein, the single Generalization bounds include：Correspond respectively to the digital content of at least two types
Results set to be recommended；In the Generalization bounds that the default corresponding relation includes, at least two Generalization bounds be present and correspond to respectively
Results set institute to be recommended foundation information recommendation algorithm it is different；The result to be recommended that Generalization bounds according to determining include
Set, information recommendation is carried out to user.Obviously, scene can be based on and carries out information recommendation to user.
As shown in Fig. 2 the embodiment of the present invention, which describes the embodiment of the present invention in detail, recommends method, the recommendation method of the present embodiment
Applied to second server side, comprise the following steps：
Step 201：Obtain the results set to be recommended that target recommends Generalization bounds corresponding to scene to include.
Wherein, it is that first server is based on each recommendation scene with recommending plan that the target, which recommends Generalization bounds corresponding to scene,
What default corresponding relation slightly determined.
During practical application, it can also utilize corresponding to the recommendation results in the results set to be recommended of first server return
Key assignments, results set to be recommended is obtained from the second database according to key assignments.Wherein, the second database can be key-value numbers
According to storehouse；Wherein, key-value databases are with a kind of a kind of database with key-value pair data storage.Second database can be with
It is stored on cluster.
Step 202：Using the corresponding relation of recommendation results and digital content, determine what the results set to be recommended included
Digital content corresponding to result to be recommended.
During practical application, recommendation results can be short-sighted frequency；Corresponding digital content can be will be short according to renewal time
The digital content that video is formed.Here, recommendation results can be the ID of digital content.Second server can be by digital content ID
It is stored in the corresponding relation of digital content in the first database.
Step 203：Judge whether the digital content of the determination can use；When it is determined that can use, by the numeral of the determination
Content is sent to client and is shown.
Here, judge whether the digital content of the determination can use, can specifically include：Judge whether copyright has expired,
When it is determined that expiring, then digital content is unavailable；Otherwise digital content can use.
In one embodiment, methods described also includes：Obtain digital content data or user data；Utilize the number of acquisition
Word content-data or user data, recommending digital contents list is generated according to the type of digital content；By the recommendation number of generation
Word contents list is stored in the first database.Here, the recommending digital contents list can be used for determining for first server
The results set to be recommended that the Generalization bounds of corresponding scene include.
In one embodiment, when generating recommending digital contents list, methods described also includes：Utilize the digital content number
According to or user data, generate recommendation results；The recommendation results of generation are carried out with key assignments corresponding, generate the digital content institute
Belong to the recommending digital contents list of type.
Specifically, newest recommendation results (recommendation results 1) can be generated according to attribute of digital content itself；According to
The recommendation results (recommendation results 3) that the behavior at family generates most hot recommendation results (recommendation results 2), bought recently；According to by with
Similarity between user characteristics vector and digital content that family behavior obtains guesses that you like recommendation results (to recommend knot to generate
Fruit 4).First server can write recommendation results 1~4 in REDIS clusters according to the classification of digital content, for second service
Device calls.Wherein, the type of digital content includes：Game, animation, music, video, the class of books five.
Here, these four recommendation results more than including but is not limited to, can also be generated according to other proposed algorithms more
Recommendation results, the proposed algorithm for determining recommendation results are the common proposed algorithm in this area.
The recommendation method provided based on each embodiment of the application, the application are also provided a kind of recommendation apparatus, can be arranged on
In first server, as shown in figure 3, described device includes：First acquisition module 31, the first determining module 32, recommending module 33；
Wherein,
First acquisition module 31, the relevant information of scene is can determine for obtaining.
First determining module 32, for utilizing the relevant information, determine that target recommends scene；Based on it is each recommendation scene with
The default corresponding relation of Generalization bounds, determine that the target recommends Generalization bounds corresponding to scene.
Wherein, the single Generalization bounds include：Correspond respectively to the knot to be recommended of the digital content of at least two types
Fruit set；In the Generalization bounds that the default corresponding relation includes, exist to be recommended corresponding at least two Generalization bounds difference
The information recommendation algorithm of results set institute foundation is different；
Recommending module 33, for the results set to be recommended included according to the Generalization bounds determined, letter is carried out to user
Breath is recommended.
In one embodiment, first acquisition module 31, specifically for judging whether default migration efficiency；It is determined that
During in the absence of default migration efficiency, the relevant information that can determine scene is obtained.
In one embodiment, first acquisition module 31, it is additionally operable to when it is determined that default migration efficiency be present, based on pre-
If migration efficiency, it is determined that results set to be recommended corresponding with the default migration efficiency.
Specifically, can be according to major holiday (such as National Day special topic, mid-autumn special topic) or major event (as locality is drilled
The meeting of singing, modern drama performance etc.) determine results set to be recommended.
In one embodiment, first determining module 32, specifically for utilizing the relevant information, the class of user is determined
Type；The type of the user of the determination information related to user is combined, determines that target recommends scene.
Here, the relevant information can include：The natural quality (such as sex, age etc.) and access record of user
Deng.During practical application, according to the natural quality of user and record is accessed, user is classified as into a certain classification, and (such as deep search type is used
Family browses type user)；Recorded if the natural quality of user can not be obtained and accessed, user can be divided into acquiescence class
(type user is not browsed such as).
In one embodiment, the recommending module 33, it is additionally operable to recommend Generalization bounds corresponding to scene from the target of determination
Including recommendation results set in, determine the information recommendation ratio of each type digital content；Information recommendation ratio based on determination
Example is recommended.
In one embodiment, first determining module 32, scene and history are recommended specifically for the target for judging to determine
Whether scene is identical；When it is determined that identical and determination has recommendation for historic scenery, from historic scenery and history Generalization bounds
Corresponding relation in search historic scenery corresponding to history Generalization bounds, using the history Generalization bounds of lookup as the determination
Target recommends the Generalization bounds of scene；When it is determined that when different, the default corresponding relation based on each recommendation scene and Generalization bounds, really
The fixed target recommends Generalization bounds corresponding to scene.
In one embodiment, described device also includes the first judge module；
First judge module, for obtaining the historical feedback information of user；Using the historical feedback information, judge
Whether Generalization bounds are preferentially determined based on scene；When it is determined that determining Generalization bounds based on scene, based on each recommendation scene with pushing away
The default corresponding relation for recommending strategy determines Generalization bounds；When determining the use of the historical feedback information and determining Generalization bounds, base
In the historical feedback information, proposed algorithm is determined, recommendation results is obtained using the proposed algorithm of determination, is obtained based on recommendation results
To Generalization bounds.
It should be noted that：The recommendation apparatus that above-described embodiment provides is when being recommended, only with above-mentioned each program module
Division progress for example, in practical application, can distribute as needed and by above-mentioned processing complete by different program modules
Into the internal structure of device being divided into different program modules, to complete all or part of processing described above.Separately
Outside, the recommendation apparatus that above-described embodiment provides is with recommending embodiment of the method to belong to same design, its specific implementation process side of referring to
Method embodiment, is repeated no more here.
In actual applications, the first acquisition module 31 is realized by the network interface on recommendation apparatus；First determines mould
Block 32, the first recommending module 33, the first judge module can be by central processing unit (CPU, the Central on recommendation apparatus
Processing Unit), microprocessor (MPU, Micro Processor Unit), digital signal processor (DSP,
Digital Signal Processor) or field programmable gate array (FPGA, Field Programmable Gate
) etc. Array realize.
The recommendation method provided based on each embodiment of the application, the application are also provided a kind of recommendation apparatus, can be arranged on
On second server, as shown in figure 4, described device includes：Second acquisition module 41, the second determining module 42, second judge mould
Block 43；Wherein,
Second acquisition module 41, the result set to be recommended for recommending Generalization bounds corresponding to scene to include for obtaining target
Close.Wherein, it is that first server recommends scene and Generalization bounds based on each that the target, which recommends Generalization bounds corresponding to scene,
What default corresponding relation determined.
Second determining module 42, for the corresponding relation using recommendation results and digital content, determine the knot to be recommended
Digital content corresponding to the result to be recommended that fruit set includes；
Second judge module 43, for judging whether the digital content of the determination can use；When it is determined that can use, by described in
The digital content of determination is sent to client and is shown.
Here, judge whether the digital content of the determination can use, can specifically include：Judge whether copyright has expired,
When it is determined that expiring, then digital content is unavailable；Otherwise digital content can use.
During practical application, it can also utilize corresponding to the recommendation results in the results set to be recommended of first server return
Key assignments, results set to be recommended is obtained from the second database according to key assignments.Wherein, the second database can be key-value numbers
According to storehouse；Wherein, key-value databases are with a kind of a kind of database with key-value pair data storage.Second database can be with
It is stored on cluster.
During practical application, recommendation results can be short-sighted frequency；Corresponding digital content can be will be short according to renewal time
The digital content that video is formed.Here, recommendation results can be the ID of digital content.Second server can be by digital content ID
It is stored in the corresponding relation of digital content in the first database.
In one embodiment, described device also includes generation module,
The generation module, for obtaining digital content data or user data；Utilize the digital content data of acquisition
Or user data, generate recommending digital contents list according to the type of digital content；By the recommending digital contents list of generation
It is stored in the first database；The recommending digital contents list is used for the recommendation results that corresponding scene is determined for first server.
In one embodiment, the generation module, it is raw specifically for utilizing the digital content data or user data
Into recommendation results；The recommendation results of generation are carried out with key assignments corresponding, generate the recommending digital of the affiliated type of the digital content
Contents list.
It should be noted that：The recommendation apparatus that above-described embodiment provides is when being recommended, only with above-mentioned each program module
Division progress for example, in practical application, can distribute as needed and by above-mentioned processing complete by different program modules
Into the internal structure of device being divided into different program modules, to complete all or part of processing described above.Separately
Outside, the recommendation apparatus that above-described embodiment provides is with recommending embodiment of the method to belong to same design, its specific implementation process side of referring to
Method embodiment, is repeated no more here.
In actual applications, the second acquisition module 41 is realized by the network interface on recommendation apparatus；Second determines mould
Block 42, the second judge module 43, generation module can be realized by the CPU on recommendation apparatus, micro- MPU, DSP or FPGA etc..
Fig. 5 is the structural representation of recommendation apparatus of the present invention, and the recommendation apparatus 500 shown in Fig. 5 includes：At least one processing
Device 501, memory 502, user interface 503, at least one network interface 504.Each component in recommendation apparatus 500 passes through total
Linear system system 505 is coupled.It is understood that bus system 505 is used to realize the connection communication between these components.Bus system
505 in addition to including data/address bus, in addition to power bus, controlling bus and status signal bus in addition.But for clear explanation
For the sake of, various buses are all designated as bus system 505 in Figure 5.
Wherein, user interface 503 can include display, keyboard, mouse, trace ball, click wheel, button, button, sense of touch
Plate or touch-screen etc..
It is appreciated that memory 502 can be volatile memory or nonvolatile memory, may also comprise volatibility and
Both nonvolatile memories.Wherein, nonvolatile memory can be read-only storage (ROM, Read Only Memory),
Programmable read only memory (PROM, Programmable Read-Only Memory), Erasable Programmable Read Only Memory EPROM
(EPROM, Erasable Programmable Read-Only Memory), Electrically Erasable Read Only Memory
(EEPROM, Electrically Erasable Programmable Read-Only Memory), magnetic random access store
Device (FRAM, ferromagnetic random access memory), flash memory (Flash Memory), magnetic surface are deposited
Reservoir, CD or read-only optical disc (CD-ROM, Compact Disc Read-Only Memory)；Magnetic surface storage can be
Magnetic disk storage or magnetic tape storage.Volatile memory can be random access memory (RAM, Random Access
Memory), it is used as External Cache.By exemplary but be not restricted explanation, the RAM of many forms can use, such as
Static RAM (SRAM, Static Random Access Memory), synchronous static RAM
(SSRAM, Synchronous Static Random Access Memory), dynamic random access memory (DRAM,
Dynamic Random Access Memory), Synchronous Dynamic Random Access Memory (SDRAM, Synchronous
Dynamic Random Access Memory), double data speed synchronous dynamic RAM (DDRSDRAM,
Double Data Rate Synchronous Dynamic Random Access Memory), enhanced synchronous dynamic random
Access memory (ESDRAM, Enhanced Synchronous Dynamic Random Access Memory), synchronized links
Dynamic random access memory (SLDRAM, SyncLink Dynamic Random Access Memory), direct rambus
Random access memory (DRRAM, Direct Rambus Random Access Memory).Description of the embodiment of the present invention is deposited
Reservoir 502 is intended to the memory of including but not limited to these and any other suitable type.
Memory 502 in the embodiment of the present invention is used to store various types of data to support the behaviour of recommendation apparatus 500
Make.The example of these data includes：For any computer program operated on recommendation apparatus 500, such as operating system 5021
With application program 5022；Wherein, operating system 5021 includes various system programs, such as ccf layer, core library layer, driving layer
Deng for realizing various basic businesses and the hardware based task of processing.Application program 5022 can apply journey comprising various
Sequence, for realizing various applied business.Realize that the program of present invention method may be embodied in application program 5022.
The method that the embodiments of the present invention disclose can apply in processor 501, or be realized by processor 501.
Processor 501 is probably a kind of IC chip, has the disposal ability of signal.In implementation process, the above method it is each
Step can be completed by the integrated logic circuit of the hardware in processor 501 or the instruction of software form.Above-mentioned processing
Device 501 can be general processor, digital signal processor, either other PLDs, discrete gate or transistor
Logical device, discrete hardware components etc..The disclosed each side in the embodiment of the present invention can be realized or be performed to processor 501
Method, step and logic diagram.General processor can be microprocessor or any conventional processor etc..With reference to of the invention real
The step of applying the method disclosed in example, hardware decoding processor can be embodied directly in and perform completion, or use decoding processor
In hardware and software module combination perform completion.Software module can be located in storage medium, and the storage medium is positioned at storage
Device 502, processor 501 read the information in memory 502, with reference to the step of its hardware completion foregoing recommendation method.
The recommendation method provided based on each embodiment of the application, the application also provide a kind of computer-readable recording medium,
Shown in reference picture 5, the computer-readable recording medium can include：It is above-mentioned for storing the memory 502 of computer program
Computer program can be performed by the processor 501 of recommendation apparatus 500, to complete step described in foregoing recommendation method.Computer can
Read storage medium can be FRAM, ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface storage, CD or
The memories such as CD-ROM.
Below to realize that the instantiation that is recommended as of digital content describes present invention realization in actual applications in detail
Journey and principle.
Fig. 6 is the specific implementation schematic flow sheet of digital content recommending of the embodiment of the present invention, and Fig. 7 is first server and the
Two servers interact the composition structural representation for realizing digital content recommending, as shown in fig. 7, corresponding first clothes of policy engine
Business device, calculating platform correspond to second server, and the first database corresponds to REDIS clusters 1.Specific implementation process, including following step
Suddenly：
Step 601：Calculating platform is directed to the type of different digital contents, different according to different proposed algorithm generations
Recommendation results, and all recommendation results are stored in REDIS clusters 1 according to the type of digital content.
Wherein, REDIS clusters 1 are the storage system based on key-value i.e. key-value databases.
Specifically, to calculating platform input the digital content data related to digital content and with the behavior correlation of user
User data；Calculating platform classifies digital content by type, is divided into game, animation, music, video, the class of books five；For every
The digital content of individual type, newest recommendation results are generated (such as according to digital content data (attribute of such as digital content itself)
Recommendation results 1)；Most hot recommendation results (such as recommendation results are generated according to user data (data related to the behavior of user)
2) recommendation results (such as recommendation results 3), bought recently；According in the user characteristics vector and numeral obtained by user data
Similarity between appearance guesses that you like recommendation results (such as recommendation results 4) to generate.Calculating platform (such as pushes away all recommendation results
Recommend result 1~4) sub-category (such as game, animation, music, video, books) whole write-in REDIS clusters 1, adjusted for policy engine
With.
Digital content ID and the corresponding relation of digital content can also be stored in the REDIS clusters 1 of calculating platform.When looking into
When looking for a certain digital content, calculating platform can be searched according to digital content ID in REDIS clusters 1 corresponding to digital content.
Step 602：Policy engine prestores the corresponding relation for recommending scene and Generalization bounds.
For example, policy engine can be by a certain scene recommendation scene of paid video (such as user watch at night) with pushing away
Recommend the corresponding storage of this Generalization bounds of result (recommendation results of the nearest purchase in such as video classification).
Step 603：The relevant information that can determine scene that policy engine inputs according to calculating platform, determine that target is recommended
Scene.
The calculating platform can carry out server such as miaow cluck kind race server, miaow cluck the Video service of different business
Device, miaow cluck reading server etc..
In the relevant information that can determine scene inputted according to calculating platform, before determining that target recommends scene, in advance
Determine whether operation to intervene；If so, then according to default migration efficiency (such as according to the major holiday (such as National Day special topic, mid-autumn special topic
Deng) or major event (local concert, modern drama performance etc.) be prefixed the rule of recommendation results), determine recommendation results, will be true
Key assignments corresponding to fixed recommendation results and recommendation results are correspondingly stored in REDIS clusters 2, while return to KEY values (key assignments) letter
Cease the digital content for determining to recommend according to KEY values to calculating platform, calculating platform；If no, perform step 603.
Step 604：Corresponding relation of the policy engine based on scene and recommendation results, the recommendation obtained from REDIS clusters 1
Determine that target recommends results set to be recommended corresponding to scene, the recommendation that results set to be recommended is included in digital content list
As a result recommendation results write-in REDIS clusters 2 corresponding to corresponding key assignments and scene, at the same return KEY value informations (key assignments) to
Calculating platform, read for calculating platform from REDIS clusters 2 and call recommendation results corresponding to scene.
Step 605：Calculating platform pushes away corresponding to according to the KEY values that policy engine returns from REDIS clusters 2 obtaining scene
Result is recommended, the recommendation results of acquisition are formed to the digital content of corresponding scene.Whether the digital content that verification is formed can use, when true
Surely when available, the digital content of corresponding scene is shown in client.
It is digital content ID corresponding to the recommendation results obtained from REDIS clusters 2, it is impossible to directly shown in client,
So need to obtain digital content corresponding with digital content ID from calculating platform using digital content ID.Meanwhile judge shape
Into digital content under current state it is whether available (as judged whether copyright has expired；It is unavailable if expiring；Otherwise may be used
With), when applicable, the digital content to be formed is shown in client.User operates in client to the digital content of recommendation
When, caused operation data will return to policy engine in real time by KAFKA message interfaces, and policy engine makees the data of feedback
The foundation recommended for next time.
The foregoing is only a preferred embodiment of the present invention, is not intended to limit the scope of the present invention.
Claims (21)
- A kind of 1. recommendation method, it is characterised in that methods described includes：Obtain the relevant information that can determine scene；Using the relevant information, determine that target recommends scene；Default corresponding relation based on each recommendation scene and Generalization bounds, determine that the target recommends to recommend plan corresponding to scene Slightly；Wherein, the single Generalization bounds include：Correspond respectively to the result set to be recommended of the digital content of at least two types Close；In the Generalization bounds that the default corresponding relation includes, result to be recommended corresponding at least two Generalization bounds difference be present The information recommendation algorithm for gathering institute's foundation is different；The results set to be recommended that Generalization bounds according to determining include, information recommendation is carried out to user.
- 2. according to the method for claim 1, it is characterised in that the acquisition can determine the relevant information of scene, including：Judge whether default migration efficiency；It is determined that when default migration efficiency is not present, the relevant information that can determine scene is obtained.
- 3. according to the method for claim 2, it is characterised in that methods described also includes：When it is determined that default migration efficiency be present, based on default migration efficiency, it is determined that corresponding with the default migration efficiency treat Recommendation results set.
- 4. according to the method for claim 1, it is characterised in that it is described to utilize the relevant information, determine that target recommends field Scape, including：Using the relevant information, the type of user is determined；The type of the user of the determination information related to user is combined, determines that target recommends scene.
- 5. according to the method for claim 1, it is characterised in that determine the target recommend Generalization bounds corresponding to scene it Afterwards, methods described also includes：Recommend from the target of determination in the recommendation results set that Generalization bounds corresponding to scene include, determine in each type numeral The information recommendation ratio of appearance；Information recommendation ratio based on determination is recommended.
- 6. according to the method for claim 1, it is characterised in that the default corresponding pass based on each recommendation scene with Generalization bounds System, determine that the target recommends Generalization bounds corresponding to scene, including：Whether the target for judging to determine recommends scene and historic scenery identical；When it is determined that identical and determination has recommendation for historic scenery, from historic scenery and the corresponding relation of history Generalization bounds History Generalization bounds corresponding to middle lookup historic scenery, recommend field using the history Generalization bounds of lookup as the target of the determination The Generalization bounds of scape；When it is determined that when different, the default corresponding relation based on each recommendation scene and Generalization bounds, determining the target Recommend Generalization bounds corresponding to scene.
- 7. according to the method for claim 1, it is characterised in that methods described also includes：Obtain the historical feedback information of user；Using the historical feedback information, judge whether that being preferentially based on scene determines Generalization bounds；When it is determined that determining Generalization bounds based on scene, the default corresponding relation based on each recommendation scene and Generalization bounds determines to push away Recommend strategy；When determining the use of the historical feedback information and determining Generalization bounds, based on the historical feedback information, it is determined that recommending to calculate Method, recommendation results are obtained using the proposed algorithm of determination, Generalization bounds are obtained based on recommendation results.
- A kind of 8. recommendation method, it is characterised in that methods described includes：Obtain the results set to be recommended that target recommends Generalization bounds corresponding to scene to include；The target is recommended corresponding to scene Generalization bounds are that default corresponding relation of the first server based on each recommendation scene and Generalization bounds determines；Using the corresponding relation of recommendation results and digital content, the result pair to be recommended that the results set to be recommended includes is determined The digital content answered；Judge whether the digital content of the determination can use；When it is determined that can use, the digital content of the determination is sent to client and is shown.
- 9. according to the method for claim 8, it is characterised in that methods described also includes：Obtain digital content data or user data；Using the digital content data or user data of acquisition, recommending digital content row are generated according to the type of digital content Table；The recommending digital contents list of generation is stored in the first database；The recommending digital contents list is used for for the first clothes The recommendation results for the corresponding scene of device determination of being engaged in.
- 10. according to the method for claim 9, it is characterised in that during generation recommending digital contents list, methods described is also wrapped Include：Using the digital content data or user data, recommendation results are generated；The recommendation results of generation are carried out with key assignments corresponding, generate the recommending digital content row of the affiliated type of the digital content Table.
- 11. a kind of recommendation apparatus, it is characterised in that described device includes：First acquisition module, the relevant information of scene is can determine for obtaining；First determining module, for utilizing the relevant information, determine that target recommends scene；Based on each recommendation scene with recommending plan Default corresponding relation slightly, determine that the target recommends Generalization bounds corresponding to scene；Wherein, the single Generalization bounds bag Include：Correspond respectively to the results set to be recommended of the digital content of at least two types；What the default corresponding relation included pushes away Recommend in strategy, the information recommendation algorithm of results set institute to be recommended foundation corresponding at least two Generalization bounds difference be present not Together；Recommending module, for the results set to be recommended included according to the Generalization bounds determined, information recommendation is carried out to user.
- 12. device according to claim 11, it is characterised in thatFirst acquisition module, specifically for judging whether default migration efficiency；It is determined that when default migration efficiency is not present, the relevant information that can determine scene is obtained.
- 13. device according to claim 12, it is characterised in thatFirst acquisition module, it is additionally operable to when it is determined that default migration efficiency be present, based on default migration efficiency, it is determined that and institute State results set to be recommended corresponding to default migration efficiency.
- 14. device according to claim 11, it is characterised in thatFirst determining module, specifically for utilizing the relevant information, determine the type of user；By the class of the user of determination The type information related to user is combined, and determines that target recommends scene.
- 15. device according to claim 11, it is characterised in thatThe recommending module, it is additionally operable to the recommendation results set for recommending Generalization bounds corresponding to scene to include from the target of determination In, determine the information recommendation ratio of each type digital content；Information recommendation ratio based on determination is recommended.
- 16. device according to claim 11, it is characterised in thatFirst determining module, recommend scene and historic scenery whether identical specifically for the target for judging to determine；When it is determined that When identical and determination has recommendation for historic scenery, history is searched from the corresponding relation of historic scenery and history Generalization bounds History Generalization bounds corresponding to scene, recommend the recommendation plan of scene using the history Generalization bounds of lookup as the target of the determination Slightly；When it is determined that when different, the default corresponding relation based on each recommendation scene and Generalization bounds, determining that the target recommends scene pair The Generalization bounds answered.
- 17. device according to claim 11, it is characterised in that described device also includes the first judge module,First judge module, for obtaining the historical feedback information of user；Using the historical feedback information, judge whether Generalization bounds are preferentially determined based on scene；When it is determined that determining Generalization bounds based on scene, based on each recommendation scene with recommending plan Default corresponding relation slightly determines Generalization bounds；When determining the use of the historical feedback information and determining Generalization bounds, based on institute Historical feedback information is stated, determines proposed algorithm, recommendation results is obtained using the proposed algorithm of determination, is pushed away based on recommendation results Recommend strategy.
- 18. a kind of recommendation apparatus, it is characterised in that described device includes：Second acquisition module, the results set to be recommended for recommending Generalization bounds corresponding to scene to include for obtaining target；It is described It is default corresponding relation of the first server based on each recommendation scene and Generalization bounds that target, which recommends Generalization bounds corresponding to scene, Determine；Second determining module, for the corresponding relation using recommendation results and digital content, determine the results set to be recommended Including result to be recommended corresponding to digital content；Second judge module, for judging whether the digital content of the determination can use；When it is determined that can use, by the determination Digital content is sent to client and is shown.
- 19. a kind of computer-readable recording medium, is stored thereon with computer program, it is characterised in that the computer program The step of any one of claim 1 to 7 methods described is realized when being executed by processor, or realize that claim 8 to 10 is any The step of item methods described.
- A kind of 20. recommendation apparatus, it is characterised in that including：Processor and the calculating that can be run on a processor for storage The memory of machine program,Wherein, when the processor is used to run the computer program, any one of perform claim requirement 1 to 7 methods described Step.
- A kind of 21. recommendation apparatus, it is characterised in that including：Processor and the calculating that can be run on a processor for storage The memory of machine program,Wherein, when the processor is used to run the computer program, any one of perform claim requirement 8 to 10 methods described Step.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710986686.XA CN107609198B (en) | 2017-10-20 | 2017-10-20 | Recommendation method and device and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710986686.XA CN107609198B (en) | 2017-10-20 | 2017-10-20 | Recommendation method and device and computer readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107609198A true CN107609198A (en) | 2018-01-19 |
| CN107609198B CN107609198B (en) | 2020-06-12 |
Family
ID=61077930
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710986686.XA Active CN107609198B (en) | 2017-10-20 | 2017-10-20 | Recommendation method and device and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107609198B (en) |
Cited By (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108319722A (en) * | 2018-02-27 | 2018-07-24 | 北京小度信息科技有限公司 | Data access method, device, electronic equipment and computer readable storage medium |
| CN109658206A (en) * | 2019-01-14 | 2019-04-19 | 京东方科技集团股份有限公司 | Information recommendation method and device |
| CN109670106A (en) * | 2018-12-06 | 2019-04-23 | 百度在线网络技术（北京）有限公司 | Things recommended method and device based on scene |
| CN110147500A (en) * | 2019-05-21 | 2019-08-20 | 北京奇艺世纪科技有限公司 | A kind of information recommendation method and device |
| CN110322139A (en) * | 2019-06-28 | 2019-10-11 | 阿里巴巴集团控股有限公司 | Policy recommendation method and device |
| CN110490683A (en) * | 2018-05-15 | 2019-11-22 | 中国移动通信集团浙江有限公司 | A kind of method and system of the offline upper collaboration multi-model mixed recommendation of line |
| CN111026979A (en) * | 2019-11-12 | 2020-04-17 | 恒大智慧科技有限公司 | Target recommendation method and system and computer-readable storage medium |
| CN111177548A (en) * | 2019-12-24 | 2020-05-19 | 广州华多网络科技有限公司 | Information recommendation method and device, electronic equipment and storage medium |
| CN111198989A (en) * | 2019-12-26 | 2020-05-26 | 东软集团股份有限公司 | Method and device for determining travel recommendation data, storage medium and electronic equipment |
| CN111460279A (en) * | 2020-02-25 | 2020-07-28 | 拉扎斯网络科技（上海）有限公司 | Information recommendation method and device, storage medium and computer equipment |
| CN111556368A (en) * | 2020-04-01 | 2020-08-18 | 深圳市酷开网络科技有限公司 | Application method and system of AB test in OTT TV and storage medium |
| CN111680254A (en) * | 2020-06-11 | 2020-09-18 | 京东方科技集团股份有限公司 | Content recommendation method and device |
| CN111723234A (en) * | 2020-06-15 | 2020-09-29 | 中国第一汽车股份有限公司 | Audio providing method, device, equipment and storage medium |
| CN111897861A (en) * | 2020-06-30 | 2020-11-06 | 苏宁金融科技(南京)有限公司 | Content recommendation method and device, computer equipment and storage medium |
| CN111966911A (en) * | 2020-08-31 | 2020-11-20 | 北京健康之家科技有限公司 | Personalized service recommendation method and device and electronic equipment |
| CN111988636A (en) * | 2020-08-21 | 2020-11-24 | 广州华多网络科技有限公司 | Anchor recommendation method and device, server and computer-readable storage medium |
| CN112232915A (en) * | 2019-12-23 | 2021-01-15 | 北京来也网络科技有限公司 | Commodity recommendation method and device combining RPA and AI |
| CN112258218A (en) * | 2020-09-29 | 2021-01-22 | 京东数字科技控股股份有限公司 | Method and device for recommending products |
| CN112288517A (en) * | 2019-12-23 | 2021-01-29 | 北京来也网络科技有限公司 | Commodity recommendation method and device combining RPA and AI |
| CN112422400A (en) * | 2020-01-21 | 2021-02-26 | 上海哔哩哔哩科技有限公司 | Content recommendation method and device and computer equipment |
| CN112507235A (en) * | 2020-12-22 | 2021-03-16 | 北京明略软件系统有限公司 | Mixed-rank material recommendation method and system, electronic equipment and storage medium |
| CN112733014A (en) * | 2020-12-30 | 2021-04-30 | 上海众源网络有限公司 | Recommendation method, device, equipment and storage medium |
| CN112784080A (en) * | 2021-01-28 | 2021-05-11 | 上海发电设备成套设计研究院有限责任公司 | Scene recommendation method, system and device based on three-dimensional digital platform of power plant |
| CN112836118A (en) * | 2021-01-26 | 2021-05-25 | 南京小灿灿网络科技有限公司 | Deep learning-based recommendation method and system in interactive scene |
| CN113009839A (en) * | 2021-02-18 | 2021-06-22 | 青岛海尔科技有限公司 | Scene recommendation method and device, storage medium and electronic equipment |
| CN113157766A (en) * | 2021-03-12 | 2021-07-23 | Oppo广东移动通信有限公司 | Application analysis method and device, electronic equipment and computer-readable storage medium |
| CN113377196A (en) * | 2021-06-04 | 2021-09-10 | 北京市商汤科技开发有限公司 | Data recommendation method and device, electronic equipment and readable storage medium |
| CN116049478A (en) * | 2022-08-26 | 2023-05-02 | 荣耀终端有限公司 | Application program recommendation method and device and user equipment |
| CN116600020A (en) * | 2023-07-13 | 2023-08-15 | 支付宝(杭州)信息技术有限公司 | Protocol generation method, terminal cloud collaborative recommendation method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103885987A (en) * | 2012-12-21 | 2014-06-25 | 中国移动通信集团公司 | Music recommendation method and system |
| CN104361085A (en) * | 2014-11-14 | 2015-02-18 | 百度在线网络技术（北京）有限公司 | Information recommendation method, device, browser, server and system |
| CN106997347A (en) * | 2016-01-22 | 2017-08-01 | 华为技术有限公司 | Information recommendation method and server |
-
2017
- 2017-10-20 CN CN201710986686.XA patent/CN107609198B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103885987A (en) * | 2012-12-21 | 2014-06-25 | 中国移动通信集团公司 | Music recommendation method and system |
| CN104361085A (en) * | 2014-11-14 | 2015-02-18 | 百度在线网络技术（北京）有限公司 | Information recommendation method, device, browser, server and system |
| CN106997347A (en) * | 2016-01-22 | 2017-08-01 | 华为技术有限公司 | Information recommendation method and server |
Cited By (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108319722A (en) * | 2018-02-27 | 2018-07-24 | 北京小度信息科技有限公司 | Data access method, device, electronic equipment and computer readable storage medium |
| CN110490683A (en) * | 2018-05-15 | 2019-11-22 | 中国移动通信集团浙江有限公司 | A kind of method and system of the offline upper collaboration multi-model mixed recommendation of line |
| CN110490683B (en) * | 2018-05-15 | 2022-04-12 | 中国移动通信集团浙江有限公司 | Offline collaborative multi-model hybrid recommendation method and system |
| CN109670106A (en) * | 2018-12-06 | 2019-04-23 | 百度在线网络技术（北京）有限公司 | Things recommended method and device based on scene |
| CN109658206A (en) * | 2019-01-14 | 2019-04-19 | 京东方科技集团股份有限公司 | Information recommendation method and device |
| CN110147500A (en) * | 2019-05-21 | 2019-08-20 | 北京奇艺世纪科技有限公司 | A kind of information recommendation method and device |
| CN110147500B (en) * | 2019-05-21 | 2021-11-16 | 北京奇艺世纪科技有限公司 | Information recommendation method and device |
| CN110322139A (en) * | 2019-06-28 | 2019-10-11 | 阿里巴巴集团控股有限公司 | Policy recommendation method and device |
| CN110322139B (en) * | 2019-06-28 | 2023-11-28 | 创新先进技术有限公司 | Policy recommendation method and device |
| CN111026979A (en) * | 2019-11-12 | 2020-04-17 | 恒大智慧科技有限公司 | Target recommendation method and system and computer-readable storage medium |
| CN112288517A (en) * | 2019-12-23 | 2021-01-29 | 北京来也网络科技有限公司 | Commodity recommendation method and device combining RPA and AI |
| CN112232915A (en) * | 2019-12-23 | 2021-01-15 | 北京来也网络科技有限公司 | Commodity recommendation method and device combining RPA and AI |
| CN111177548A (en) * | 2019-12-24 | 2020-05-19 | 广州华多网络科技有限公司 | Information recommendation method and device, electronic equipment and storage medium |
| CN111177548B (en) * | 2019-12-24 | 2024-04-16 | 广州方硅信息技术有限公司 | Information recommendation method and device, electronic equipment and storage medium |
| CN111198989A (en) * | 2019-12-26 | 2020-05-26 | 东软集团股份有限公司 | Method and device for determining travel recommendation data, storage medium and electronic equipment |
| CN112422400A (en) * | 2020-01-21 | 2021-02-26 | 上海哔哩哔哩科技有限公司 | Content recommendation method and device and computer equipment |
| CN111460279A (en) * | 2020-02-25 | 2020-07-28 | 拉扎斯网络科技（上海）有限公司 | Information recommendation method and device, storage medium and computer equipment |
| CN111556368A (en) * | 2020-04-01 | 2020-08-18 | 深圳市酷开网络科技有限公司 | Application method and system of AB test in OTT TV and storage medium |
| CN111680254A (en) * | 2020-06-11 | 2020-09-18 | 京东方科技集团股份有限公司 | Content recommendation method and device |
| CN111680254B (en) * | 2020-06-11 | 2024-04-09 | 京东方科技集团股份有限公司 | Content recommendation method and device |
| CN111723234A (en) * | 2020-06-15 | 2020-09-29 | 中国第一汽车股份有限公司 | Audio providing method, device, equipment and storage medium |
| CN111897861A (en) * | 2020-06-30 | 2020-11-06 | 苏宁金融科技(南京)有限公司 | Content recommendation method and device, computer equipment and storage medium |
| CN111988636A (en) * | 2020-08-21 | 2020-11-24 | 广州华多网络科技有限公司 | Anchor recommendation method and device, server and computer-readable storage medium |
| CN111966911A (en) * | 2020-08-31 | 2020-11-20 | 北京健康之家科技有限公司 | Personalized service recommendation method and device and electronic equipment |
| CN112258218A (en) * | 2020-09-29 | 2021-01-22 | 京东数字科技控股股份有限公司 | Method and device for recommending products |
| CN112507235A (en) * | 2020-12-22 | 2021-03-16 | 北京明略软件系统有限公司 | Mixed-rank material recommendation method and system, electronic equipment and storage medium |
| CN112733014A (en) * | 2020-12-30 | 2021-04-30 | 上海众源网络有限公司 | Recommendation method, device, equipment and storage medium |
| CN112836118A (en) * | 2021-01-26 | 2021-05-25 | 南京小灿灿网络科技有限公司 | Deep learning-based recommendation method and system in interactive scene |
| CN112784080A (en) * | 2021-01-28 | 2021-05-11 | 上海发电设备成套设计研究院有限责任公司 | Scene recommendation method, system and device based on three-dimensional digital platform of power plant |
| CN113009839A (en) * | 2021-02-18 | 2021-06-22 | 青岛海尔科技有限公司 | Scene recommendation method and device, storage medium and electronic equipment |
| CN113157766A (en) * | 2021-03-12 | 2021-07-23 | Oppo广东移动通信有限公司 | Application analysis method and device, electronic equipment and computer-readable storage medium |
| CN113377196A (en) * | 2021-06-04 | 2021-09-10 | 北京市商汤科技开发有限公司 | Data recommendation method and device, electronic equipment and readable storage medium |
| CN116049478A (en) * | 2022-08-26 | 2023-05-02 | 荣耀终端有限公司 | Application program recommendation method and device and user equipment |
| CN116049478B (en) * | 2022-08-26 | 2023-09-19 | 荣耀终端有限公司 | Application program recommendation method and device and user equipment |
| CN116600020A (en) * | 2023-07-13 | 2023-08-15 | 支付宝(杭州)信息技术有限公司 | Protocol generation method, terminal cloud collaborative recommendation method and device |
| CN116600020B (en) * | 2023-07-13 | 2023-10-10 | 支付宝(杭州)信息技术有限公司 | Protocol generation method, terminal cloud collaborative recommendation method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107609198B (en) | 2020-06-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107609198A (en) | One kind recommends method, apparatus and computer-readable recording medium | |
| US11853354B2 (en) | Override of automatically shared meta-data of media | |
| CN110110203B (en) | Resource information pushing method, server, resource information display method and terminal | |
| US8393544B2 (en) | Intelligent identification of multimedia content for synchronization | |
| US7882111B2 (en) | User interactive precision targeting principle | |
| CN103902656B (en) | Media object metadata association and ranking | |
| US10891679B2 (en) | Generating media recommendations based upon beats per minute | |
| US8838641B2 (en) | Content recommendation system, content recommendation method, content recommendation device, and information storage medium | |
| CN110110262A (en) | Browser EMS memory management process, device and equipment | |
| CN110958493B (en) | Bullet screen adjusting method and device, electronic equipment and storage medium | |
| US20120117026A1 (en) | Play list management | |
| US20070220552A1 (en) | Automatic delivery of personalized content to a portable media player with feedback | |
| KR101403836B1 (en) | Virtualizing consumer behavior as a financial instrument | |
| CN107256215A (en) | Mobile computing device is loaded using media file | |
| CN101689174A (en) | Carry out selective media access by recommended engine | |
| KR20150135196A (en) | Tailoring user experience for unrecognized and new users | |
| US11113315B2 (en) | Search keyword generation | |
| TW201030542A (en) | System and method for URL based query for retrieving data related to a context | |
| CN101689284B (en) | Exchange collection of statistical data based on container | |
| CN114697683A (en) | Intelligent scheduling method, equipment and computer program product for streaming media file | |
| KR101628054B1 (en) | Virtual space providing apparatus and method | |
| CN114390299A (en) | Song on-demand method, device, equipment and computer readable storage medium | |
| CN103577481B (en) | A kind of method and apparatus of ad data search | |
| CN113407277B (en) | Display element color setting method, device, equipment and storage medium | |
| JP2007328699A (en) | Similar character data retrieval method, server, and program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |