CN104021182B - A kind of personalization resource recommendation method - Google Patents
A kind of personalization resource recommendation method Download PDFInfo
- Publication number
- CN104021182B CN104021182B CN201410254566.7A CN201410254566A CN104021182B CN 104021182 B CN104021182 B CN 104021182B CN 201410254566 A CN201410254566 A CN 201410254566A CN 104021182 B CN104021182 B CN 104021182B
- Authority
- CN
- China
- Prior art keywords
- user
- resource
- service
- relationship
- owned
- 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.)
- Expired - Fee Related
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
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/9032—Query formulation
- G06F16/90324—Query formulation using system suggestions
Abstract
The invention discloses a kind of personalized resource recommendation methods, this method is that the service flow extraction user customized using user's history uses resource preference, it forms user resources recommended models and obtains the user resources recommendation list of great personalization using resource recommendation algorithm on the basis of this model.The specific steps are (1), describe and have recorded relationship between service flow, service, user and resource;(2), buddy subscriber relationship is defined using the similar preference of resource according to user;(3), the relationship between buddy subscriber is accustomed to using resource according to user and generates user resources recommended models;(4), according to user resources recommended models, design realizes user resources proposed algorithm.
Description
Technical field
The invention belongs to cloud service fields, concretely relate to a kind of personalized resource recommendation method.
Background technology
Cloud service refers to a kind of standard interface that can be used on the internet, user can usually use HTTP and
HTTPS transport protocols call.Its realization method mainly include software i.e. service (SaaS), platform i.e. service (PaaS) and
Three kinds of modes of Web service, service referred to herein generally refer to the software provided in cloud platform, i.e. software services.
The serviced component provided in cloud platform combines according to specific control flow to be completed particular task and constitutes
It executes flow and is known as cloud service stream.Seemingly with common workflow class, cloud service stream need to initially set up procedural model, then execute
The flow is executed in engine.Cloud service stream is once set up, and just contains specific data flow and control is flowed, enforcement engine root
Determine which cloud service called according to the relationship between being serviced in flow.There are two types of modes for the composition of cloud service stream:User customize and
System provides, and for the service flow that system provides, user can not change, and can be only done the distinctive function of service flow, and user is fixed
Service flow processed can according to the specific demand of user, by different cloud services without prejudice under the premise of constraint between service,
The customization of service flow is realized according to the demand of user.
The requirement that the visualization of service flow customization can meet user's on-demand customization service flow still provides in cloud platform
Service it is multifarious far more than, allow user oneself to find the cloud service of oneself needs in cloud platform, actual feasibility is very small.This
Text proposes a kind of user resources recommended models according to the characteristics of cloud service stream, and on the basis of the model, it is proposed that user
Resource recommendation algorithm.
Invention content
In conjunction with the advantage of cloud service stream cloud clothes are improved present invention aims at a kind of personalized resource recommendation method is provided
Be engaged in the service quality of stream, promote the user experience of cloud platform, allow user can physical experience to cloud service platform be exactly for user volume
Body is customized.
In order to achieve the above objectives, design of the invention is as follows:
Relationship between Analysis Service stream, service, user and resource proposes the theory of buddy subscriber, indicates inter-resource relation
User resources recommended models, user resources proposed algorithm.
Conceive according to above-mentioned invention, the present invention uses following technical proposals:
A kind of personalization resource recommendation method, it is characterised in that concrete operation step is as follows:
(1), describe and record the relationship between service flow, service, user and resource;
(2), buddy subscriber relationship is defined using the similar preference of resource according to user;
(3), the relationship between buddy subscriber is accustomed to using resource according to user and generates user resources recommended models;
(4), according to user resources recommended models, user resources proposed algorithm is realized in design.
It is described in the above (1) and records the relationship between service flow, service, user and resource, detailed relation is as follows:
11) service is the daughter element of service flow;
12) resource is to realize that service is essential basic, including data resource and Service Source;
13) service flow is by realizing that each resource of service is formed;
14) according to whether being the owner of resource, the user is divided into resource user and resource owner, this depends on
Whether it is arranged to publicly-owned in resource;
15) the implicit relationship needs between user and user and Service Source and Service Source are found by excavating.
Buddy subscriber relationship is defined using resource preference according to user in the above (2), it is specific as follows:
Assuming that existing publicly-owned Service Source m, publicly-owned data resource n:
MpThe frequency that-pth (0 < p≤m) a Service Source is used by a user;
NqThe frequency that-the q (0 < q≤n) a data resource is used;
Then user is expressed as P (M using the feature vector of publicly-owned resource1...Mp...Mm N1...Nq...Nn);
The similarity between user can be calculated using the feature vector of resource according to user.
If there are two user i, j, the frequency vector of their the used publicly-owned resources of history uses P (i), P (j) tables respectively
Show.The similarity between user i and user j is indicated with Sim (i, j), then the calculation formula of Sim (i, j) is as follows:
Wherein, same (P1, P2) indicates the vector that same alike result respective value is constituted in vector P1 and P2, length (P) table
Show the length of vectorial P, same (P (i), P (j))kIndicate two users i, j the publicly-owned resource of used history frequency vector
Same alike result respective value constitutes k-th of attribute component of vector in P (i), P (j).P(i)hIndicate user i using resource frequency to
H-th of component of amount, P (j)lThen indicate that user j uses first of component of resource frequency vector.
Definition:For any two user i, j, the similarity Sim (i, j) between user i and user j is calculated, defines threshold
Value t (0 t≤1 <) claims user i and user j buddy subscribers each other if there are Sim (i, j) > t.
It is accustomed to and the relationship generation user resources recommendation mould between buddy subscriber using resource according to user in the above (3)
Type.It refers to that user uses resource preference and buddy subscriber relationship that the user is accustomed to the relationship between buddy subscriber using resource,
And buddy subscriber relationship is defined using the similar preference of resource according to user, if user U1 has customized 2 service flows, then user provides
Source recommended models are as shown in Fig. 2, be described in detail below:
31) user U2 is contacted by partnership in the top of model, indicates that user U2 is the buddy subscriber of U1;
32) user that the resource that user U1 is used when customizing this two service flows is saved in below model uses money
In the list of source;
33) model node relationship part saves the relationship between each resource in service flow.
In the above (4) user resources proposed algorithm (SRR) is realized according to user resources recommended models, design.If
Active user U, it is current to select resource CR.Data resource shows that list dis_D, Service Source show list dis_S, indicated with HA
The number that some resource is used by some user's history, and referred to as history liveness of the resource to user.
Algorithm is as follows:
41) whether detection service device end memory has imported resource recommendation model, if not importing, imports the money of N number of user
Source recommended models;
42) according to resource CR (CR can be sky) is currently selected, publicly-owned resource can be obtained respectively using resource recommendation model
(multiple) to the history liveness vector PUB_HA of user U, history liveness vector PUB_HA is publicly-owned resource vector, public
There is resource (multiple) to the private privileges of history liveness the matrix PUB_MHA and U of buddy subscriber (multiple) to user U's
Liveness vector PRI_HA;
43) it utilizes PUB_HA vectors, PUB_MHA matrixes and PRI_HA vectors, resource preference, U is used by user U
Buddy subscriber and resource input and output constraint obtain the data of user U and Service Source shows list dis_D and dis_S;
44) user U selects required resource CR from dis_D and dis_S, goes to step 2), and cycle has been customized until service flow
At.
The present invention compared with prior art, have following obvious prominent substantive distinguishing features and notable technology into
Step:This method generates user resources recommended models according to user using resource preference and buddy subscriber relationship, based on the model
User resources proposed algorithm effectively improves selection difficult problem of the user plane to vast resources, improves user customized service
The efficiency of stream.
Description of the drawings
Fig. 1 is the resource recommendation flow chart in the service flow customization of the present invention;
Fig. 2 is user resources recommended models;
Fig. 3 is browsing number of resources distribution when customizing service flow using random presentation, SRR and collaborative filtering;
Fig. 4 is to customize service flow amount using SRR algorithms and averagely browse the relationship of number of resources;
Fig. 5 is number of users and the relationship for averagely browsing number of resources when customizing service flow using SRR algorithms.
Specific implementation mode
The present invention is described in further detail with preferred implementation below in conjunction with the accompanying drawings.
Embodiment one:
Referring to Fig. 1, this personalization resource recommendation method operating procedure is as follows:1) history for obtaining user is arranged using resource
Table;2) the buddy subscriber list of active user is obtained;3) history for obtaining buddy subscriber uses resource matrix;4) according to partner
The similarity of user and active user obtain the weight of each buddy subscriber;5) it is obtained according to the weight of each buddy subscriber and recommends money
Source list.
Embodiment two:
The present embodiment and embodiment one are essentially identical, and special feature is as follows:1) it describes and records service flow, service, user
Relationship between resource;2) buddy subscriber relationship is defined using resource preference according to user;3) resource preference is used according to user
User resources recommended models are proposed with buddy subscriber relationship;4) according to user resources recommended models, design realizes that user resources push away
Recommend algorithm.
Embodiment three:
The running environment of this example is that hardware includes 15 PC machine and a server.PC machine is respectively mounted the behaviour of CentOS 5.6
Make system, wherein 3 are Lenovo household machines, is configured to 4GB memories, Pentium (R) Dual-core CPU E5700
3.0GHz, 500GB hard disk;12 cheap household machines for Dell, Dell machines are configured to 2GB memories, Pentium (R) 4
3.0GHz Dual-core CPU, 80GB hard disks;Server is as cloud service stream engine, the service for receiving user's submission
Stream.Software includes tomcat 6.0 and Hadoop 0.2.Including number of users be 100, while each user has private data
Resource and each 50 of private services resource, also comprise 5000 publicly-owned data resources and 5000 publicly-owned Service Sources.If working as
Preceding user is U, and the resource that user currently selects is CR, and data resource shows that list dis_D, Service Source show list dis_
S。
(1) it describes and records the relationship between service flow, service, user and resource, concrete operations are as follows:
(11) service is the daughter element of service flow, if service flow is SF, is serviced as S, then S ∈ SF;
(12) resource is to realize that service is essential basic, including data resource, Service Source and icon resource.If
Resource collection is Rset, data resource collection is combined into Dset, Service Source collection is combined into Sset, then Rset=Dset∪Sset。
(13) service flow is by realizing that each resource of service is formed, if RiFor the resource that service flow includes, service flow is provided by n
Source is constituted, then
(14) according to whether being the owner of resource, the user is divided into resource user and resource owner, this takes
Certainly whether it is arranged to publicly-owned in resource;
(15) implicit relationship between user and user and Service Source and Service Source recommends mould using user resources
Type excavates.
(2) buddy subscriber relationship is defined using the similar preference of resource according to user, be as follows:
(21) it obtains active user's U history and uses resource vector, it is assumed that existing publicly-owned Service Source m, publicly-owned data money
Source n:
MpThe frequency that-pth (0 < p≤m) a Service Source is used by a user;
NqThe frequency that-the q (0 < q≤n) a data resource is used;
Then user U is expressed as P (M using the feature vector of publicly-owned resource1...Mp...Mm N1...Nq...Nn);
(22) other users UOHistory is P using resource vectorOIf threshold value t (0 < t < 1), calculated using similarity formula
P and POSimilarity Sim (P, PO), if Sim (P, PO) > t, then UOFor the buddy subscriber of U, method finds out the institute of user U according to this
There is buddy subscriber.
(3) relationship between buddy subscriber is accustomed to using resource according to user and generates user resources recommended models, the use
Family is accustomed to the relationship between buddy subscriber referring to that user uses resource preference and buddy subscriber relationship using resource, and according to user
Buddy subscriber relationship is defined using the similar preference of resource, is as follows:
(31) user U2 is contacted by partnership in the top of model, indicates that user U2 is the buddy subscriber of U1.It utilizes
The buddy subscriber list of active user U1 is found out in the definition of buddy subscriber, is recorded in the top of model.
(32) user U1 customizes the user that the resource used when service flow is saved in below model and uses the Resources list
In.
(33) set by a Service Source in service flow and node that multiple data resources for being connected directly with it form as
Model node relationship in service flow is stored in the middle section of user resources recommended models by model node.
(4) according to user resources recommended models, design is realized user resources proposed algorithm, is as follows:
(41) whether detection service device end memory has imported resource recommendation model.If not importing, 100 are imported from external memory
The resource recommendation model of user.
(42) according to resource CR (CR can be sky) is currently selected, indicate that some resource is used by some user's history with HA
Number, and referred to as history liveness of the resource to user U.Publicly-owned resource can be obtained respectively to user using resource recommendation model
The history liveness vector PUB_HA of U, private privileges of the publicly-owned resource to history liveness the matrix PUB_MHA, U of buddy subscriber
To the frequency vector PRI_HA of U.
If the forerunner and successor set for model node M are pre_M, next_M, then the subsequent node collection of M is combined into
FOLLOWMs=pre_M∪next_M.Such as Fig. 2, for model node M0, forerunner, which collects, is combined into { M1, M3, successor set is { M2,
M4, then M0Following model node set be { M1, M2, M3, M4}。
If user U has M buddy subscriber, some publicly-owned resource is HAP=(HA to the HA vectors of M user1,…,HAm),
Weight vectors WP=(the W of buddy subscriber1,…,Wm), the similarity vector of U and buddy subscriber is SIM=(S1,…,Sm), then it provides
Expection liveness PA=HAP*WP of the source to user U.Because of W=f (S), and require W=f (S) to meet S is bigger, and W is bigger, and
SIM=(1 ..., 1)mWhen,So desirable f (S)=S.I.e. the expection liveness of user U is PA=HAP*WP
=HAP* (W1,…,Wm)=HAP*SIM.The publicly-owned resource PA=HA that private privileges and user itself are used.
If the resource for including in model node is known as resource instances;
If mtimes is the number that some model node occurs in the service flow that user's history customizes.
If rtimes is the number that resource instances occur in model node.
If fmtimes is degree of the model node to following model node.
When CR is empty, the recommended models user for taking out user U uses publicly-owned resource and private privileges pair in the Resources list
The history liveness of user U, is separately recorded in PUB_HA and PRI_HA vectors.If wherein private services resource pri_x
It is a, publicly-owned Service Source pub_x, private data resource pri_y, publicly-owned data resource pub_y.
Private privileges are expressed as the HA vectors of user U:
PRI_HA=(pri_SHA1,…,pri_SHApri_x,pri_DHA1,…,pri_DHApri_y)
Publicly-owned resource is expressed as the HA vectors of user U:
PUB_HA=(pub_SHA1,…,pub_SHApub_x,pub_DHA1,…,pub_DHApub_y)
Wherein:pri_SHAi- the i-th (0<i<=pri_x) a private services resource is to the HA of user U, pri_SHAi>=0;
pri_DHAi- the i-th (0<i<=pri_y) a private data resource is to the HA of user U, pri_DHAi>=0;
pub_SHAi- the i-th (0<i<=pub_x) a publicly-owned Service Source is to the HA of user U, pub_SHAi>=0;
pub_DHAi- the i-th (0<i<=pub_y) a publicly-owned data resource is to the HA of user U, pub_DHAi>=0;
When CR is not empty, according to the resource recommendation model of user U, from user using finding resource CR in the Resources list,
The model node identification list for including resource CR is taken out, corresponding model node is taken out in model node list, constitutes CML.
If CR is data resource, takes out from CML and provided in all Service Sources of the same model node, data with CR
Source merges the history liveness of same asset using corresponding mtimes*rtimes as their history resource liveness, if
Wherein private services resource pri_x, publicly-owned Service Source pub_x, private data resource pri_y, publicly-owned data resource
Pub_y.
Privately owned following resource is expressed as the HA vectors of user U:
PRI_HA=(pri_SHA1,…,pri_SHApri_x,pri_DHA1,…,pri_DHApri_y)
Publicly-owned following resource is expressed as the HA vectors of user U:
PUB_HA=(pub_SHA1,…,pub_SHApub_x,pub_DHA1,…,pub_DHApub_y)
If CR is Service Source, taken out with CR from CML in the data resource of the same model node, it will be corresponding
Mtimes*rtimes merges the history liveness of same data resource, if wherein privately owned as their history resource liveness
Data resource pri_y, publicly-owned data resource pub_y.The following model of the model node comprising resource CR is taken out from CML
Node listing takes out Service Source therein, using corresponding fmtimes as the history liveness of Service Source, merges identical
The history liveness of Service Source, if wherein private services resource pri_x, publicly-owned Service Source pub_x.
Privately owned following resource is expressed as the HA vectors of user U:
PRI_HA=(pri_SHA1,…,pri_SHApri_x,pri_DHA1,…,pri_DHApri_y)
Publicly-owned following resource is expressed as the HA vectors of user U:
PUB_HA=(pub_SHA1,…,pub_SHApub_x,pub_DHA1,…,pub_DHApub_y)
Wherein:pri_SHAi- the i-th (0<i<=pri_x) a privately owned follow-up service resource is to the HA of user U, pri_SHAi>
=0;
pri_DHAi- the i-th (0<i<=pri_y) a privately owned follow-up data resource is to the HA of user U, pri_DHAi>=0;
pub_SHAi- the i-th (0<i<=pub_x) a publicly-owned follow-up service resource is to the HA of user U, pub_SHAi>=0;
pub_DHAi- the i-th (0<i<=pub_y) a publicly-owned follow-up data resource is to the HA of user U, pub_DHAi>=0;
PUB_ of the publicly-owned resource to buddy subscriber is obtained from the resource recommendation model of buddy subscriber using same method
HA forms PUB_MHA matrixes.
(43) PUB_HA vectors, PUB_MHA matrixes and PRI_HA vectors are utilized, the use resource preference of user U, U are passed through
Buddy subscriber and resource input and output constraint obtain the data and Service Source recommendation list dis_D and dis_S of user U.
Using PUB_HA and PUB_MHA, is constrained and obtained using resource preference, the buddy subscriber of U and input and output by U
Publicly-owned the Resources list pub_D and pub_S;
Using PRI_HA, by the use resource preference of U and input and output constraint obtain private privileges list pri_D and
pri_S;
Pri_S and pub_S is merged, and is sorted from big to small according to PA, dis_S is obtained;Pri_D and pub_D is merged,
And sort from big to small according to resource PA, obtain dis_D.
(44) user U selects required resource CR from dis_D and dis_S, goes to step (42), and cycle is customized until service flow
It completes.
The display between the service of considering and service, between service and data is can be seen that from the experimental result of Fig. 3,4 and 5
No matter the user resources proposed algorithm based on user resources recommended models of constraint and implicit association, customize the big of service flow scale
Small, the number that resource is browsed using intelligent recommendation is all significantly less than the random number for presenting and browsing resource;For different
User, intelligent recommendation algorithm are also effective always.
Claims (2)
1. a kind of personalization resource recommendation method, includes the reason of relationship, buddy subscriber between service flow, service, user and resource
User resources recommended models, the user resources proposed algorithm of relationship between user resources are read, indicate, concrete operation step is as follows:
1) it describes and records the relationship between service flow, service, user and resource;
2) buddy subscriber relationship is defined using the similar preference of resource according to user;The specific method is as follows:
Assuming that existing publicly-owned Service Source m, publicly-owned data resource n:
MpThe frequency that-p-th Service Source is used by a user, 0 < p≤m;
NqThe frequency that-q-th data resource is used, 0 < q≤n;
Then user is expressed as P (M using the feature vector of publicly-owned resource1...Mp......Mm,N1...Nq...Nn);
The similarity between user can be calculated using the feature vector of resource according to user, if the used history of user i, j
For publicly-owned resource to i, the frequency vector of j is respectively P (i), P (j), defines threshold value t, 0 < t < 1, the phase between user i and user j
It is indicated like degree Sim (i, j), define the length that length (P) indicates vector, same (P1, P2) indicates phase in vector P1 and P2
With the vector that attribute respective value is constituted, calculation formula is as follows:
Definition:Equipped with N number of user, if there are Sim (i, j) > t, claim user i and user j buddy subscribers each other;
3) relationship between buddy subscriber is accustomed to using resource according to user and generates user resources recommended models;
4) according to user resources recommended models, user resources proposed algorithm is realized in design;
The step 3) is accustomed to the side that the relationship between buddy subscriber generates user resources recommended models according to user using resource
Method is:It refers to that user is closed using resource preference and buddy subscriber that the user is accustomed to the relationship between buddy subscriber using resource
System, and defines buddy subscriber relationship according to user using the similar preference of resource, if user U1 has customized 2 service flows, then user
Resource recommendation model is described in detail below:
31) user U2 is contacted by partnership in the top of model, indicates that user U2 is the buddy subscriber of U1;
32) user that the resource that user U1 is used when customizing this two service flows is saved in below model uses the Resources list
In;
33) model node relationship part saves the relationship between each resource in service flow;
For the step 4) according to user resources recommended models, design realizes that user resources proposed algorithm is:If active user U, when
Preceding selection resource CR;Data resource shows that list dis_D, Service Source show list dis_S, be as follows:
41) whether detection service device end memory has imported resource recommendation model, if not importing, the resource for importing N number of user pushes away
Recommend model;
42) according to current selection resource CR, it is active to the history of user U that publicly-owned resource can be obtained respectively using resource recommendation model
Vector PUB_HA is spent, history liveness vector PUB_HA is publicly-owned resource vector, and publicly-owned resource lives to the history of buddy subscriber
Frequency vector PRI_HA of the private privileges of jerk matrix PUB_MHA, U to U;
43) PUB_HA vectors, PUB_MHA matrixes and PRI_HA vectors are utilized, resource preference, the partner of U are used by user U
The resource that user and resource input and output constraint obtain user U shows list dis_D and dis_S;
44) user U is shown from data resource and Service Source selects required resource CR in list dis_D and dis_S, go to step
42), cycle is completed until service flow customizes.
2. personalization resource recommendation method according to claim 1, it is characterised in that the description of the step 1) and record
Relationship between service flow, service, user and resource is as follows:
11) service is the daughter element of service flow;
12) resource is to realize that service is essential basic, including data resource and Service Source;
13) service flow is by realizing that each resource of service is formed;
14) according to whether being the owner of resource, the user is divided into resource user and resource owner, this depends on money
Whether source is arranged to publicly-owned;
15) implicit relationship between user and user and Service Source and Service Source needs to find by excavating.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410254566.7A CN104021182B (en) | 2014-06-10 | 2014-06-10 | A kind of personalization resource recommendation method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410254566.7A CN104021182B (en) | 2014-06-10 | 2014-06-10 | A kind of personalization resource recommendation method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104021182A CN104021182A (en) | 2014-09-03 |
CN104021182B true CN104021182B (en) | 2018-10-23 |
Family
ID=51437936
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410254566.7A Expired - Fee Related CN104021182B (en) | 2014-06-10 | 2014-06-10 | A kind of personalization resource recommendation method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104021182B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105740392A (en) * | 2016-01-27 | 2016-07-06 | 浪潮软件股份有限公司 | Resource sharing apparatus, system and method |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101271559A (en) * | 2008-05-16 | 2008-09-24 | 华东师范大学 | Cooperation recommending system based on user partial interest digging |
CN101751448A (en) * | 2009-07-22 | 2010-06-23 | 中国科学院自动化研究所 | Commendation method of personalized resource information based on scene information |
CN101819572A (en) * | 2009-09-15 | 2010-09-01 | 电子科技大学 | Method for establishing user interest model |
CN102004767A (en) * | 2010-11-10 | 2011-04-06 | 北京航空航天大学 | Abstract service logic-based interactive semantic Web service dynamic combination method |
CN103544623A (en) * | 2013-11-06 | 2014-01-29 | 武汉大学 | Web service recommendation method based on user preference feature modeling |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2481018A4 (en) * | 2009-09-21 | 2013-06-12 | Ericsson Telefon Ab L M | Method and apparatus for executing a recommendation |
-
2014
- 2014-06-10 CN CN201410254566.7A patent/CN104021182B/en not_active Expired - Fee Related
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101271559A (en) * | 2008-05-16 | 2008-09-24 | 华东师范大学 | Cooperation recommending system based on user partial interest digging |
CN101751448A (en) * | 2009-07-22 | 2010-06-23 | 中国科学院自动化研究所 | Commendation method of personalized resource information based on scene information |
CN101819572A (en) * | 2009-09-15 | 2010-09-01 | 电子科技大学 | Method for establishing user interest model |
CN102004767A (en) * | 2010-11-10 | 2011-04-06 | 北京航空航天大学 | Abstract service logic-based interactive semantic Web service dynamic combination method |
CN103544623A (en) * | 2013-11-06 | 2014-01-29 | 武汉大学 | Web service recommendation method based on user preference feature modeling |
Non-Patent Citations (1)
Title |
---|
杜金峰等;杜金峰等;《计算机工程与设计》;20130116;第34卷(第1期);第344页第2节、图1-2 * |
Also Published As
Publication number | Publication date |
---|---|
CN104021182A (en) | 2014-09-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103678647B (en) | A kind of method and system for realizing information recommendation | |
CN102075851B (en) | Method and system for acquiring user preference in mobile network | |
US10489845B2 (en) | System and method for providing context driven hyper-personalized recommendation | |
CN105224606B (en) | A kind of processing method and processing device of user identifier | |
CN107515878B (en) | Data index management method and device | |
US20180336202A1 (en) | System and method to represent documents for search in a graph | |
Rakesh et al. | Personalized recommendation of twitter lists using content and network information | |
CN103106285A (en) | Recommendation algorithm based on information security professional social network platform | |
US9129296B2 (en) | Augmenting recommendation algorithms based on similarity between electronic content | |
US10554613B2 (en) | Dynamic hashtag ordering based on projected interest | |
CN106528894B (en) | The method and device of label information is set | |
CN105005582A (en) | Recommendation method and device for multimedia information | |
CN104869048B (en) | Packet processing method, the apparatus and system of microblog data | |
CN104077415A (en) | Searching method and device | |
CN104077723A (en) | Social network recommending system and social network recommending method | |
CN102959539B (en) | Item recommendation method during a kind of repeat in work and system | |
CN104767810A (en) | Cloud-client cooperative service system and cloud-client cooperative work method | |
CN105718951A (en) | User similarity estimation method and system | |
CN104978406A (en) | User behavior analysis method of Internet platform | |
CN109033190A (en) | A kind of method for pushing of recommendation information, device and equipment | |
CN114830080B (en) | Data distribution flow configuration method and device, electronic equipment and storage medium | |
CN104021182B (en) | A kind of personalization resource recommendation method | |
CN103838773A (en) | User judgment method and device for search result | |
CN105677699A (en) | Method and apparatus for generating news pages for user | |
CN110717095B (en) | Service item pushing method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20181023 Termination date: 20210610 |