CN104268760B - A kind of user interest is obtained and transmission method and its system - Google Patents
A kind of user interest is obtained and transmission method and its system Download PDFInfo
- Publication number
- CN104268760B CN104268760B CN201410494809.4A CN201410494809A CN104268760B CN 104268760 B CN104268760 B CN 104268760B CN 201410494809 A CN201410494809 A CN 201410494809A CN 104268760 B CN104268760 B CN 104268760B
- Authority
- CN
- China
- Prior art keywords
- interest
- user
- collection
- model
- client
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Abstract
Obtained the present invention relates to a kind of user interest includes step S1 with transmission method and its system, method:Client initialization common user interest model;Step S2:Generate initial interest collection;Step S3:User accesses e-commerce platform;Step S4:Propose to propagate interest request, personalized recommendation module receives the interest collection of client and updates by interest model translation interface;Step S5:Judge whether the interest collection of correspondence user in personalized recommendation module is identical with interest collection in client, if it is not, performing step S6, if so, interest is propagated completing, perform step S7;Step S6:Personalized recommendation module extracts new interest collection, and converted interface is sent to client, updates client interest collection;Step S7:User accesses other e-commerce platforms, jump procedure S4.Compared with prior art, the present invention can rapidly catch the interest of user, and user interest gradual perfection is caused by constantly updating, and obtain and propagate beneficial to interest.
Description
Technical field
The present invention relates to personalized recommendations in E-business field, obtained more particularly, to a kind of user interest and transmission method
And its system.
Background technology
Personalized recommendation system is a kind of behavioural characteristic and Characteristic of Interest according to user, recommends it may to feel emerging to user
The commodity of interest or the software systems of resource, have of crucial importance and are widely applied in e-commerce field.Personalized recommendation
Set up on the user's history data of magnanimity, by data mining and proposed algorithm, help e-commerce platform (below referred to as
Platform) provide a user with the solution of a set of personalized decision support.Proposed algorithm is generally divided into three kinds:Pushing away based on content
Recommend algorithm, collaborative filtering and mixing proposed algorithm.Content-based recommendation is matching between resource and resource, in combination with
User interest is recommended.Collaborative filtering recommending is user and matching that user asks, i.e., the scoring of commodity is entered using user
Row is recommended.It is then that the above two way of recommendation is combined that mixing is recommended.
No matter commending system uses above-mentioned any proposed algorithm, and its common ground is all to be necessarily dependent upon in respective platform to use
The historical record that family browses, so as to calculate user interest model or user-related information using historical record.This is resulted in
Following two defects.
1) platform dependence:Each platform uses a set of Users' Interests Mining algorithm of oneself, and independently safeguards
With one group of interest related data of renewal.These data are that specific to the platform, once leaving this platform, these data are just lost
Meaning is gone, because the basic None- identified of other platforms.
2) locality of data and openness:The interest related data of user is existed only in specific platform, and
It is just meaningful only in the platform.If user be switched to one he from the platform having not visited, then in the new platform
His historical data and interest related data is sky.The platform cannot make recommendation for his interest at all.Even this is put down
Platform is very similar with the platform that above he accessed (such as being all internet book store).
At present, the processing mode for above-mentioned two problems is all based on what single platform was carried out, and the present invention is then with more
The generally starting point of platform, proposes that a set of brand-new user interest is obtained and mechanism of transmission, so as to be largely overcoming
Above-mentioned two defect present in existing commending system.
The content of the invention
The purpose of the present invention is exactly to provide a kind of user interest for the defect for overcoming above-mentioned prior art to exist to obtain
With transmission method and its system, the interest of user can be quickly caught, user interest gradual perfection is caused by constantly updating, from
And be beneficial to interest and obtain and propagate.
The purpose of the present invention can be achieved through the following technical solutions:
A kind of user interest is obtained and transmission method, is comprised the following steps:
Step S1:Client sets up the memory space of storage common user interest model, and by common user interest model
In interest collection be initialized as empty set;
Step S2:Based on common user interest model, user interest investigation result or electronics business of the client according to feedback
The historical interest collection of business platform, initial interest collection in generation common user interest model;
Step S3:User accesses e-commerce platform;
Step S4:Client proposes to propagate user interest request to the personalized recommendation module of current E-commerce platform,
Current Personalized recommending module receives common user interest model in client by interest model translation interface, according to general use
Interest collection in the interest model of family updates the interest collection of correspondence user in current Personalized recommending module;
Step S5:Judge correspondence user in current Personalized recommending module interest collection whether with common user in client
Interest collection in interest model is identical, if it is not, then performing step S6, if so, interest is propagated completing, performs step S7;
Step S6:The new interest collection of correspondence user is extracted from current Personalized recommending module, is changed by interest model
Interface is sent to client, updates the interest collection in common user interest model in client, and interest updates and completes;
Step S7:User continues to access the personalized recommendation of other e-commerce platforms, client and e-commerce platform
Module is communicated, jump procedure S4.
The method for building up of described common user interest model includes:
101:Set up the basic structure of description user interest feature, i.e., first interest, each first interest is by several key-value pairs
Composition, is expressed as first interest in=∪ < ki, vi> (i=1,2,3 ...), ki, viIt is key assignments, by all first interest composition set
It is first interest complete or collected works, is designated as INall={ in1, in2..., inN, wherein N is the length of first interest complete or collected works, then emerging by several yuan
Interest composition collection is combined into first interest collection, is designated as IN, and
102:Set up the structure of correlation between description first interest, i.e. interest relation, each interest relation is with a ternary
Group expression, is expressed as interest relation rij=< ini, inj, ρ > (ini, inj∈ IN), wherein ρ represents first interest iniAnd injBetween
Relation function, and rijAnd rjiIt is different;
103:By step 101,102, for first interest collection IN, there is one group of interest relation, be designated as:R=∪ rij=∪ <
ini, inj, ρ > (ini, inj∈ IN), first interest collection IN interest relation R corresponding with its is referred to as user interest collection, is designated as S=
(IN, R), then for user u, its common user interest model is expressed as an interest collection or the union of several interest collection, note
For:
Described step S2 is specially:
201:Client judges whether to receive the user interest survey of feedback, if so, step 205 is then performed, if it is not,
Then user accesses e-commerce platform, performs step 202;
202:Set up between client and the personalized recommendation module of e-commerce platform and communicated;
203:Judge in current Personalized recommending module whether the historical interest collection comprising active user, if so, then performing
204, if it is not, then user accesses other e-commerce platforms, redirect 202;
204:Client obtains historical interest collection from current Personalized recommending module, in generation common user interest model
Initial interest collection;
205:Client generates the initial interest collection of common user interest model according to user interest Questionnaire results.
A kind of user interest acquisition realized using method described above and broadcasting system, including e-commerce platform is individual
Property recommending module, also including client and interest model translation interface, described client includes memory module and interest mould
Type management module, described memory module connection interest model management module, described interest model management module connection interest
Model conversion interface, described interest model translation interface connects personalized recommendation module in e-commerce platform.
Described interest model translation interface is by the interest collection form and individual character in common user interest model in client
The interest collection form changed in recommending module is mutually changed.
Common user interest model I of the described interest model management module to user uuBasic operation is carried out, for giving
Determine operation function f and n dimension common user interest model vectorF is acted onComputing be designated asWherein, I1, I2..., InIt is user u1, u2..., unCommon user interest model;
Described basic operation is specifically included:
1) user interest inquiry:From common user interest model IuThe interest collection of middle inquiry user u, is designated as read (Iu);
2) user interest addition:To the common user interest model I of user uuThe new interest collection S of middle additionα, it is designated as
insert(Iu, Sα);
3) user interest updates:By the common user interest model I of user uuIn interest collection SαIt is updated to S 'α, it is designated as
update(Iu, Sα, S 'α);
4) user interest is deleted:Delete the common user interest model I of user uuIn interest collection Sα, it is designated as delete
(Iu, Sα);
5) user interest matching:The common user interest model I of given user uuWith the common user interest model of user v
Iv, similarity between the two is calculated, it is designated as similarity (Iu, Iv)。
Compared with prior art, the present invention has advantages below:
1) present invention proposes and employs a kind of general user interest descriptive model and corresponding arithmetic operation, for
Common interesting measure method and form are used between different platforms, in order to understanding mutually and exchanging.It can reflect
The different Characteristic of Interest that each user has, including it is user behavior feature, professional domain, ability grade, Long-term Interest, short-term
Association between interest, interest etc., with accuracy, versatility, it is easy to promote and realize.
2) present invention proposes and employs the system that user interest is obtained and propagated, including interest model translation interface, real
The function of the cross-platform multiplexing of user interest is showed.In the individuation data that a platform is obtained, by interest model translation interface
Conversion, the user interest descriptive model as standard can be supplied to other platforms to use, and as user is on platform
Historical record increase, interest model can be continuously available renewal.
3) present invention proposes and employs that a kind of brand-new user interest is obtained and transmission method, can overcome existing individual character
Change commending system for the platform dependence and interesting data locality of user interest and openness defect.User is to electronics
On business platform during accomodation of activities, it will be actively broadcast to platform by client from the interest model for carrying, and in turn again
The interest model of oneself can be updated using the new data of platform excavation, obtain it constantly perfect, with high efficiency.
Brief description of the drawings
Fig. 1 is that user interest of the present invention is obtained and transmission method flow chart;
Fig. 2 is that the present invention obtains initial common user interest model method flow diagram;
Fig. 3 is that user interest of the present invention is obtained and broadcasting system structural representation;
Fig. 4 is user interest acquisition and communication process schematic diagram in the embodiment of the present invention.
In figure:1st, client, 2, e-commerce platform, 3, memory module, 4, interest model management module, 5, interest model
Translation interface, 6, personalized recommendation module.
Specific embodiment
The present invention is described in detail with specific embodiment below in conjunction with the accompanying drawings.The present embodiment is with technical solution of the present invention
Premised on implemented, give detailed implementation method and specific operating process, but protection scope of the present invention is not limited to
Following embodiments.
As shown in figure 1, obtaining the method with propagating using a kind of user interest, realize accessing e-commerce platform in user
Carry out interest acquisition when 2 in client 1 and e-commerce platform 2 between commending system, propagate and update, comprise the following steps:
Step S1:Client 1 sets up the memory space of storage common user interest model, and by common user interest model
In interest collection be initialized as empty set.The standard and framework of a set of interesting measure of common user interest model, store in model
Interest collection is to need to propagate and update, and its method for building up includes:
101:Set up the basic structure of description user interest feature, i.e., first interest, each first interest is by several key-value pairs
Composition, is expressed as first interest in=∪ < ki, vi> (i=1,2,3 ...), ki, viIt is key assignments, by all first interest composition set
It is first interest complete or collected works, is designated as INall={ in1, in2..., inN, wherein N is the length of first interest complete or collected works, then emerging by several yuan
Interest composition collection is combined into first interest collection, is designated as IN, and
102:Set up the structure of correlation between description first interest, i.e. interest relation, each interest relation is with a ternary
Group expression, is expressed as interest relation rij=< ini, inj, ρ > (ini, inj∈ IN), wherein ρ represents first interest iniAnd injBetween
Relation function, and rijAnd rjiIt is different;
103:By step 101,102, for first interest collection IN, there is one group of interest relation, be designated as:R=∪ rij=∪ <
ini, inj, ρ > (ini, inj∈ IN), first interest collection IN interest relation R corresponding with its is referred to as user interest collection, is designated as S=
(IN, R), then for user u, its common user interest model is expressed as an interest collection or the union of several interest collection, note
For:
Step S2:Based on common user interest model, user interest investigation result or electronics business of the client 1 according to feedback
The historical interest collection of business platform 2, initial interest collection in generation common user interest model.As shown in Fig. 2 being specially:
201:Client 1 judges whether user would like to do interest questionnaire investigation, specially judges whether to receive the use of feedback
Family interest questionnaire investigation, if so, then performing step 205, if it is not, then user accesses e-commerce platform 2, performs step 202;
202:Set up between client 1 and the personalized recommendation module 6 of e-commerce platform 2 and communicated;
203:Judge in current Personalized recommending module 6 whether the historical interest collection comprising active user, if so, then performing
204, if it is not, then user accesses other e-commerce platforms 2, redirect 202;
204:Client 1 obtains historical interest collection from current Personalized recommending module 6, generates common user interest model
In initial interest collection;
205:Client 1 generates the initial interest collection of common user interest model according to user interest Questionnaire results.
Step S3:User accesses e-commerce platform 2.
Step S4:Client 1 proposes that propagating user interest asks to the personalized recommendation module 6 of current E-commerce platform 2
Ask, current Personalized recommending module 6 receives common user interest model in client 1 by interest model translation interface 5, according to
Interest collection in client 1 in common user interest model updates the interest collection of correspondence user in current Personalized recommending module 6.
Step S5:Judge correspondence user in current Personalized recommending module 6 interest collection whether with general use in client 1
Interest collection in the interest model of family is identical, if it is not, then performing step S6, if so, interest is propagated completing, performs step S7.
Step S6:The new interest collection of correspondence user is extracted from current Personalized recommending module 6, is changed by interest model
Interface 5 is sent to client 1, updates the interest collection in common user interest model in client 1, and interest updates and completes.
Step S7:User continues to access other e-commerce platforms 2, and client 1 is pushed away with the personalization of e-commerce platform 2
Recommend module 6 to be communicated, jump procedure S4.
As shown in figure 3, being obtained and the system propagated, including e-commerce platform using the user interest that the above method is realized
2 personalized recommendation module 6, also including client 1 and interest model translation interface 5, client 1 includes memory module 3 and emerging
Interesting model management module 4, the connection interest model of memory module 3 management module 4, for being preserved with common user interest model form
User interest collection, the connection interest model of interest model management module 4 translation interface 5, interest model translation interface 5 is located at electronics business
In business platform 2, connection personalized recommendation module 6.
Interest model translation interface 5 is by the common user interest model form in client 1 and personalized recommendation module 6
Interest collection form is mutually changed.I.e. it is that a kind of to be used for the user in common user interest model with specific commending system emerging
The part interacted between interesting data, consolidation form and specific interest number in the commending system for realizing user interest
According to the mutual conversion between form and mutual operation.The function of the interface includes:
A. interest model is parsed:Interest collection is converted into personalized recommendation mould from the consolidation form of common user interest model
The specific format of block 6.
B. interest model is reduced:Interest collection is changed into common user interest from the specific format of personalized recommendation module 6
The consolidation form of model.
Common user interest model I of the interest model management module 4 to user uuBasic operation is carried out, for giving computing
Function f and n dimension common user interest model vectorF is acted onComputing be designated asWherein,
I1, I2..., InIt is user u1, u2..., unCommon user interest model.
Basic operation includes:
1) user interest inquiry:From common user interest model IuThe interest collection of middle inquiry user u, is designated as read (Iu);
2) user interest addition:To the common user interest model I of user uuThe new interest collection S of middle additionα, it is designated as
insert(Iu, Sα);
3) user interest updates:By the common user interest model I of user uuIn interest collection SαIt is updated to S 'α, it is designated as
update(Iu, Sα, S 'α);
4) user interest is deleted:Delete the common user interest model I of user uuIn interest collection Sα, it is designated as delete
(Iu, Sα);
5) user interest matching:The common user interest model I of given user uuWith the common user interest model of user v
Iv, similarity between the two is calculated, it is designated as similarity (Iu, Iv)。
It is apparent from, above-mentioned basic operation is all general computingSpecial shape.
Therefore, each user possesses a memory space in its client 1, for depositing common user interest model,
Interest model management module 4 then provides the arithmetic operation interface of the model.User is participating in various ecommerce using client 1
While movable, equivalent to " carrying with " interest model of oneself.Client 1 is constantly obtained from e-commerce platform 2
About the data of the user interest, and the model of oneself is updated accordingly.Then new model " carrying with ", travel to behind
In the platform of access.
Based on the above method and system, user u carries own interests collection, is carried out on platform A, B, C, D by client I
In the process of activity, client 1 and corresponding commending system A, B, C, D (including interest model translation interface 5 and personalized recommendation
Module 6) carry out interest acquisition, propagate as follows with the specific workflow for updating, referring to Fig. 4:
A. initialize.When user u first switches on client 1, the foundation storage common user of client 1 interest model (after
Face abbreviation interest model) memory space, and empty set is initialized with by interest model management module 4.
B. interest is obtained.Can be obtained by carrying out survey to user, it is also possible to from the platform that user accessed
In obtain.If user u is unwilling to carry out survey, then, in the platform accessed before user accesses, client
1 interest that the user is obtained from the commending system of the platform, and " carrying with ".As shown in (a) part in Fig. 4, when client 1
When proposing to obtain the request of interest collection to commending system A, the interest model translation interface 5 in commending system A is first pushed away from personalization
Extraction interest collection α in module 6 is recommended, reference format is then converted into, finally returned to client 1.
C. interest is propagated.When user u logs in certain platform for the first time, the interest collection that its client 1 " will be carried with "
The platform is broadcast to, so the platform just can immediately provide recommendation service, and this is that impossible accomplish in current platform
's.As shown in (b) part in Fig. 4, when client 1 proposes to propagate the request of interest collection α to platform commending system B, platform is pushed away
The interesting data form that interest collection α is first converted into the interest model translation interface 5 in system B their own is recommended, it is then that this is emerging
Interest collection adds its personalized recommendation module 6.
D. when the platform accessed before user u is logged in, it is possible to propagate its current interest, step ibid, such as Fig. 4
In (c) part shown in.
E. interest updates.After the platform carries on business, the commending system of the platform is according to his activity note for user
Record information calculates new interest coalescence and returns to client 1, the renolation interest model accordingly of client 1.Such as (c) portion in Fig. 4
Shown in point, after user u activity a period of times on platform C, when exiting, client 1 proposes to obtain to the commending system C of platform
Take the request of new interest collection.Personalized recommendation module 6 in commending system C can calculate the new interest collection β of the user.Then will
β is sent to interest model translation interface 5, after the interface conversion data form, β is sent into client 1.By treatment, client
Existing interest collection α and interest collection β are merged into new interest collection alpha+beta by end 1.
F. interest is propagated again.Brand-new interest model is traveled to other platforms by client 1.(d) part institute in Fig. 4
Show, when activity is carried out on user u to platform D, its client 1 can pass the interest collection alpha+beta obtained from platform B and platform C
Cast in commending system D, commending system D is merged and updated with interest alpha+beta that will be new and existing interest collection γ.
G. the rest may be inferred, and the client 1 of user u can constantly obtain the new of oneself from the commending system of each platform
Interesting data, and the interest model of oneself carrying is updated accordingly.Conversely, can also by newest interest model be broadcast to herein it
The platform for accessing afterwards, and update the interest of the user in the commending system of the platform.
Claims (6)
1. a kind of user interest is obtained and transmission method, it is characterised in that comprised the following steps:
Step S1:Client sets up the memory space of storage common user interest model, and by common user interest model
Interest collection is initialized as empty set;
Step S2:Based on common user interest model, client is flat according to the user interest investigation result of feedback or ecommerce
The historical interest collection of platform, initial interest collection in generation common user interest model;
Step S3:User accesses e-commerce platform;
Step S4:Client proposes to propagate user interest request to the personalized recommendation module of current E-commerce platform, currently
Personalized recommendation module receives common user interest model in client by interest model translation interface, emerging according to common user
Interest collection in interesting model updates the interest collection of correspondence user in current Personalized recommending module;
Step S5:Judge correspondence user in current Personalized recommending module interest collection whether with common user interest in client
Interest collection in model is identical, if it is not, then performing step S6, if so, interest is propagated completing, performs step S7;
Step S6:The new interest collection of correspondence user is extracted from current Personalized recommending module, by interest model translation interface
Client is sent to, the interest collection in common user interest model in client is updated, interest updates and completes;
Step S7:User continues to access other e-commerce platforms, client and the e-commerce platform of user's current accessed
Personalized recommendation module is communicated, jump procedure S4.
2. a kind of user interest according to claim 1 is obtained and transmission method, it is characterised in that described common user
The method for building up of interest model includes:
101:The basic structure of description user interest feature is set up, i.e., first interest, each first interest is made up of several key-value pairs,
It is expressed as first interest in=∪ < ki,vi> (i=1,2,3 ...), ki,viIt is key assignments, is unit by all first interest composition set
Interest complete or collected works, are designated as INall={ in1,in2,…,inN, wherein N is the length of first interest complete or collected works, then by several yuan of interest group
First interest collection is combined into collection, IN is designated as, and
102:Set up the structure of correlation between description first interest, i.e. interest relation, each interest relation is with a triple table
Show, be expressed as interest relation rij=< ini,inj, ρ > (ini,inj∈ IN), wherein ρ represents first interest iniAnd injBetween pass
It is function, and rijAnd rjiIt is different;
103:By step 101,102, for first interest collection IN, there is one group of interest relation, be designated as:R=∪ rij=∪ < ini,
inj, ρ > (ini,inj∈ IN), first interest collection IN interest relation R corresponding with its is referred to as user interest collection, be designated as S=(IN,
R), then for user u, its common user interest model is expressed as an interest collection or the union of several interest collection, is designated as:
3. a kind of user interest according to claim 1 is obtained and transmission method, it is characterised in that described step S2 tools
Body is:
201:Client judges whether to receive the user interest survey of feedback, if so, step 205 is then performed, if it is not, then using
Family accesses e-commerce platform, performs step 202;
202:Set up between client and the personalized recommendation module of e-commerce platform and communicated;
203:Judge in current Personalized recommending module whether the historical interest collection comprising active user, if so, then perform 204,
If it is not, then user accesses other e-commerce platforms, 202 are redirected;
204:Client obtains historical interest collection from current Personalized recommending module, initial in generation common user interest model
Interest collection;
205:Client generates the initial interest collection of common user interest model according to user interest Questionnaire results.
4. the user interest that a kind of use the method for claim 1 is realized is obtained and broadcasting system, including ecommerce
The personalized recommendation module of platform, it is characterised in that also including client and interest model translation interface, described client bag
Include memory module and interest model management module, described memory module connection interest model management module, described interest mould
Type management module connects interest model translation interface, and described interest model translation interface is connected in e-commerce platform
Personalized recommendation module.
5. a kind of user interest according to claim 4 is obtained and broadcasting system, it is characterised in that described interest model
Translation interface is by the interest collection lattice in the interest collection form in common user interest model in client and personalized recommendation module
Formula is mutually changed.
6. a kind of user interest according to claim 4 is obtained and broadcasting system, it is characterised in that described interest model
Common user interest model I of the management module to user uuBasic operation is carried out, general use is tieed up for given operation function f and n
Family interest model vectorF is acted onComputing be designated asWherein, I1,I2,…,InIt is user
u1,u2,…,unCommon user interest model;
Described basic operation is specifically included:
1) user interest inquiry:From common user interest model IuThe interest collection of middle inquiry user u, is designated as read (Iu);
2) user interest addition:To the common user interest model I of user uuThe new interest collection S of middle additionα, it is designated as insert
(Iu,Sα);
3) user interest updates:By the common user interest model I of user uuIn interest collection SαIt is updated to S 'α, it is designated as update
(Iu,Sα,S′α);
4) user interest is deleted:Delete the common user interest model I of user uuIn interest collection Sα, it is designated as delete (Iu,Sα);
5) user interest matching:The common user interest model I of given user uuWith the common user interest model I of user vv, meter
Similarity between the two is calculated, similarity (I are designated asu,Iv)。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410494809.4A CN104268760B (en) | 2014-09-24 | 2014-09-24 | A kind of user interest is obtained and transmission method and its system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410494809.4A CN104268760B (en) | 2014-09-24 | 2014-09-24 | A kind of user interest is obtained and transmission method and its system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104268760A CN104268760A (en) | 2015-01-07 |
CN104268760B true CN104268760B (en) | 2017-06-13 |
Family
ID=52160280
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410494809.4A Active CN104268760B (en) | 2014-09-24 | 2014-09-24 | A kind of user interest is obtained and transmission method and its system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104268760B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105740381B (en) * | 2016-01-27 | 2019-05-17 | 北京工业大学 | A method of user interest is excavated based on complex network characteristic and neural network clustering |
CN110796505B (en) * | 2018-08-03 | 2023-07-04 | 淘宝(中国)软件有限公司 | Business object recommendation method and device |
CN109389427B (en) * | 2018-09-26 | 2023-11-21 | 中国平安人寿保险股份有限公司 | Questionnaire pushing method, questionnaire pushing device, computer device and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1967533A (en) * | 2006-07-17 | 2007-05-23 | 北京航空航天大学 | Gateway personalized recommendation service method and system introduced yuan recommendation engine |
CN101271558A (en) * | 2008-05-16 | 2008-09-24 | 华东师范大学 | Multi-policy commercial product recommending system based on context information |
CN101339562A (en) * | 2008-08-15 | 2009-01-07 | 北京航空航天大学 | Portal personalized recommendation service system introducing into interest model feedback and update mechanism |
CN103488705A (en) * | 2013-09-06 | 2014-01-01 | 电子科技大学 | User interest model incremental update method of personalized recommendation system |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7613664B2 (en) * | 2005-03-31 | 2009-11-03 | Palo Alto Research Center Incorporated | Systems and methods for determining user interests |
-
2014
- 2014-09-24 CN CN201410494809.4A patent/CN104268760B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1967533A (en) * | 2006-07-17 | 2007-05-23 | 北京航空航天大学 | Gateway personalized recommendation service method and system introduced yuan recommendation engine |
CN101271558A (en) * | 2008-05-16 | 2008-09-24 | 华东师范大学 | Multi-policy commercial product recommending system based on context information |
CN101339562A (en) * | 2008-08-15 | 2009-01-07 | 北京航空航天大学 | Portal personalized recommendation service system introducing into interest model feedback and update mechanism |
CN103488705A (en) * | 2013-09-06 | 2014-01-01 | 电子科技大学 | User interest model incremental update method of personalized recommendation system |
Also Published As
Publication number | Publication date |
---|---|
CN104268760A (en) | 2015-01-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Wang et al. | Assessing hotel-related smartphone apps using online reviews | |
CN104166668B (en) | News commending system and method based on FOLFM models | |
US9483580B2 (en) | Estimation of closeness of topics based on graph analytics | |
US9348916B2 (en) | Method and system for providing search services for a social media ecosystem | |
Leung et al. | CLR: a collaborative location recommendation framework based on co-clustering | |
CN103918001B (en) | Comprehension effects of communications propagated through social networking system | |
US20170351774A1 (en) | Real-time online-learning object recommendation engine | |
US7885986B2 (en) | Enhanced browsing experience in social bookmarking based on self tags | |
US20150066624A1 (en) | Method and system for providing a social media ecosystem cooperative marketplace | |
CN106202331A (en) | The commending system of secret protection and operational method based on this commending system by different level | |
CN107480714B (en) | Cross-social network user identification method based on full-view characteristics | |
CN105095279B (en) | File recommendation method and device | |
CN104899763A (en) | Personalized recommendation method based on bilateral diffusion of bipartite network | |
US9292886B2 (en) | Method and system for providing social media ecosystem classified listings | |
US9639581B1 (en) | Query completions | |
CN105142028A (en) | Television program content searching and recommending method oriented to integration of three networks | |
CN107301247B (en) | Method and device for establishing click rate estimation model, terminal and storage medium | |
US20150066689A1 (en) | Method and system for providing social media ecosystem compensation | |
Daskalopoulos et al. | Borderline estimates for fully nonlinear elliptic equations | |
CN106372101A (en) | Video recommendation method and apparatus | |
Saïdi et al. | Relaxation of optimal control problems involving time dependent subdifferential operators | |
CN104268760B (en) | A kind of user interest is obtained and transmission method and its system | |
JP5264246B2 (en) | Information recommendation device and computer program | |
Zhang et al. | Proposing a new friend recommendation method, FRUTAI, to enhance social media providers' performance | |
CN104123321B (en) | A kind of determining method and device for recommending picture |
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 |