CN108132963A - Resource recommendation method and device, computing device and storage medium - Google Patents

Resource recommendation method and device, computing device and storage medium Download PDF

Info

Publication number
CN108132963A
CN108132963A CN201711180121.9A CN201711180121A CN108132963A CN 108132963 A CN108132963 A CN 108132963A CN 201711180121 A CN201711180121 A CN 201711180121A CN 108132963 A CN108132963 A CN 108132963A
Authority
CN
China
Prior art keywords
resource
user
recommendation
recommended
represent
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201711180121.9A
Other languages
Chinese (zh)
Inventor
潘岸腾
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Alibaba China Co Ltd
Original Assignee
Guangzhou Youshi Network Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Guangzhou Youshi Network Technology Co Ltd filed Critical Guangzhou Youshi Network Technology Co Ltd
Priority to CN201711180121.9A priority Critical patent/CN108132963A/en
Publication of CN108132963A publication Critical patent/CN108132963A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9535Search customisation based on user profiles and personalisation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00Computing arrangements using knowledge-based models
    • G06N5/02Knowledge representation; Symbolic representation

Abstract

The present invention discloses a kind of resource recommendation method based on belief network and device, computing device and storage medium.The resource recommendation method includes:The resource adopted using user is trained the data model for being used for resource recommendation, obtains Rating Model of the user to resource to be recommended;Based on the Rating Model, scoring of the user to resource to be recommended is calculated;According to scoring sequence from high to low, resource to be recommended is ranked up;By sort near preceding predetermined quantity resource recommendation to user.Utilize the resource recommendation method and device based on belief network of the present invention, it can be ensured that the resource set of recommendation is maximized by the success rate that user adopts, so as to improve the personalization level of recommendation results.

Description

Resource recommendation method and device, computing device and storage medium
Technical field
The present invention relates to resource recommendation technical field, more particularly to a kind of resource recommendation method based on belief network and Device, computing device and storage medium.
Background technology
With the development of Internet technology, the information content on internet is growing day by day.In face of the information of magnanimity, how is user It rapidly searches for becoming very difficult to oneself required information.Resource recommendation technology is a kind of important information filtering means, Can the interested information of user automatically be searched by information filtering, so as to provide effective personalized clothes to the user Business.For example, during the operation of application shop, there are some scenes to need to like recommendation a batch application according to user interest, such as: " guessing that you like ".
Existing resource recommendation technology can be divided into three kinds of content-based recommendation, collaborative filtering, mixing recommendation recommendation skills Art.Content-based recommendation technology refers to the preference information according to user's history, recommends the resource with like attribute.It is insufficient In terms of part is the unicity of recommendation resource and there are problems that the Content Feature Extraction to multimedia resource, therefore the technology It is chiefly used in the recommendation of web page resources.Collaborative filtering recommending technology has the user group of same interest hobby by searching with user, The resource that other users are liked into user's recommended user's group.For example, collaborative filtering recommending technology is used to be given according to current application User recommends the relevant application of a batch, specially:Limit applying for recommendation first is having same label, Ran Houtong with intended application It crosses the download of user, browsing, the user behavior space vector that behavioral datas establish each application has been installed etc., finally according to cosine system Number (or the German number of outstanding card, Pearson's coefficient etc.) calculates the similarity value for recommending application and intended application, takes similarity ranking most A batch application of front is as recommendation collection.Collaborative filtering recommending technology is there is also many problems, for example, to new user or new resources Cold start-up problem during recommendation, the sparse sex chromosome mosaicism of score data and the scalability problem of algorithm etc..Mix recommended technology It is that both the above recommended technology is applied in combination, it is intended to make up the deficiency of various recommended technologies.So far, collaborative filtering recommending skill Art is one of recommended technology the most successful in e-commerce field.But in the individualized resource field for being divided into target with higher assessment Scape, traditional collaborative filtering recommending technology are to define similarity between article by cosine relative coefficient, and this mode is simultaneously It cannot be guaranteed that the optimal solution of high scoring.
Invention content
In order to overcome the above-mentioned deficiency in collaborative filtering recommending technology, the present invention provides a kind of resource based on belief network Recommend method and apparatus, to ensure that the resource set recommended is maximized by the success rate that user adopts.
To achieve these goals, the first aspect of the present invention provides a kind of resource recommendation method based on belief network, The resource recommendation method includes the following steps:The resource adopted using user carries out the data model for being used for resource recommendation Training, obtains Rating Model of the user to resource to be recommended;Based on the Rating Model, calculate user and resource to be recommended is commented Point;According to scoring sequence from high to low, resource to be recommended is ranked up;The resource to sort near preceding predetermined quantity is pushed away It recommends to user.
The second aspect of the present invention provides a kind of resource recommendation device based on belief network, the resource recommendation device packet It includes:Training module for being trained using the resource that user has adopted to the data model for being used for resource recommendation, obtains user To the Rating Model of resource to be recommended;Computing module comments resource to be recommended for calculating user based on the Rating Model Point;Sorting module, for being ranked up according to scoring sequence from high to low to resource to be recommended;Recommending module, for that will arrange Sequence near preceding predetermined quantity resource recommendation to user.
The third aspect of the present invention provides a kind of computing device, which includes:Processor;And memory, On be stored with executable code, when the executable code is performed by the processor, the processor is made to perform the present invention The resource recommendation method that is provided of first aspect.
The fourth aspect of the present invention provides a kind of non-transitory machinable medium, is stored thereon with executable generation Code when the executable code is performed by the processor of electronic equipment, makes the processor perform the first aspect of the present invention The resource recommendation method provided.
Technical scheme of the present invention, the resource adopted by using user, to be used for the data model of resource recommendation into Row training, obtains Rating Model of the user to resource to be recommended, it is ensured that the success rate that the resource set of recommendation is adopted by user is maximum Change, so as to improve the personalization level of recommendation results.
Description of the drawings
Exemplary embodiment of the invention is described in more detail in conjunction with the accompanying drawings, it is of the invention above-mentioned and its Its purpose, feature and advantage will be apparent, wherein, in exemplary embodiment of the invention, identical reference label Typically represent same parts.
Fig. 1 is the flow chart of the resource recommendation method according to the present invention based on belief network.
Fig. 2 is the schematic diagram of the belief network of resource built according to the present invention.
Fig. 3 is the structure diagram of the resource recommendation device according to the present invention based on belief network.
Specific embodiment
The preferred embodiment of the present invention is more fully described below with reference to accompanying drawings.Although the present invention is shown in attached drawing Preferred embodiment, however, it is to be appreciated that may be realized in various forms the present invention without the embodiment party that should be illustrated here Formula is limited.On the contrary, these embodiments are provided so that the present invention is more thorough and complete, and can be by the present invention's Range is completely communicated to those skilled in the art.
Before technical scheme of the present invention is specifically described, the term mentioned in the present invention is fitted first When explanation." resource " mentioned herein refers to so that computer or processor or other function execution means are able to carry out The component software or hardware mechanism of specific function, including multimedia (for example, audio, video, picture etc.), document, using and Other online Internet resources." belief network " mentioned herein refers to a kind of uncertainty based on probability analysis, graph theory The model of expression and the reasoning of knowledge, also referred to as Bayesian network, Belief Network or causal net.It is mentioned herein that " electronics is set It is standby " refer to it is any with the equipment of electronic unit used by terminal user or consumer, including mobile phone, personal digital assistant (PDA), computer, tablet computer, music player, camera, video recorder, electronic reader, radio set equipment and trip Gaming machine.
In order to make technical scheme of the present invention clearer, clear, below with reference to accompanying drawings and combine to electronic equipment use Recommend the example of resource that the preferred embodiment of the present invention is described in detail in family.
Fig. 1 is the flow chart of the resource recommendation method according to the present invention based on belief network.The resource recommendation method can Suitable for recommending the situation of online Internet resources to electronic device user, include but not limited to recommend multimedia, document, application The situation of APP.As shown in Figure 1, the resource recommendation method according to the present invention based on belief network includes:
Step S101:The resource adopted using user is trained the data model for being used for resource recommendation, is used Family is to the Rating Model of resource to be recommended.The step can also be referred to as the model training stage.For example, (compare to electronic equipment Such as, mobile phone) user recommend application in the case of, the stage can by using installed application build application belief network.Tool Body, based on the assumption that build the belief network of resource shown in Fig. 2:User adopts the probability of a certain resource recommendation by working as The preceding all real estate impacts adopted, that is, if user has adopted the recommendation of resource, then user will adopt resource Probability is influenced by the resource adopted.For example, in the case where recommending application to electronic equipment (for example, mobile phone) user, The probability of a certain resource of user installation influenced by all applications being currently installed on, that is, if user installation, using A, B, C are so The probability of user installation application D is by installation using A, the influence of B, C.
Assuming that U represents all resource sample sets, u represents user, and m represents the number of all resources in resources bank, i, j tables Show resource to be recommended, qj,iRepresent belief network parameter, lu,jRepresent normalized scores of the user u to resource j, that is,Here, vu,jIt represents whether user u has adopted resource j, works as vu,jRepresent that user u has adopted resource when=0 J works as vu,jRepresent that user u does not adopt resource j when=1.scoreu,iRepresent user u to the scoring of resource i (for example, for table Requisition family u likes degree to resource i), calculation formula is as follows:
There is certain stable, pervasive relationship, i.e., the relationship q between resource and resource between resource and resourcej,iFor not It is consistent with user.Build belief network process it is practical be exactly correlation q between computing resource and resourcej,i.Separately Outside, it is noted that, due to considering user behavior path, so the belief network of resource is digraph, i.e. qj,i≠qi,j.For example, with Family is mounted with the probability using B using installation after A, is using the probability of A not equal to installation after user installation application B.Then, The parameter q of belief network is calculated by following two stepj,i
The first step:Likelihood function (also referred to as loss function) g (q) is constructed, i.e.,
Second step:Seek parameter q during minimum value (object function is min g (q)) of likelihood functionj,iTo get arriving resource Belief network.
In the present embodiment the parameter q during minimum value of likelihood function is sought using gradient descent methodj,i, still, this field Technical staff can also seek the parameter q during minimum value of likelihood function using other equivalent algorithms known in the fieldj,i。 Gradient descent method is specially as follows:
Step (a):One group of transfer matrix { q between 0 to 1 is given at randomi,j| i, j ∈ I }, it is set as q(0), initialization changes Ride instead of walk several k=0;
Step (b):Iterative calculationWherein θ is the step number of iteration, for example, can be with Take θ=0.01;
Step (c):Judge Δ g (q(k+1))=| g (q(k+1))-g(q(k)) | whether restrain, if | Δ g (q(k+1))-Δg(q(k)) | < α then return to q(k+1), otherwise the transfer matrix as estimated back to step (b) and continues to calculate, wherein α is one The value of very little, for example, the θ of α=0.01 can be taken.
Step S102:Based on the obtained Rating Models of step S101, i.e.,Calculate user Scoring to resource to be recommended.For example, in the case where recommending application to electronic equipment (for example, mobile phone) user, the scoring mould Type is clicking rate model of the user to different application.According to above-mentioned formula, based on different application and the corresponding feature of user to Amount, you can obtain click probability of the user to different application.
Step S103:According to scoring sequence from high to low, resource to be recommended is ranked up.
Step S104:By sort near preceding predetermined quantity resource recommendation to user.
For example, in the case where recommending application to electronic equipment (for example, mobile phone) user, if all application collection of resources bank A is combined into, can estimate user in advance by step S101 and step S102 clicks clicking rate of any one in set A using b scoreu,i.Then, according to clicking rate scoreu,iDescending is done to set A, intercept near 100 preceding applications and is recommended User.
Above resource to be recommended to be pushed away as example to describe the resource based on belief network of the present invention to electronic device user Recommend method.Those skilled in the art can along the exemplary thinking under other scenes using the present invention based on conviction net The resource recommendation method of network.
Compared to traditional collaborative filtering method, the resource recommendation method based on belief network of the present embodiment passes through construction Loss function simultaneously asks the minimum value of loss function to acquire optimal solution, it is ensured that the success rate that the resource set of recommendation is adopted by user is most Bigization, so as to improve the personalization level of recommendation results.Specifically, in the case where recommending application to electronic device user, lead to It crosses construction installation rate and estimates function, then construct loss function, optimal solution is acquired by the minimum value for seeking loss function, it is ensured that this The recommendation collection that kind mode calculates is mounted to power maximization.Moreover, the present embodiment by be exposed to the conversion ratio of installation compared to Traditional collaborative filtering method has to be promoted by a relatively large margin.
In addition, the present invention also provides a kind of resource recommendations for being used to implement the above-mentioned resource recommendation method based on belief network Device.As shown in figure 3, resource recommendation device 300 includes training module 301, computing module 302, sorting module 303 and recommends mould Block 304.
Training module 301 is used to be trained the data model for being used for resource recommendation using the resource that user has adopted, Obtain Rating Model of the user to resource to be recommended.Rating Model can be expressed as:
Wherein, u represents user, and m represents the number of all resources in resources bank, and i, j represent resource to be recommended, qj,iIt represents Belief network parameter, lu,jRepresent normalized scores of the user u to resource j, that is,Here, vu,jIt represents to use Whether family u has adopted resource j, works as vu,jRepresent that user u has adopted resource j, works as v when=0u,jRepresent that user u does not adopt when=1 Resource j.
Computing module 302 is used to calculate scoring of the user to resource to be recommended based on the Rating Model.Computing module can be wrapped specifically It includes:Structural unit, for constructing likelihood function g (q), i.e., Wherein, U represents all resource sample sets;Unit is solved, for seeking the parameter q during minimum value of likelihood functionj,i.It solves single Member can seek the parameter q during minimum value of likelihood function g (q) by performing following stepsj,i
Step (a):One group of transfer matrix { q between 0 to 1 is given at randomi,j| i, j ∈ I }, it is set as q(0), initialization changes Ride instead of walk several k=0;
Step (b):Iterative calculationWherein θ is the step number of iteration, for example, can be with Take θ=0.01;
Step (c):Judge Δ g (q(k+1))=| g (q(k+1))-g(q(k)) | whether restrain, if | Δ g (q(k+1))-Δg(q(k)) | < α then return to q(k+1), otherwise the transfer matrix as estimated back to step (b) and continues to calculate, wherein α is one The value of very little, for example, the θ of α=0.01 can be taken.
Sorting module 303 is used to be ranked up resource to be recommended according to scoring sequence from high to low.Recommending module 304 For the resource recommendation that will sort near preceding predetermined quantity to user.
Above by reference to attached drawing be described in detail the resource recommendation method according to the present invention based on belief network and Device.
In addition, it is also implemented as a kind of computer program or computer program product according to the method for the present invention, the meter Calculation machine program or computer program product include the calculating of above steps limited in the above method for performing the present invention Machine program code instruction.
In addition, the present invention can also be embodied as a kind of computing device, which includes:Processor;And memory, Executable code is stored thereon with, when the executable code is performed by the processor, the processor is made to perform basis Each step of the method for the present invention.
Alternatively, the present invention can also be embodied as a kind of (or the computer-readable storage of non-transitory machinable medium Medium or machine readable storage medium), executable code (or computer program or computer instruction code) is stored thereon with, When the executable code (or computer program or computer instruction code) is by electronic equipment (or computing device, server When) processor perform when, the processor is made to perform each step of the above method according to the present invention.
Those skilled in the art will also understand is that, with reference to the described various illustrative logical blocks of disclosure herein, mould Block, circuit and algorithm steps may be implemented as the combination of electronic hardware, computer software or both.
Flow chart and block diagram in attached drawing show that the possibility of the system and method for multiple embodiments according to the present invention is real Existing architectural framework, function and operation.In this regard, each box in flow chart or block diagram can represent module, a journey A part for sequence section or code, as defined in the part of the module, program segment or code is used to implement comprising one or more The executable instruction of logic function.It should also be noted that in some implementations as replacements, the function of being marked in box also may be used To be occurred with being different from the sequence marked in attached drawing.For example, two continuous boxes can essentially perform substantially in parallel, They can also be performed in the opposite order sometimes, this is depended on the functions involved.It is also noted that block diagram and/or stream The combination of each box in journey figure and the box in block diagram and/or flow chart can use functions or operations as defined in performing Dedicated hardware based system realize or can be realized with the combination of specialized hardware and computer instruction.
Various embodiments of the present invention are described above, above description is exemplary, and non-exclusive, and It is not limited to disclosed each embodiment.In the case of without departing from the scope and spirit of illustrated each embodiment, for this skill Many modifications and changes will be apparent from for the those of ordinary skill in art field.The selection of term used herein, purport In the principle for best explaining each embodiment, practical application or to the improvement of the technology in market or make the art Other those of ordinary skill are understood that each embodiment disclosed herein.

Claims (14)

1. a kind of resource recommendation method based on belief network, includes the following steps:
The resource adopted using user is trained the data model for being used for resource recommendation, obtains user to money to be recommended The Rating Model in source;
Based on the Rating Model, scoring of the user to resource to be recommended is calculated;
According to scoring sequence from high to low, resource to be recommended is ranked up;
By sort near preceding predetermined quantity resource recommendation to user.
2. resource recommendation method according to claim 1, which is characterized in that the Rating Model is expressed as:
Wherein, u represents user, and m represents the number of all resources in resources bank, and i, j represent resource to be recommended, qj,iRepresent conviction Network parameter, lu,jRepresent normalized scores of the user u to resource j, that is,Here, vu,jRepresenting user u is It is no to have adopted resource j, work as vu,jRepresent that user u has adopted resource j, works as v when=0u,jRepresent that user u does not adopt resource when=1 j。
3. resource recommendation method according to claim 2, which is characterized in that acquire belief network parameter as follows qj,i
Likelihood function g (q) is constructed, i.e.,Wherein, U Represent all resource sample sets;
Seek the parameter q during minimum value of likelihood functionj,i
4. resource recommendation method according to claim 3, which is characterized in that seek likelihood function g (q) as follows Minimum value when parameter qj,i
Step (a):One group of transfer matrix { q between 0 to 1 is given at randomi,j| i, j ∈ I }, it is set as q(0), initialize iteration step Number k=0;
Step (b):Iterative calculationWherein θ is the step number of iteration;
Step (c):Judge Δ g (q(k+1))=| g (q(k+1))-g(q(k)) | whether restrain, if | Δ g (q(k+1))-Δg(q(k))| < α, then return to q(k+1), otherwise the transfer matrix as estimated back to step (b) and continues to calculate, wherein α is a very little Value.
5. resource recommendation method according to any one of claim 1 to 4, which is characterized in that the resource includes more matchmakers Body, document, using and other online Internet resources, also, it is described scoring correspond to user to the click of the resource, download, The probability of browsing, installation or collection.
6. resource recommendation method according to any one of claim 1 to 4, which is characterized in that the resource is to electronics The application of equipment recommendation, also, the scoring corresponds to click probability of the user to the application.
7. a kind of resource recommendation device based on belief network, including:
Training module for being trained using the resource that user has adopted to the data model for being used for resource recommendation, is used Family is to the Rating Model of resource to be recommended;
Computing module, for calculating scoring of the user to resource to be recommended based on the Rating Model;
Sorting module, for being ranked up according to scoring sequence from high to low to resource to be recommended;
Recommending module, for the resource recommendation that will sort near preceding predetermined quantity to user.
8. resource recommendation device according to claim 7, which is characterized in that the Rating Model is expressed as:
Wherein, u represents user, and m represents the number of all resources in resources bank, and i, j represent resource to be recommended, qj,iRepresent conviction Network parameter, lu,jRepresent normalized scores of the user u to resource j, that is,Here, vu,jRepresenting user u is It is no to have adopted resource j, work as vu,jRepresent that user u has adopted resource j, works as v when=0u,jRepresent that user u does not adopt resource when=1 j。
9. resource recommendation device according to claim 8, which is characterized in that the computing module includes:
Structural unit, for constructing likelihood function g (q), i.e., Wherein, U represents all resource sample sets;
Unit is solved, for seeking the parameter q during minimum value of likelihood functionj,i
10. resource recommendation device according to claim 9, which is characterized in that the solution unit is by performing following step It is rapid to seek the parameter q during minimum value of likelihood function g (q)j,i
Step (a):One group of transfer matrix { q between 0 to 1 is given at randomi,j| i, j ∈ I }, it is set as q(0), initialize iteration step Number k=0;
Step (b):Iterative calculationWherein θ is the step number of iteration;
Step (c):Judge Δ g (q(k+1))=| g (q(k+1))-g(q(k)) | whether restrain, if | Δ g (q(k+1))-Δg(q(k))| < α, then return to q(k+1), otherwise the transfer matrix as estimated back to step (b) and continues to calculate, wherein α is a very little Value.
11. the resource recommendation device according to any one of claim 7 to 10, which is characterized in that the resource includes more Media, document, using and other online Internet resources, also, the scoring correspond to user to the click of the resource, under It carries, browsing, the probability installed or collected.
12. the resource recommendation device according to any one of claim 7 to 10, which is characterized in that the resource is to electricity The application of sub- equipment recommendation, also, the scoring corresponds to click probability of the user to the application.
13. a kind of computing device, including:
Processor;And
Memory is stored thereon with executable code, when the executable code is performed by the processor, makes the processing Device performs the method as any one of claim 1-6.
14. a kind of non-transitory machinable medium, is stored thereon with executable code, when the executable code is electric When the processor of sub- equipment performs, the processor is made to perform such as method according to any one of claims 1 to 6.
CN201711180121.9A 2017-11-23 2017-11-23 Resource recommendation method and device, computing device and storage medium Pending CN108132963A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711180121.9A CN108132963A (en) 2017-11-23 2017-11-23 Resource recommendation method and device, computing device and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711180121.9A CN108132963A (en) 2017-11-23 2017-11-23 Resource recommendation method and device, computing device and storage medium

Publications (1)

Publication Number Publication Date
CN108132963A true CN108132963A (en) 2018-06-08

Family

ID=62389774

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711180121.9A Pending CN108132963A (en) 2017-11-23 2017-11-23 Resource recommendation method and device, computing device and storage medium

Country Status (1)

Country Link
CN (1) CN108132963A (en)

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108804670A (en) * 2018-06-11 2018-11-13 腾讯科技(深圳)有限公司 Data recommendation method, device, computer equipment and storage medium
CN110443636A (en) * 2019-07-15 2019-11-12 阿里巴巴集团控股有限公司 Resource is distributed between user to regulate and control the method and apparatus of its e-payment behavior
CN110889029A (en) * 2018-08-17 2020-03-17 北京京东金融科技控股有限公司 City target recommendation method and device
CN110941727A (en) * 2019-11-29 2020-03-31 北京达佳互联信息技术有限公司 Resource recommendation method and device, electronic equipment and storage medium
CN110995828A (en) * 2019-11-29 2020-04-10 北京邮电大学 Network resource caching method, device and system
CN111191142A (en) * 2018-11-14 2020-05-22 腾讯科技(深圳)有限公司 Electronic resource recommendation method and device and readable medium
CN111340522A (en) * 2019-12-30 2020-06-26 支付宝实验室(新加坡)有限公司 Resource recommendation method, device, server and storage medium
CN111625713A (en) * 2020-04-30 2020-09-04 平安国际智慧城市科技股份有限公司 Resource recommendation method and device based on big data, electronic equipment and medium
CN112199187A (en) * 2019-07-08 2021-01-08 北京字节跳动网络技术有限公司 Content display method and device, electronic equipment and computer readable storage medium
CN116701772A (en) * 2023-08-03 2023-09-05 广东美的暖通设备有限公司 Data recommendation method and device, computer readable storage medium and electronic equipment
CN110889029B (en) * 2018-08-17 2024-04-05 京东科技控股股份有限公司 Urban target recommendation method and device

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103064856A (en) * 2011-10-21 2013-04-24 中国移动通信集团重庆有限公司 Resource recommendation method and device based on belief network
WO2016077127A1 (en) * 2014-11-11 2016-05-19 Massachusetts Institute Of Technology A distributed, multi-model, self-learning platform for machine learning
CN105740430A (en) * 2016-01-29 2016-07-06 大连理工大学 Personalized recommendation method with socialization information fused
CN105975564A (en) * 2016-04-29 2016-09-28 天津大学 Relative entropy similarity-based knowledge recommendation method
CN106202515A (en) * 2016-07-22 2016-12-07 浙江大学 A kind of Mobile solution based on sequence study recommends method and commending system thereof
CN106920147A (en) * 2017-02-28 2017-07-04 华中科技大学 A kind of commodity intelligent recommendation method that word-based vector data drives

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103064856A (en) * 2011-10-21 2013-04-24 中国移动通信集团重庆有限公司 Resource recommendation method and device based on belief network
WO2016077127A1 (en) * 2014-11-11 2016-05-19 Massachusetts Institute Of Technology A distributed, multi-model, self-learning platform for machine learning
CN105740430A (en) * 2016-01-29 2016-07-06 大连理工大学 Personalized recommendation method with socialization information fused
CN105975564A (en) * 2016-04-29 2016-09-28 天津大学 Relative entropy similarity-based knowledge recommendation method
CN106202515A (en) * 2016-07-22 2016-12-07 浙江大学 A kind of Mobile solution based on sequence study recommends method and commending system thereof
CN106920147A (en) * 2017-02-28 2017-07-04 华中科技大学 A kind of commodity intelligent recommendation method that word-based vector data drives

Cited By (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108804670A (en) * 2018-06-11 2018-11-13 腾讯科技(深圳)有限公司 Data recommendation method, device, computer equipment and storage medium
CN108804670B (en) * 2018-06-11 2023-03-31 腾讯科技(深圳)有限公司 Data recommendation method and device, computer equipment and storage medium
CN110889029A (en) * 2018-08-17 2020-03-17 北京京东金融科技控股有限公司 City target recommendation method and device
CN110889029B (en) * 2018-08-17 2024-04-05 京东科技控股股份有限公司 Urban target recommendation method and device
CN111191142A (en) * 2018-11-14 2020-05-22 腾讯科技(深圳)有限公司 Electronic resource recommendation method and device and readable medium
CN111191142B (en) * 2018-11-14 2023-03-14 腾讯科技(深圳)有限公司 Electronic resource recommendation method and device and readable medium
CN112199187A (en) * 2019-07-08 2021-01-08 北京字节跳动网络技术有限公司 Content display method and device, electronic equipment and computer readable storage medium
CN110443636A (en) * 2019-07-15 2019-11-12 阿里巴巴集团控股有限公司 Resource is distributed between user to regulate and control the method and apparatus of its e-payment behavior
CN110995828A (en) * 2019-11-29 2020-04-10 北京邮电大学 Network resource caching method, device and system
CN110941727B (en) * 2019-11-29 2023-09-29 北京达佳互联信息技术有限公司 Resource recommendation method and device, electronic equipment and storage medium
CN110941727A (en) * 2019-11-29 2020-03-31 北京达佳互联信息技术有限公司 Resource recommendation method and device, electronic equipment and storage medium
CN111340522A (en) * 2019-12-30 2020-06-26 支付宝实验室(新加坡)有限公司 Resource recommendation method, device, server and storage medium
CN111340522B (en) * 2019-12-30 2024-03-08 支付宝实验室(新加坡)有限公司 Resource recommendation method, device, server and storage medium
CN111625713A (en) * 2020-04-30 2020-09-04 平安国际智慧城市科技股份有限公司 Resource recommendation method and device based on big data, electronic equipment and medium
CN116701772A (en) * 2023-08-03 2023-09-05 广东美的暖通设备有限公司 Data recommendation method and device, computer readable storage medium and electronic equipment
CN116701772B (en) * 2023-08-03 2024-03-19 广东美的暖通设备有限公司 Data recommendation method and device, computer readable storage medium and electronic equipment

Similar Documents

Publication Publication Date Title
CN108132963A (en) Resource recommendation method and device, computing device and storage medium
Zhou et al. Userrec: A user recommendation framework in social tagging systems
Lu et al. Content-based collaborative filtering for news topic recommendation
Yin et al. A temporal context-aware model for user behavior modeling in social media systems
Liu et al. Personalized travel package recommendation
US20180375949A1 (en) Provisioning personalized content recommendations
Yang et al. Like like alike: joint friendship and interest propagation in social networks
WO2017181612A1 (en) Personalized video recommendation method and device
Shmueli et al. Care to comment? Recommendations for commenting on news stories
JP5454357B2 (en) Information processing apparatus and method, and program
CN110209807A (en) A kind of method of event recognition, the method for model training, equipment and storage medium
WO2018121700A1 (en) Method and device for recommending application information based on installed application, terminal device, and storage medium
CN103577549A (en) Crowd portrayal system and method based on microblog label
CN110008397B (en) Recommendation model training method and device
US20160189218A1 (en) Systems and methods for sponsored search ad matching
US20080270549A1 (en) Extracting link spam using random walks and spam seeds
CN104199836B (en) A kind of mark user model constructing method divided based on sub- interest
CN109918662A (en) A kind of label of e-sourcing determines method, apparatus and readable medium
CN111522886B (en) Information recommendation method, terminal and storage medium
CN107809370B (en) User recommendation method and device
Prando et al. Content-based Recommender System using Social Networks for Cold-start Users.
CN107943910B (en) Personalized book recommendation method based on combined algorithm
CN109117442A (en) A kind of application recommended method and device
Ji et al. Jointly modeling content, social network and ratings for explainable and cold-start recommendation
Au Yeung et al. Capturing implicit user influence in online social sharing

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
TA01 Transfer of patent application right
TA01 Transfer of patent application right

Effective date of registration: 20200901

Address after: 310052 room 508, floor 5, building 4, No. 699, Wangshang Road, Changhe street, Binjiang District, Hangzhou City, Zhejiang Province

Applicant after: Alibaba (China) Co.,Ltd.

Address before: 510627 Guangdong city of Guangzhou province Whampoa Tianhe District Road No. 163 Xiping Yun Lu Yun Ping square B radio tower 15 layer self unit 02

Applicant before: GUANGZHOU UC NETWORK TECHNOLOGY Co.,Ltd.

RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20180608