CN107577823B - The medical information of diversity enhancing recommends method and device - Google Patents
The medical information of diversity enhancing recommends method and device Download PDFInfo
- Publication number
- CN107577823B CN107577823B CN201710954748.9A CN201710954748A CN107577823B CN 107577823 B CN107577823 B CN 107577823B CN 201710954748 A CN201710954748 A CN 201710954748A CN 107577823 B CN107577823 B CN 107577823B
- Authority
- CN
- China
- Prior art keywords
- user
- information
- scoring
- target
- target user
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides the medical informations of species diversity enhancing to recommend method and device, to solve the problems, such as to recommend information content more single in the related technology.This method includes：Obtain first scoring of first user to the pre-selection information in pre-selection information set in the first user set；Wherein, pre-selection information includes medical information, and the similarity between target user and the first user is calculated the second scoring for preselecting information according to the first scoring and target user；According to the diversity metric between calculated similarity calculation target user and the first user；Arest neighbors of the first user of predetermined number as target user is selected according to the diversity metric；Scoring of the target user to pre-selection information is calculated to the scoring of pre-selection information according to the arest neighbors；Target information is determined according to the scoring of calculated pre-selection information, and the target information determined is recommended into target user.The present invention improves the diversity for recommending information content.
Description
Technical field
The medical information enhanced the present invention relates to field of computer technology more particularly to a species diversity recommends method and dress
It sets.
Background technology
Currently, when recommending information to user, it generally need to first determine the keyword that information to be recommended includes, then obtain mesh
The interest keyword for marking user, determines the keyword of information to be recommended and the higher feelings of similarity of target user's interest keyword
Under condition, information to be recommended is recommended and is presented to target user.However, this information way of recommendation, only accounts for money to be recommended
Contacting between the keyword of news and target user's keyword interested does not consider the recommendation problem of content variety, can lead
It causes to recommend information content more single, the covering surface of recommendation can not be extended.
Invention content
For the defects of the relevant technologies, the present invention provides the medical informations of species diversity enhancing to recommend method and dress
It sets, to solve the problems, such as to recommend information content more single in the related technology.
According to an aspect of the present invention, a kind of information recommendation method is provided, including：Obtain the in the first user set
First scoring of one user to the pre-selection information in pre-selection information set；The pre-selection information includes medical information, according to described
First scoring and target user to the second scoring of the pre-selection information calculate the target user and first user it
Between similarity；According to the diversity degree between target user and first user described in the calculated similarity calculation
Magnitude；Select first user of predetermined number as the target user's according to the calculated diversity metric
Arest neighbors；The target user is calculated to the pre-selection information to the scoring for preselecting information according to the arest neighbors of the target user
Scoring；Target information is determined according to the scoring of calculated pre-selection information, and target information is recommended into target user.
Optionally, described according between target user and first user described in the calculated similarity calculation
Diversity metric, including：First user is established for first of the pre-selection information in the pre-selection information set
Rating matrix；First rating matrix is expanded into the second rating matrix, user's classification based on user's classification shape parameter
Shape parameter is the species parameter of first user；Utilize second rating matrix by first user according to clustering algorithm
First user in set is divided into multiple classes, obtains clustering cluster；Calculate the target user with it is all kinds of in the clustering cluster
In user diversity metric.
Optionally, it is described calculate the target user with it is all kinds of in the clustering cluster in user diversity metric,
Including：The diversity between class in the target user and the clustering cluster is calculated using following formula (1)：
Wherein, it is the class V in clustering cluster V that v, which belongs to clustering cluster V, k,kIn number of users, Diversion (u, V) be target
Diversity metric between user u and clustering cluster V, sim (u, v) similarities between user u and user v；
The diversity metric of the target user and the user in the clustering cluster are calculated according to following formula (2)：
Wherein, diversity metrics of the Diversion (u, v) between target user u and user v, λ are adjustable parameter,
When λ is intended to 1, the diversity metric highest between target user u and user v, when λ is not intended to 1, target user u
The similitude highest between user v.
Optionally, the arest neighbors according to the target user calculates the target user couple to the scoring for preselecting information
The scoring of the pre-selection information, including：Scoring of the target user to the advance information is calculated using following formula (3)：
Wherein, rv,iScoring for user v to information i,Scoring for target user u to information i, v belong to the target
User in the arest neighbors n (u) of user.
Optionally, described that institute is calculated to the second scoring of the pre-selection information according to first scoring and target user
The similarity between target user and first user is stated, including：The target user and institute are calculated using following formula (4)
State the similarity between the first user：
Wherein, sim (u, v) represents similarity between user u and user v, Ru,tScoring for user u to information t, Rv,tFor
Scorings of the user v for information t, Iu,vFor the set of the common information of user v and user u, AuIt is user u to described common
The average value of information scoring, AvIt is the average value that user v scores to the common information.
According to another aspect of the present invention, a kind of information recommendation apparatus is provided, including：Acquisition module, for obtaining
First scoring of first user to the pre-selection information in pre-selection information set in first user set, the pre-selection information include
Medical information；First computing module, for being commented the second of the pre-selection information according to first scoring and target user
Divide the similarity calculated between the target user and first user；Second computing module, for according to calculated institute
State the diversity metric between target user described in similarity calculation and first user；Selecting module, by based on
The diversity metric calculated selects arest neighbors of first user of predetermined number as the target user；Third
Computing module, for calculating the target user to described pre- to the scoring for preselecting information according to the arest neighbors of the target user
Select the scoring of information；Recommending module pushes away target information for determining target information according to the scoring of calculated pre-selection information
It recommends to target user.
Optionally, second computing module, including：Unit is established, for establishing first user for described pre-
Select the first rating matrix of the pre-selection information in information set；Expansion unit, for being based on user's classification shape parameter by institute
It states the first rating matrix and is expanded into the second rating matrix, user's classification shape parameter is the species parameter of first user；
Taxon is used for described first in being gathered first user using second rating matrix according to clustering algorithm
Family is divided into multiple classes, obtains clustering cluster；Computing unit, for calculate in the target user and the clustering cluster it is all kinds of in use
The diversity metric at family.
Optionally, the computing unit is used for：The diversity between class in the target user and the clustering cluster makes
It is calculated with following formula (1)：
Wherein, it is the class V in clustering cluster V that v, which belongs to clustering cluster V, k,kIn number of users, Diversion (u, V) be target
Diversity metric between user u and clustering cluster V, sim (u, v) similarities between user u and user v；
The diversity metric of the target user and the user in the clustering cluster are calculated according to following formula (2)：
Wherein, diversity metrics of the Diversion (u, v) between target user u and user v, λ are adjustable parameter,
When λ is intended to 1, the diversity metric highest between target user u and user v, when λ is not intended to 1, target user u
The similitude highest between user v.
Optionally, the third computing module is used for：The target user is calculated to described advance using following formula (3)
The scoring of information：
Wherein, rv,iScoring for user v to information i,Scoring for target user u to information i, v belong to the target
User in the arest neighbors n (u) of user.
Optionally, first computing module is used for：The target user and described first is calculated using following formula (4)
Similarity between user：
Wherein, sim (u, v) represents similarity between user u and user v, Ru,tScoring for user u to information t, Rv,tFor
Scorings of the user v for information t, Iu,vFor the set of the common information of user v and user u, AuIt is user u to described common
The average value of information scoring, AvIt is the average value that user v scores to the common information.
Compared with the relevant technologies, the scoring the present invention is based on user to information, according to neighbour similar with target user
User carries out information recommendation to the preference of information to target user so that the information recommended for user becomes more diversity, solves
It has determined and has recommended the more single problem of information content in the related technology.
Description of the drawings
Technical solution in order to illustrate the embodiments of the present invention more clearly or in the related technology, below will be to embodiment or phase
Attached drawing is briefly described needed in the technology description of pass, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
Other attached drawings are obtained according to these figures.
Fig. 1 is that the medical information of diversity enhancing provided in an embodiment of the present invention recommends the flow chart of method；
Fig. 2 is the block diagram of the medical information recommendation apparatus of diversity enhancing provided in an embodiment of the present invention.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Fig. 1 is that the medical information for the diversity enhancing that one embodiment of the invention provides recommends method, as shown in Figure 1, the party
Method may include handling as follows：
Step 101：The first user obtained in the first user set comments first of pre-selection information in pre-selection information set
Point；
In one exemplary embodiment, pre-selection information can be medical information or may further be portable medical money
News.
Step 102：The second scoring of the information in pre-selection information set is calculated according to the first scoring and target user
Similarity between target user and the first user；
In one exemplary embodiment, step 101 and step 102 can be realized in the following way：
Establish rating matrix Rs of the first user's set U for information set KHK, increase user's classification shape parameter, use this
Shape parameter of classifying expands RHKAnd generate RCHK, using Enhance KP clustering algorithms and utilize RCHKUser's collection U is divided into multiple
Class utilizes matrix RHKCalculate the Pearson came similarity between all users, the meter of the similarity in target user u and user's collection U
It is as follows to calculate formula：
Wherein, sim (u, v) represents similarity between user u and user v, Ru,tRepresent user u commenting for information t
Point, Rv,tRepresent scorings of the user v for information t, Iu,vRefer to that the formula includes the common money of user v and user u
The set of news, AuRepresent the average value that user u scores to common information, AvUser v is represented to be averaged to common information scoring
Value.
Wherein, for above-mentioned rating matrix RHK, can also user further be obtained according to the travel log of user and browses note
The information such as record, like time and information attribute are reprinted in record, the pretreatments such as are weighted to user's unrated information, from
And obtain more complete rating matrix RHK.
Wherein, above-mentioned Enhance KP are obtained by changing K-prototypes algorithms, EnhanceKP algorithm steps
As K-prototypes, include the following steps：
Step (1)：In data set X={ X1,X2,X3…XnIn random selection K data object as in initial clustering
The heart, initial cluster center collection are combined into V={ V1,V2,V3…VK}；
Step (2)：All data object X in data set are calculated successivelyiTo each cluster centre VjDistance d (i, j)；
Step (3)：According to the minimum value of d (i, j) by XiIt is included in corresponding VjAmong the clustering cluster at place；
Step (4)：All XiAfter being assigned, according to categorical attribute value and numeric type attribute value and cluster centre away from
From updating the cluster centre of each cluster；
Step (5) repeats step above-mentioned (2), (3) and (4), until clustering criteria function convergence；
(6) algorithm is completed, and exports final result；
Enhance KP are as follows to calculation formula of the categorical attribute value at a distance from cluster centre in (2)：
Wherein, xniWith vmiRepresent object n and object m, i refer to object n and object m categorical attribute i.DlRepresent cluster
Gather in first of attribute (categorical attribute), the difference value of all properties value and cluster centre and.DlmRepresent object to be sorted
xniThe attribute value of categorical attribute i and the difference value of cluster centre.T be adjustable parameters (be trained by training set, to
Determine it is optimal t).Wherein, the difference value between same alike result value is 0, and the difference value between different attribute value is 1.Therefore data
Object XnWith XmThe distance between calculation formula it is as follows：
Wherein, d1(n, m) is Euclidean distance, d2(n, m) is categorical attribute calculation formula, the subscript i's and V of X herein
Subscript j represents the different value under l attributes.
Illustratively, above-mentioned target user is also a member in first user's set U, therefore it is also after dividing U
In some obtained class.
Step 103：According to the diversity metric between calculated similarity calculation target user and the first user；
In one exemplary embodiment, the diversity metric calculated between target user and the first user can wrap
It includes：Establish first rating matrix of first user for the pre-selection information in pre-selection information set；Based on user's classification shape parameter
First rating matrix is expanded into the second rating matrix, user's classification shape parameter is the species parameter of the first user；According to cluster
The first user during algorithm is gathered the first user using the second rating matrix is divided into multiple classes, obtains clustering cluster；Calculate target
In user and clustering cluster it is all kinds of in user diversity metric.Wherein, user classify shape parameter may include user's gender,
The parameters such as user's schooling.
Calculate it is all kinds of in target user and clustering cluster in the diversity metric of user may be used such as under type：
Wherein, diversity metrics of the Diversion (u, v) between target user u and user v；Diversion(u,
V) calculation is as follows：
The diversity between class in target user and clustering cluster is calculated using following formula：
Wherein, it is the class V in clustering cluster V that v, which belongs to clustering cluster V, k,kIn number of users, Diversion (u, V) be target
Diversity metric between user u and clustering cluster V, sim (u, v) similarities between user u and user v；
The diversity metric of target user and the user in clustering cluster are calculated according to following formula：
Wherein, λ is adjustable parameter, when λ is intended to 1, the diversity metric highest between target user u and user v,
Conversely, the similitude highest between user v of target user u, by that the content recommended can be made more more the adjustment of λ
Sample.
Step 104：Select the first user of predetermined number as target user's according to calculated diversity metric
Arest neighbors；
Determine target user u, according to Diversion (u, v) value of target user u and other users, and according to from greatly to
Arest neighbors of the small k user of sequential selection as target user u forms the nearest-neighbors collection n (u) of target user u.
Step 105：Target user is calculated to pre-selection information to the scoring of pre-selection information according to the arest neighbors of target user
Scoring；
In one exemplary embodiment, target user is calculated to the scoring for preselecting information according to the arest neighbors of target user
To preselect information scoring may include：
Scoring of the target user to advance information is calculated using following formula：
Wherein, rv,iScoring for user v to information i,Scoring for target user u to information i, v belong to target use
User in the arest neighbors n (u) at family.
Step 106：Target information is determined according to the scoring of calculated pre-selection information, and target information is recommended into the mesh
Mark user.
In step 106, scoring of the target user to specified information can be predicted, L highests scorings, which generate, before taking recommends
List provides accurate and multifarious information to the user.
Information provided in an embodiment of the present invention recommends method, based on user to the scoring of information according to similar to target user
The preference of neighbour user carries out information recommendation to target user, is added during calculating the similar neighborhoods of target user various
Sexual factor, to increase the diversity of information recommendation；Based on the scoring of the prediction unvalued information of target user, default is chosen
The high information of number prediction scoring is recommended so that user can know its interested information.
The present invention also provides the medical information recommendation apparatus of species diversity enhancing, which carries for realizing the present invention
The information of confession recommends method, and Fig. 2 is the block diagram of the device, as shown in Fig. 2, the device 20 includes following component part：
Acquisition module 21, for obtaining the first user in the first user set to first of information in pre-selection information set
Scoring；Wherein, pre-selection information includes medical information.
First computing module 22, for according to the first scoring and target user to the of the information in pre-selection information set
Two scorings calculate the similarity between target user and the first user；
Second computing module 23, for according to various between calculated similarity calculation target user and the first user
Property metric；
Selecting module 24, for selecting the first user of predetermined number as target according to calculated diversity metric
The arest neighbors of user；
Third computing module 25, for calculating target user couple to the scoring for preselecting information according to the arest neighbors of target user
Preselect the scoring of information；
Recommending module 26 recommends target information for determining target information according to the scoring of calculated pre-selection information
To the target user.
In one exemplary embodiment, above-mentioned second computing module 23 can be such as lower unit：
Unit is established, the first rating matrix for establishing the first user for the pre-selection information in pre-selection information set；
First rating matrix is expanded into the second rating matrix, user by expansion unit for being based on user's classification shape parameter
Classification shape parameter is the species parameter of the first user；
Taxon, for the first user point in being gathered the first user using the second rating matrix according to clustering algorithm
At multiple classes, clustering cluster is obtained；
Computing unit, for calculate in target user and clustering cluster it is all kinds of in user diversity metric.
Wherein, the diversity between the class in target user and clustering cluster is calculated using following formula：
Wherein, it is the class V in clustering cluster V that v, which belongs to clustering cluster V, k,kIn number of users, Diversion (u, V) be target
Diversity metric between user u and clustering cluster V, sim (u, v) similarities between user u and user v；
Computing unit can be used for：The diversity that target user and the user in clustering cluster are calculated according to following formula is measured
Value：
Wherein, diversity metrics of the Diversion (u, v) between target user u and user v, λ are adjustable parameter,
When λ is intended to 1, the diversity metric highest between target user u and user v, conversely, target user u with user v it
Between similitude highest.
Third computing module 25 is used for following formula and calculates scoring of the target user to advance information：
Wherein, rv,iScoring for user v to information i,Scoring for target user u to information i, v belong to target user
Arest neighbors n (u) in user.
Above-mentioned first computing module 22 can be used for：The phase between target user and the first user is calculated using following formula
Like degree：
Wherein, sim (u, v) represents similarity between user u and user v, Ru,tScoring for user u to information t, Rv,tFor
Scorings of the user v for information t, Iu,vFor the set of user v and user the u information to score, AuIt is user u to Iu,v
The average value of scoring, AvUser v is represented to Iu,vThe average value of scoring.
Information recommendation apparatus provided in an embodiment of the present invention, based on user to the scoring of information according to similar to target user
The preference of neighbour user carries out information recommendation to target user, is added during calculating the similar neighborhoods of target user various
Sexual factor, to increase the diversity of information recommendation；Based on the scoring of the prediction unvalued information of target user, default is chosen
The high information of number prediction scoring is recommended so that user can know its interested information.
In the specification of the present invention, numerous specific details are set forth.It is to be appreciated, however, that the embodiment of the present invention can be with
It puts into practice without these specific details.In some instances, well known method, structure and skill is not been shown in detail
Art, so as not to obscure the understanding of this description.
Similarly, it should be understood that disclose to simplify the present invention and help to understand one or more in each inventive aspect
A, in the above description of the exemplary embodiment of the present invention, each feature of the invention is grouped together into individually sometimes
In embodiment, figure or descriptions thereof.It is intended in reflection is following however, should not explain the method for the disclosure：Wanted
Ask protection the present invention claims the more features of feature than being expressly recited in each claim.More precisely, such as
As following claims reflect, inventive aspect is all features less than single embodiment disclosed above.
Therefore, it then follows thus claims of specific implementation mode are expressly incorporated in the specific implementation mode, wherein each right is wanted
Ask itself all as a separate embodiment of the present invention.
It will be understood by those skilled in the art that can adaptively be changed to the module in the equipment in embodiment
And they are provided in the different one or more equipment of the embodiment.Can in embodiment module or unit or
Component is combined into a module or unit or component, and can be divided into multiple submodule or subelement or subgroup in addition
Part.In addition to such feature and/or at least some of process or unit are mutually exclusive places, any combinations may be used
To all features disclosed in this specification (including adjoint claim, abstract and attached drawing) and such disclosed any side
All processes or unit of method or equipment are combined.Unless expressly stated otherwise, this specification (including want by adjoint right
Ask, make a summary and attached drawing) disclosed in each feature can be replaced by providing the alternative features of identical, equivalent or similar purpose.
In addition, it will be appreciated by those of skill in the art that although some embodiments described herein include other embodiments
In included certain features rather than other feature, but the combination of the feature of different embodiments means in of the invention
Within the scope of and form different embodiments.For example, in the following claims, embodiment claimed is appointed
One of meaning mode can use in any combination.
The all parts embodiment of the present invention can be with hardware realization, or to run on one or more processors
Software module realize, or realized with combination thereof.It will be understood by those of skill in the art that can use in practice
In the equipment of microprocessor or digital signal processor (DSP) to realize a kind of browser terminal according to the ... of the embodiment of the present invention
Some or all components some or all functions.The present invention is also implemented as executing side as described herein
Some or all equipment or program of device (for example, computer program and computer program product) of method.It is such
Realize that the program of the present invention can may be stored on the computer-readable medium, or can be with the shape of one or more signal
Formula.Such signal can be downloaded from internet website and be obtained, and either be provided on carrier signal or with any other shape
Formula provides.
It should be noted that the present invention will be described rather than limits the invention for above-described embodiment, and ability
Field technique personnel can design alternative embodiment without departing from the scope of the appended claims.In the claims,
Any reference mark between bracket should not be configured to limitations on claims.Word "comprising" does not exclude the presence of not
Element or step listed in the claims.Word "a" or "an" before element does not exclude the presence of multiple such
Element.The present invention can be by means of including the hardware of several different elements and being come by means of properly programmed computer real
It is existing.In the unit claims listing several devices, several in these devices can be by the same hardware branch
To embody.The use of word first, second, and third does not indicate that any sequence.These words can be explained and be run after fame
Claim.
Finally it should be noted that：The above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations；To the greatest extent
Present invention has been described in detail with reference to the aforementioned embodiments for pipe, it will be understood by those of ordinary skill in the art that：Its according to
So can with technical scheme described in the above embodiments is modified, either to which part or all technical features into
Row equivalent replacement；And these modifications or replacements, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme should all cover in the claim of the present invention and the range of specification.
Claims (6)
1. the medical information of species diversity enhancing recommends method, which is characterized in that including：
Obtain first scoring of the first user in the first user set to the pre-selection information in pre-selection information set, the pre-selection
Information includes medical information；
The target user and institute are calculated to the second scoring of the pre-selection information according to first scoring and target user
State the similarity between the first user；
According to the diversity metric between target user and first user described in the calculated similarity calculation；
Select first user of predetermined number as the target user's according to the calculated diversity metric
Arest neighbors；
The target user is calculated according to the arest neighbors of the target user to the scoring of the pre-selection information to provide the pre-selection
The scoring of news；
Target information is determined according to the scoring of the calculated pre-selection information, and the target information is recommended into the target and is used
Family；
The diversity according between target user and first user described in the calculated similarity calculation is measured
Value, including：
Establish first rating matrix of first user for the pre-selection information in the pre-selection information set；
First rating matrix is expanded into the second rating matrix, user's classification shape parameter based on user's classification shape parameter
For the species parameter of first user；
First user in being gathered first user using second rating matrix according to clustering algorithm is divided into more
A class, obtains clustering cluster；
Calculate it is all kinds of in the target user and the clustering cluster in user diversity metric；
It is described calculate the target user with it is all kinds of in the clustering cluster in user diversity metric, including：
Diversity metric between class in the target user and the clustering cluster is calculated using following formula (1)：
Wherein, it is the class V in clustering cluster V that v, which belongs to clustering cluster V, k,kIn number of users, Diversion (u, V) be target user u
Diversity metric between clustering cluster V, sim (u, v) similarities between user u and user v；
The diversity metric of the target user and the user in the clustering cluster are calculated according to following formula (2)：
Wherein, diversity metrics of the Diversion (u, v) between target user u and user v, λ are adjustable parameter, when λ becomes
To in 1 when, the diversity metric highest between target user u and user v, when λ is not intended to 1, target user u with
Similitude highest between the v of family.
2. according to the method described in claim 1, it is characterized in that, the arest neighbors according to the target user provides pre-selection
The scoring of news calculates scoring of the target user to the pre-selection information, including：
Scoring of the target user to the pre-selection information is calculated using following formula (3)：
Wherein, rv,iScoring for user v to information i,Scoring for target user u to information i, v belong to the target user
Arest neighbors n (u) in user.
3. according to the method for claims 1 or 2, which is characterized in that it is described according to it is described first scoring and target user to institute
The similarity between the second scoring calculating target user of pre-selection information and first user is stated, including：
The similarity between the target user and first user is calculated using following formula (4)：
Wherein, sim (u, v) represents similarity between user u and user v, Ru,tScoring for user u to information t, Rv,tFor user
Scorings of the v for information t, Iu,vFor the set of the common information of user v and user u, AuIt is user u to the common information
The average value of scoring, AvIt is the average value that user v scores to the common information.
4. the medical information recommendation apparatus of species diversity enhancing, which is characterized in that including：
Acquisition module, for obtaining the first user in the first user set to first of the pre-selection information in pre-selection information set
Scoring, the pre-selection information includes medical information；
First computing module, for being calculated the second scoring of the pre-selection information according to first scoring and target user
Similarity between the target user and first user；
Second computing module, for according between target user and first user described in the calculated similarity calculation
Diversity metric；
Selecting module, for selecting first user of predetermined number as institute according to the calculated diversity metric
State the arest neighbors of target user；
Third computing module, for calculating the target to the scoring of the pre-selection information according to the arest neighbors of the target user
Scoring of the user to the pre-selection information；
Recommending module pushes away the target information for determining target information according to the scoring of the calculated pre-selection information
It recommends to the target user；
Second computing module, including：
Unit is established, is commented for first of the pre-selection information in the pre-selection information set for establishing first user
Sub-matrix；
First rating matrix is expanded into the second rating matrix by expansion unit for being based on user's classification shape parameter, described
User's classification shape parameter is the species parameter of first user；
Taxon, for described the in being gathered first user using second rating matrix according to clustering algorithm
One user is divided into multiple classes, obtains clustering cluster；
Computing unit, for calculate in the target user and the clustering cluster it is all kinds of in user diversity metric；
The computing unit is used for：
Diversity metric between class in the target user and the clustering cluster is calculated using following formula (1)：
Wherein, it is the class V in clustering cluster V that v, which belongs to clustering cluster V, k,kIn number of users, Diversion (u, V) be target user u
Diversity metric between clustering cluster V, sim (u, v) similarities between user u and user v；
The diversity metric of the target user and the user in the clustering cluster are calculated according to following formula (2)：
Wherein, diversity metrics of the Diversion (u, v) between target user u and user v, λ are adjustable parameter, when λ becomes
To in 1 when, the diversity metric highest between target user u and user v, when λ is not intended to 1, target user u with
Similitude highest between the v of family.
5. device according to claim 4, which is characterized in that the third computing module is used for：
Scoring of the target user to the pre-selection information is calculated using following formula (3)：
Wherein, rv,iScoring for user v to information i,Scoring for target user u to information i, v belong to the target user
Arest neighbors n (u) in user.
6. device according to claim 4 or 5, which is characterized in that first computing module is used for：
The similarity between the target user and first user is calculated using following formula (4)：
Wherein, sim (u, v) represents similarity between user u and user v, Ru,tScoring for user u to information t, Rv,tFor user
Scorings of the v for information t, Iu,vFor the set of the common information of user v and user u, AuIt is user u to the common information
The average value of scoring, AvIt is the average value that user v scores to the common information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710954748.9A CN107577823B (en) | 2017-10-13 | 2017-10-13 | The medical information of diversity enhancing recommends method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710954748.9A CN107577823B (en) | 2017-10-13 | 2017-10-13 | The medical information of diversity enhancing recommends method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107577823A CN107577823A (en) | 2018-01-12 |
| CN107577823B true CN107577823B (en) | 2018-08-28 |
Family
ID=61037127
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710954748.9A Active CN107577823B (en) | 2017-10-13 | 2017-10-13 | The medical information of diversity enhancing recommends method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107577823B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116504354B (en) * | 2023-06-28 | 2024-01-09 | 合肥工业大学 | Intelligent service recommendation method and system based on intelligent medical treatment |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102156747A (en) * | 2011-04-21 | 2011-08-17 | 清华大学 | Method and device for forecasting collaborative filtering mark by introduction of social tag |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7512582B2 (en) * | 2003-12-10 | 2009-03-31 | Microsoft Corporation | Uncertainty reduction in collaborative bootstrapping |
-
2017
- 2017-10-13 CN CN201710954748.9A patent/CN107577823B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102156747A (en) * | 2011-04-21 | 2011-08-17 | 清华大学 | Method and device for forecasting collaborative filtering mark by introduction of social tag |
Non-Patent Citations (3)
| Title |
|---|
| Antonia Kyriakopoulou.Using Clustering and Co-Training to Boost Classification Performance.《19th IEEE International Conference on Tools with Artificial Intelligence》.2007,第325页-330页. * |
| 基于用户量化属性的多维相似度的协同过滤推荐算法;胡健 等;《江西理工大学学报》;20170630;第38卷(第3期);第86页-91页 * |
| 基于聚类的个性化推荐算法研究;雷震;《中国优秀硕士论文科技辑》;20160603;第28页-46页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107577823A (en) | 2018-01-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104123332B (en) | The display methods and device of search result | |
| CN105701216B (en) | A kind of information-pushing method and device | |
| CN110458649A (en) | Information recommendation method, device, electronic equipment and readable storage medium storing program for executing | |
| CN109118336A (en) | Information recommendation method, device, computer equipment and storage medium | |
| CN104899246B (en) | Collaborative filtering recommending method based on blurring mechanism user scoring neighborhood information | |
| CN107341176B (en) | Sample weight setting method and device and electronic equipment | |
| CN106131601A (en) | Video recommendation method and device | |
| CN108648058B (en) | Product sorting method and device, electronic equipment and storage medium | |
| CN102902717A (en) | Method, system and device for organizing a plurality of applications in app store | |
| CN110008397B (en) | Recommendation model training method and device | |
| CN104615631B (en) | A kind of method and device of information recommendation | |
| CN108595493A (en) | Method for pushing and device, storage medium, the electronic device of media content | |
| CN110473040A (en) | A kind of Products Show method and device, electronic equipment | |
| CN107274242A (en) | A kind of Method of Commodity Recommendation based on association analysis algorithm | |
| CN106168980A (en) | Multimedia resource recommends sort method and device | |
| CN106776930A (en) | A kind of location recommendation method for incorporating time and geographical location information | |
| CN110020152B (en) | Application recommendation method and device | |
| CN108304399A (en) | The recommendation method and device of Web content | |
| CN105824961B (en) | A kind of label determines method and device | |
| WO2015134489A1 (en) | Utilizing product and service reviews | |
| CN110706783B (en) | Recipe recommendation method, apparatus, computing device and computer storage medium | |
| CN111737473B (en) | Text classification method, device and equipment | |
| KR101567684B1 (en) | Method for selecting recommendation type in product recommeding system | |
| CN103942264B (en) | The method and apparatus for pushing the webpage comprising news information | |
| CN106656741A (en) | Information push method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |