CN104796478B - A kind of resource recommendation method and device - Google Patents
A kind of resource recommendation method and device Download PDFInfo
- Publication number
- CN104796478B CN104796478B CN201510201816.5A CN201510201816A CN104796478B CN 104796478 B CN104796478 B CN 104796478B CN 201510201816 A CN201510201816 A CN 201510201816A CN 104796478 B CN104796478 B CN 104796478B
- Authority
- CN
- China
- Prior art keywords
- resource
- user
- keyword
- resources
- confidence level
- 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
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000011156 evaluation Methods 0.000 claims description 76
- 238000003780 insertion Methods 0.000 claims description 18
- 230000037431 insertion Effects 0.000 claims description 18
- 230000001186 cumulative effect Effects 0.000 claims description 15
- 230000006399 behavior Effects 0.000 description 13
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000002349 favourable effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000000155 isotopic effect Effects 0.000 description 2
- BQCADISMDOOEFD-UHFFFAOYSA-N Silver Chemical compound [Ag] BQCADISMDOOEFD-UHFFFAOYSA-N 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000029058 respiratory gaseous exchange Effects 0.000 description 1
- 229910052709 silver Inorganic materials 0.000 description 1
- 239000004332 silver Substances 0.000 description 1
Abstract
The embodiment of the present invention provides a kind of resource recommendation method and device, is related to network technique field, is able to solve the low problem of the accuracy of the recommendation of the user resources as caused by the insincere behavior of user.The specific scheme is that when receiving resource request, obtain the matching degree of the keyword carried in all corresponding confidence levels of user and the corresponding keyword of all resources and resource request in resource sharing system, the corresponding confidence level of owning user is distinguished according to all resources and matching degree obtains the corresponding recommendation of all resources, recommendation is ranked up according to the size of recommendation, and v recommendation before being chosen after sequence according to sequence from big to small, and show the corresponding resource of preceding v recommendation.The embodiment of the present invention is for carrying out resource recommendation.
Description
Technical field
The present invention relates to network technique field more particularly to a kind of resource recommendation methods and device.
Background technique
In resource sharing system, user provides resource by network, while can also obtain other users by network
The resource of offer.After user logs in, resource request can be sent to resource sharing system according to the resource requirement of oneself, and providing
Keyword is carried in the request of source, to meet the resource of user demand by resource sharing system search.Resource sharing system is usual
It is ranked up and according to the matching degree of the keyword carried in the keyword and resource request of resource in resources bank with the shape of list
Formula is sequentially shown, so that the resource of recommendation is returned to user, so that user selects.
In the prior art, certain customers in order to promoted oneself provided resource download and shared rate, often will be some
The popular keyword unrelated with provided resource is appointed as the keyword of the resource to cheat other users, and certain customers' can not
Letter behavior mismatches resource sharing system according to the resource of keyword recommendation and the resource request of user, to reduce money
The accuracy that source is recommended.For example, a certain user A has shared a film " a " in resource sharing system, in order to promote downloading
Instantly the keyword " b " of silver screen hits " b " is appointed as the keyword of the film by rate, the user A.When other users want viewing
The film " b " and according to keyword " b " carry out resource searching when, resource sharing system is according to keyword " b " preferential recommendation user A
There is provided the film assets " a " is the resource high with the keyword match degree that carries in the resource request of user, however the resource is real
It is to have no associated film " a " with " b " on border, does not meet the resource requirement of other users, reduce the accuracy of resource recommendation.
Summary of the invention
The embodiment of the present invention provides a kind of resource recommendation method and device, is able to solve since the insincere behavior of user is led
The low problem of the accuracy that the user resources of cause are recommended.
In order to achieve the above objectives, the embodiment of the present invention adopts the following technical scheme that
In a first aspect, providing a kind of resource recommendation method, comprising:
When receiving resource request, the corresponding confidence level of all users in resource sharing system is obtained, and obtain
The matching degree of the keyword carried in all corresponding keywords of resource and the resource request；
According to all resource difference corresponding confidence levels of owning user and the corresponding key of all resources
The matching degree of the keyword carried in word and the resource request obtains the corresponding recommendation of all resources；
The recommendation is ranked up according to the size of the recommendation, and according to sequence from big to small after sequence
The v recommendation before choosing, and the corresponding resource of the v recommendations before showing.
With reference to first aspect, in the first mode in the cards of first aspect, for appointing in all users
One user, obtaining the corresponding confidence level of any user described in resource sharing system includes:
Determine whether the resource quantity in all users in the resource collection of any user is 0；
If not 0, then obtain the confidence level of any resource in the resource collection of any user in all users；
Then, obtain each resource in the resource collection confidence level it is cumulative after value and the resource collection in resource
The ratio of quantity, the ratio are the corresponding confidence level of any user in all users；
Alternatively, then the corresponding confidence level of any user is the first preset value in all users if 0；
Wherein, the corresponding confidence level of any user indicates in all users are as follows:
T (u) indicates the corresponding confidence level of any user in all users, and T (u, r) indicates to appoint in the resource collection
The confidence level of one resource, RuIndicate the resource collection, | Ru| indicate the quantity of resource in the resource collection, q indicates described the
One preset value.
With reference to first aspect the first mode in the cards, in second of mode in the cards of first aspect
In, the confidence level for obtaining any resource in the resource collection of any user in all users includes:
According to the product of the quantity for the user for providing evaluation to any resource in the resource collection and the first preset value, with
By the product of value and second preset value of the evaluation of estimate of each user after cumulative and, then be added with first preset value
Value afterwards obtains the confidence level of any resource in the resource collection, wherein the evaluation of estimate of the user is corresponding for the user
Confidence level and the user product of evaluation indicated value that any resource in the resource collection is provided, with the user couple
The ratio of the quantity for the evaluation that all resources provide；
Wherein, the confidence level of any resource indicates in the resource collection of any user in all users are as follows:
T (u, r)=q+q*s+ (1-q) ∑J=1,2 ..., sT(uj)*kj/tj；
T (u, r) indicates the confidence level of any resource in the resource collection, and q indicates first preset value, s expression pair
Any resource provides the quantity of the user of evaluation in the resource collection, and (1-q) indicates second preset value, T (uj) indicate
The corresponding confidence level of j-th of user, k in s user of evaluation are provided to any resource in the resource collectionjIndicate described
The evaluation indicated value that j user provides any resource in the resource collection, tjIndicate j-th of user to described all
The quantity for the evaluation that resource provides.
With reference to first aspect, in the third mode in the cards of first aspect, for appointing in all resources
One resource obtains of the keyword carried in the keyword and the resource request of any resource described in resource sharing system
Include: with degree
By the corresponding keyword of resource any in all resources after Hash operation in Bloom Filter into
Row insertion operation obtains the mark of any resource in all resources, and the keyword carried in the resource request is passed through
Insertion operation is carried out in the Bloom Filter after crossing the Hash operation, to obtain the mark of the resource request；
Obtain the Hamming distance of the mark of any resource in all resources and the mark of the resource request from；
Obtain ratio of the Hamming distance from the length n with the Bloom Filter, it is any in all resources
The matching degree of the keyword carried in the keyword of resource and the resource request is the difference of 1 with the ratio；
Wherein, the matching of the keyword carried in the keyword of any resource and the resource request in all resources
Degree indicates are as follows:
S indicates the matching of the keyword carried in the keyword of any resource and the resource request in all resources
Degree, L indicate the Hamming distance from n indicates the length of the Bloom Filter.
With reference to first aspect to the third mode in the cards of first aspect, the 4th kind in first aspect may be real
In existing mode, the corresponding recommendation of any resource is indicated in all resources are as follows:
Mi=T (up)*Si；
Wherein, MiIndicate the corresponding recommendation of any resource i, T (u in all resourcesp) indicate in all resources
The confidence level of any resource i owning user p, SiIndicate that the keyword of any resource i is asked with the resource in all resources
Seek the matching degree of the keyword of middle carrying.
Second aspect provides a kind of control device, comprising:
Acquiring unit, it is corresponding for when receiving resource request, obtaining all users in resource sharing system
Confidence level, and obtain the matching degree of the keyword carried in the corresponding keyword of all resources and the resource request；
According to all resource difference corresponding confidence levels of owning user and the corresponding key of all resources
The matching degree of the keyword carried in word and the resource request obtains the corresponding recommendation of all resources；
Control unit, for being ranked up according to the size of the recommendation to the recommendation, and after sequence according to
The v recommendations before sequence from big to small is chosen, and the corresponding resource of the v recommendations before showing.
In conjunction with second aspect, in the first mode in the cards of second aspect, the acquiring unit is specifically used for:
Determine whether the resource quantity in all users in the resource collection of any user is 0；
If not 0, then obtain the confidence level of any resource in the resource collection of any user in all users；
Then, obtain each resource in the resource collection confidence level it is cumulative after value and the resource collection in resource
The ratio of quantity, the ratio are the corresponding confidence level of any user in all users；
Alternatively, then the corresponding confidence level of any user is the first preset value in all users if 0；
Wherein, the corresponding confidence level of any user indicates in all users are as follows:
T (u) indicates the corresponding confidence level of any user in all users, and T (u, r) indicates to appoint in the resource collection
The confidence level of one resource, RuIndicate the resource collection, | Ru| indicate the quantity of resource in the resource collection, q indicates described the
One preset value.
In conjunction with the first mode in the cards of second aspect, in second of mode in the cards of second aspect
In, the acquiring unit is specifically used for:
According to the product of the quantity for the user for providing evaluation to any resource in the resource collection and the first preset value, with
By the product of value and second preset value of the evaluation of estimate of each user after cumulative and, then be added with first preset value
Value afterwards obtains the confidence level of any resource in the resource collection, wherein the evaluation of estimate of the user is corresponding for the user
Confidence level and the user product of evaluation indicated value that any resource in the resource collection is provided, with the user couple
The ratio of the quantity for the evaluation that all resources provide；
Wherein, the confidence level of any resource indicates in the resource collection of any user in all users are as follows:
T (u, r)=q+q*s+ (1-q) ∑J=1,2 ..., sT(uj)*kj/tj；
T (u, r) indicates the confidence level of any resource in the resource collection, and q indicates first preset value, s expression pair
Any resource provides the quantity of the user of evaluation in the resource collection, and (1-q) indicates second preset value, T (uj) indicate
The corresponding confidence level of j-th of user, k in s user of evaluation are provided to any resource in the resource collectionjIndicate described
The evaluation indicated value that j user provides any resource in the resource collection, tjIndicate j-th of user to described all
The quantity for the evaluation that resource provides.
In conjunction with second aspect, in the third mode in the cards of second aspect, the acquiring unit is specifically used for:
By the corresponding keyword of resource any in all resources after Hash operation in Bloom Filter into
Row insertion operation obtains the mark of any resource in all resources, and the keyword carried in the resource request is passed through
Insertion operation is carried out in the Bloom Filter after crossing the Hash operation, to obtain the mark of the resource request；
Obtain the Hamming distance of the mark of any resource in all resources and the mark of the resource request from；
Obtain ratio of the Hamming distance from the length n with the Bloom Filter, it is any in all resources
The matching degree of the keyword carried in the keyword of resource and the resource request is the difference of 1 with the ratio；
Wherein, the matching of the keyword carried in the keyword of any resource and the resource request in all resources
Degree indicates are as follows:
S indicates the matching of the keyword carried in the keyword of any resource and the resource request in all resources
Degree, L indicate the Hamming distance from n indicates the length of the Bloom Filter.
In conjunction with the third mode in the cards of second aspect to second aspect, the 4th kind in second aspect may be real
In existing mode, the corresponding recommendation of any resource is indicated in all resources are as follows:
Mi=T (up)*Si；
Wherein, MiIndicate the corresponding recommendation of any resource i, T (u in all resourcesp) indicate in all resources
The confidence level of any resource i owning user p, SiIndicate that the keyword of any resource i is asked with the resource in all resources
Seek the matching degree of the keyword of middle carrying.
The embodiment of the present invention provides a kind of resource recommendation method and device, and the control device in resource sharing system receives
When to resource request, the corresponding confidence level of user in resource sharing system is obtained, which can to characterize the user
With the degree trusted by other users, and according in the corresponding confidence level of resource owning user and the resource sharing system of acquisition
The matching degree of keyword in the keyword and resource request of resource calculates recommendation, thus big for user's preferential recommendation recommendation
Resource, i.e., owning user is corresponding with a high credibility and the matching degree height of the keyword in its keyword and resource request is to more
Meet the resource of user resources demand.Due to resource sharing system can preferential recommendation more can by other users trust user mention
The resource of confession, and the User reliability with insincere behavior is low, it is difficult to trusted by other users, the resource which provides
Also tend to not meet the resource requirement of other users to will not be by preferential recommendation, it is thus possible to solve in the prior art due to
The low problem of the accuracy of resource recommendation caused by the insincere behavior at family.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, 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
It obtains other drawings based on these drawings.
Fig. 1 is a kind of resource recommendation method schematic diagram provided in an embodiment of the present invention；
Fig. 2 is a kind of resource recommendation method schematic diagram provided in an embodiment of the present invention；
Fig. 3 is a kind of process schematic for the confidence level for obtaining any resource provided in an embodiment of the present invention；
Fig. 4 is a kind of process schematic for obtaining the corresponding confidence level of any user provided in an embodiment of the present invention；
Fig. 5 is a kind of structural schematic diagram of control device provided in an embodiment of the present invention.
Specific embodiment
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 description, 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.
Referring to Fig. 1, the embodiment of the present invention provides a kind of resource recommendation method, and key step may include:
101, for control device when receiving resource request, all users are corresponding in acquisition resource sharing system can
Reliability, and obtain the matching degree of the keyword carried in the corresponding keyword of all resources and resource request.
Wherein, control device here can be to be responsible for calculating in resource sharing system and set with the computer of resource management
Degree standby, that the corresponding confidence level of user can be trusted to characterize user by other users, matching degree are total to characterize resource
Enjoy the matching degree of the keyword carried in the corresponding keyword of all resources in system and resource request.
102, control device distinguishes the corresponding confidence level of owning user according to all resources and all resources are corresponding
The matching degree of the keyword carried in keyword and resource request, obtains the corresponding recommendation of all resources.
103, control device is ranked up recommendation according to the size of recommendation, and according to from big to small after sequence
V recommendation before sequence is chosen, and show the corresponding resource of preceding v recommendation.
Wherein, the corresponding resource of preceding v recommendation is the resource that resource sharing system is recommended, and certain v's is specific
Numerical value, which can according to need, to be set, here without limitation.Wherein, it is particularly shown here and can be recommendation resource to user
Resource name, the mark (such as resource number, the storage location etc. of resource) which can identify with system
It is associated, recommend the corresponding resource name of resource that can navigate to the page of recommendation resource to divide in order to which user passes through selection
Enjoy the recommendation resource.
Optionally, for any user in all users, it is corresponding that control device obtains any user in resource sharing system
Confidence level may include:
Control device determines whether the resource quantity in all users in the resource collection of any user is 0；
If not 0, then control device obtains the confidence level of any resource in the resource collection of any user in all users；
Then, resource in the value and resource collection after the confidence level of each resource is cumulative in control device acquisition resource collection
The ratio of quantity, ratio are the corresponding confidence level of any user in all users；
Alternatively, then the corresponding confidence level of any user is the first preset value in all users of control device if 0；
Wherein, the corresponding confidence level of any user indicates in all users are as follows:
T (u) indicates the corresponding confidence level of any user in all users, and T (u, r) indicates any resource in resource collection
Confidence level, RuIndicate resource collection, | Ru| indicate the quantity of resource in resource collection, q indicates the first preset value.
Optionally, the confidence level that control device obtains any resource in the resource collection of any user in all users can be with
Include:
Control device is according to the quantity of user and multiplying for the first preset value for providing evaluation to resource any in resource collection
Product, with the product of value and the second preset value by the evaluation of estimate of each user after cumulative and, then after being added with the first preset value
Value, obtain resource collection in any resource confidence level, wherein the evaluation of estimate of user be the corresponding confidence level of user and user
To the product for the evaluation indicated value that resource any in resource collection provides, the quantity for the evaluation that all resources are provided with user
Ratio；
Wherein, the confidence level of any resource indicates in the resource collection of any user in all users are as follows:
T (u, r)=q+q*s+ (1-q) ∑J=1,2 ..., sT(uj)*kj/tj；
T (u, r) indicates the confidence level of any resource in resource collection, and q indicates that the first preset value, s are indicated to resource collection
In any resource provide evaluation user quantity, (1-q) indicate the second preset value, T (uj) indicate to any in resource collection
Resource provides the corresponding confidence level of j-th of user, k in s user of evaluationjIndicate j-th of user to any in resource collection
The evaluation indicated value that resource provides, tjIndicate the quantity for the evaluation that j-th of user provides all resources.
Optionally, for any resource in all resources, control device obtains the pass of any resource in resource sharing system
The matching degree of the keyword carried in key word and resource request may include:
Control device is by the corresponding keyword of resource any in all resources in Bloom Filter after Hash operation
Middle carry out insertion operation obtains the mark of any resource in all resources, and by the keyword carried in resource request by breathing out
Insertion operation is carried out in Bloom Filter after uncommon operation, to obtain the mark of resource request；
Control device obtain the Hamming distance of the mark of the mark and resource request of any resource in all resources from；
Control device obtains ratio of the Hamming distance from the length n with Bloom Filter, any resource in all resources
Keyword and resource request in the matching degree of keyword that carries be 1 with ratio difference；
Wherein, the matching degree of the keyword carried in the keyword of any resource and resource request in all resources indicates
Are as follows:
S indicates the matching degree of the keyword carried in the keyword of any resource and resource request in all resources, and L is indicated
Hamming distance is from n indicates the length of Bloom Filter.
Optionally, the corresponding recommendation of any resource can indicate in all resources are as follows:
Mi=T (up)*Si；
Wherein, MiIndicate the corresponding recommendation of any resource i, T (u in all resourcesp) indicate any resource in all resources
The confidence level of i owning user p, SiIndicate the keyword carried in the keyword of any resource i and resource request in all resources
Matching degree.
The embodiment of the present invention provides a kind of resource recommendation method, and the control device in resource sharing system receives resource
When request, the corresponding confidence level of user in resource sharing system is obtained, which can be by it to characterize the user
The degree of his users to trust, and according to resource in the corresponding confidence level of resource owning user and the resource sharing system of acquisition
The matching degree of keyword in keyword and resource request calculates recommendation, thus the money big for user's preferential recommendation recommendation
Source, i.e. owning user be corresponding with a high credibility and the matching degree height of its keyword and the keyword in resource request is to more meet
The resource of user resources demand.Due to resource sharing system can preferential recommendation more can by other users trust user provide
Resource, and the User reliability with insincere behavior is low, it is difficult to trusted by other users, the resource which provides is also past
Toward the resource requirement for not meeting other users thus will not be by preferential recommendation, it is thus possible to solve in the prior art due to user's
The low problem of the accuracy of resource recommendation caused by insincere behavior.
Referring to fig. 2, the embodiment of the present invention provides a kind of resource recommendation method, mainly may comprise steps of:
201, for any user in all users, control device determines in all users when receiving resource request
Whether the resource quantity in the resource collection of any user is 0, then enters step 202 or step 204.
Here control device, which can be, to be responsible for calculating the computer with resource management, any resource in resource sharing system
The resource collection of owning user refers to the set for all resources that any resource owning user provides.User is needing to carry out resource
When search, resource request can be sent to control device according to the resource requirement of itself, and keyword is carried in resource request,
To search for the resource for meeting own resource demand.
Control device, can be in the resource collection by any user in all users of determination when receiving resource request
Resource quantity whether be 0, it is any in all users so as to be obtained when resource quantity is not 0 according to step 202,203
The corresponding confidence level of user, or obtaining according to step 204 when resource quantity is 0 in all users that any user is corresponding can
Reliability.Wherein, the corresponding confidence level of any user can characterize the journey that the user can be trusted by other users in all users
Degree.
If not 202,0, then control device is given according to any resource in the resource collection to any user in all users
The product of the quantity of the user evaluated out and the first preset value is preset with by value of the evaluation of estimate of each user after cumulative with second
The sum of the product of value, then with the first preset value value after being added, obtain resource collection in any resource confidence level, wherein use
The evaluation of estimate at family is the product for the evaluation indicated value that the corresponding confidence level of user and user provide resource any in resource collection,
The ratio of the quantity of the evaluation provided with user to all resources, then enters step 203.
Wherein, the confidence level of any resource can characterize any resource in the resource collection of any user in all users
The degree that can be trusted by other users.The higher resource of quality can more be trusted by other users, wherein any resource
Resource quality may include the accuracy of Resource Key description, the influence power of bibliography class resource, the resolution of video class resource
The many aspects such as rate.Illustratively, if the keyword of any resource is consistent with the particular content of resource, any resource is more
It can be trusted by other users.If control device determines in all users of resource sharing system in the resource collection of any user
Resource quantity is not 0, then the credible of any resource in the resource collection of any user can be obtained in all users according to formula 1
Degree.Wherein, formula 1 can indicate are as follows:
T (u, r)=q+q*s+ (1-q) ∑J=1,2 ..., sT(uj)*kj/tj；
In formula 1, T (u, r) indicates the confidence level of any resource in resource collection, and q indicates the first preset value, s expression pair
Any resource provides the quantity of the user of evaluation in resource collection, and (1-q) indicates the second preset value, T (uj) indicate to resource set
Any resource provides the corresponding confidence level of j-th of user, k in s user of evaluation in conjunctionjIndicate j-th of user to resource set
The evaluation indicated value that any resource provides in conjunction, tjIndicate the quantity for the evaluation that j-th of user provides all resources.Wherein,
The process schematic for obtaining the confidence level of any resource may refer to Fig. 3.
According to the formula 1 for obtaining the confidence level of any resource in the resource collection of any user in all users, at least may be used
To know the following aspects:
1, the comment that the higher user of confidence level provides promotes effect to the confidence level of any resource and becomes apparent from, and is presented as
T (u in formula 1j)；
2, the more user of the quantity of the evaluation provided to all resources evaluates the contribution for other users confidence level
Degree will reduce, the T (u that can be presented as in formula 1j)*kj/tj；
3, the more user of the evaluation quantity obtained, the available a degree of promotion of trust value.It is presented as formula 1
In q*s；
4, any resource that do not evaluated has default confidence level, the q being presented as in formula 1.
From the foregoing, it will be observed that obtain in all users any resource in the resource collection of any user in the embodiment of the present invention can
The corresponding confidence level of user that evaluation is provided to any resource is introduced in the formula 1 of reliability, thus inventive embodiments consider
The different user that confidence level has differences generated Different Effects when evaluating other users, calculate any resource can
Formula 1 and its algorithm used by reliability are more reasonable, so that the confidence level for calculating any resource obtained is more credible.
In addition, the q in formula 1 indicates the first weight, that is, any resource default confidence level, specific value can basis
It is set, here without limitation.K in formula 1jIndicate what j-th of user provided resource any in resource collection
Evaluate indicated value, kjCan be different according to the difference for the evaluation type that user j is provided, other users can be according to any money
The height of the quality in source provides different types of evaluation.For example, the evaluation type that j-th of user provides can for favorable comment or
Difference is commented, corresponding kjIt respectively can be 1, -1, illustratively, when user thinks the keyword and particular content phase of any resource
Symbol or other reasons can provide favorable comment when making the quality of any resource lower, when any resource owning user not
Credible behavior (such as current popular keyword is used as to the keyword of provided resource, and the content of provided resource and the heat
Door keyword is unrelated) such that the keyword of any resource is not inconsistent with particular content or other reasons make any resource
When quality is lower, user can be commented any resource to going on business.Alternatively, the evaluation type that j-th of user provides may be
Favorable comment, in comment or difference is commented, corresponding kjIt respectively can be 1,0.5, -1, the embodiment of the present invention is for specifically evaluating type and correspondence
Evaluation indicated value size without limitation.
203, resource in the value and resource collection after the confidence level that control device obtains each resource in resource collection adds up
The ratio of quantity, ratio are the corresponding confidence level of any user in all users, then enter step 208.
It is got in the resource collection of any user after the confidence level of any resource in step 202, it can be according to public affairs
Resource quantity in value and resource collection after the confidence level of each resource is cumulative in the resource collection of the acquisition of formula 2 any user
Ratio, the ratio are the corresponding confidence level of any user in all users.Wherein, formula 2 can indicate are as follows:
In formula 2, T (u) indicates the corresponding confidence level of any user in all users, and T (u, r) indicates to appoint in resource collection
The confidence level of one resource, RuIndicate resource collection, | Ru| indicate resource collection in resource quantity and | Ru| the is not indicated for 0, q
One preset value.
By formula 2 it is found that making any resource when the quality of any resource in the resource collection of any user is higher
When confidence level is higher, the mean value of resource is higher in any user resource collection, i.e., each in the resource collection of any user
The ratio of resource quantity is larger in value and resource collection after the confidence level of resource is cumulative, and the corresponding confidence level of any user is got over
It is high.And the keyword for the resource that the insincere behavior for working as any user provides it is not inconsistent with resource content, or due to it
When its reason causes the quality of the resource lower, the resource that is provided in the resource collection of any user it is with a low credibility, thus
So that any user is corresponding with a low credibility.
204, if 0, then the corresponding confidence level of any user is the first preset value in all users, is then entered step
208。
If control device determines the resource quantity in all users of resource sharing system in the resource collection of any user
It is 0, then the corresponding confidence level of any user is the first preset value.In this step, any user is corresponding in all users
Confidence level can be expressed as formula 3:
T (u)=q, | Ru|=0, wherein T (u) indicates the corresponding confidence level of any user in all users, RuIndicate resource
Set, | Ru| indicate resource collection in resource quantity and | Ru| the first preset value is indicated for 0, q.
In equation 3, work as resource quantity | Ru| when being 0, the default confidence level of any user is the first weight q.That is, if
Any user does not provide any resource, then the user has default confidence level q.
In conjunction with formula 2 and formula 3, the corresponding confidence level of any user can be expressed as formula 4 in all users:
T (u) indicates the corresponding confidence level of any user in all users, and T (u, r) indicates any resource in resource collection
Confidence level, RuIndicate resource collection, | Ru| indicate the quantity of resource in resource collection, q indicates the first preset value.
In conjunction with step 201 to 204, the process schematic for obtaining the corresponding confidence level of any user may refer to Fig. 4.
In addition, control device when receiving resource request, can also obtain all resource difference in resource sharing system
The matching degree of the keyword carried in corresponding keyword and resource request, specifically may refer to retouching for following steps 205 to 207
It states.
205, for any resource in all resources, control device passes through the corresponding keyword of resource any in all resources
It crosses and carries out insertion operation after Hash operation in Bloom Filter, obtain the mark of any resource in all resources, and will money
The keyword carried in the request of source carries out insertion operation after Hash operation in Bloom Filter, to obtain resource request
Mark.
Wherein, the binary value that Hash operation obtains is the unique and extremely compact numerical value representation of one piece of data, is closed
Key word can be converted to binary Hash after Hash operation.By the two of the corresponding keyword of resource any in all resources
System cryptographic Hash carries out insertion operation in Bloom Filter, can obtain the mark of any resource in all resources.It will money
The binary Hash of the keyword carried in the request of source carries out insertion operation in Bloom Filter, can obtain resource and ask
The mark asked.
Illustratively, if the corresponding keyword of any resource is the set K={ k of c keyword in all resources1,
k2,…,ki,…,kc, then: 1 keyword carries out the result after insertion operation after Hash operation in Bloom Filter
F(k1) formula 5 can be expressed as:
F(k1)=V0∪h{k1}=< v0|d10,v1|d11,…,vn-1|d1(n-1)>；
I keyword carries out the result F (k after insertion operation after Hash operation in Bloom Filteri) can be with
It is expressed as formula 6:
F(ki)=V0∪h{k1}∪h{k2}∪…∪h{ki}=F (ki-1)∪h{ki}；
C keyword carries out the result F (k after insertion operation after Hash operation in Bloom Filterc) can be with
It is expressed as formula 7:
F(kc)=V0∪h{k1}∪h{ki}∪…∪h{kc}=F (kc-1)∪h{kc}
Thus, the mark id of any resource can be expressed as formula:
Id=F (kc)=V0∪h{k1}∪h{k2}∪…∪h{kc}
Wherein, V0The original state for indicating Bloom Filter is the Binary Zero vector that length is n, v0,…,vn-1It indicates
0th to (n-1)th binary number 0, h { k in the Bloom Filter of original stateiIndicate that keyword ki is transported by Hash
Cryptographic Hash after calculation is the binary vector that length is n, < di0,di1,…,di(n-1), > indicate keyword kiCryptographic Hash h { ki}
In the 0th to (n-1)th binary numeral, " ∪ " indicates insertion operation defined in Bloom Filter, and " | " indicates step-by-step
Or operation.
After above-mentioned c keyword carries out c insertion operation, the length of Bloom Filter is still n, at this point, the length
Binary vector for n is the mark for being any resource in all resources obtained according to the keyword of resource any in all resources
Know.
Wherein, according to the process of the mark of the resource request of the keyword acquisition carried in resource request and according to all moneys
In source the keyword of any resource obtain the mark of any resource process it is similar, which is not described herein again.
206, control device obtain the Hamming distance of the mark of the mark and resource request of any resource in all resources from.
After obtaining in all resources the mark of any resource and the mark of resource request respectively in step 205, control
Device can obtain respectively the Hamming distance of the mark of the mark and resource request of any resource in all resources from.Wherein, appoint
1 quantity that the mark of one resource and resource request mark obtain after step-by-step or operation be corresponding Hamming distance from,
That is the Hamming distance digit different from the numerical value that can be indicated in two marks in identical bits, thus Hamming distance is from smaller, two
Isotopic number is not fewer in a mark, and two marks are more similar, which more meets the resource requirement of user.
207, control device obtains ratio of the Hamming distance from the length n with Bloom Filter, any in all resources
The matching degree of the keyword carried in the keyword and resource request of resource is the difference of 1 with ratio.
Hamming distance is obtained in step 206 from rear, and the available Hamming distance of control device is from L and Bloom Filter
Length n ratio, the matching degree of the keyword carried in all resources in the keyword of any resource and resource request is 1
With the difference of the ratio, i.e. matching degree can be expressed as formula 8:
Wherein, S indicates the matching degree of the keyword carried in the keyword of any resource and resource request in all resources,
L indicates the Hamming distance of the mark of any resource and the mark of resource request from n indicates the length of Bloom Filter.
By formula 8 it is found that the Hamming distance of the mark of the mark and resource request of any resource is from smaller, i.e., two marks
In not isotopic number it is fewer, then the ratio of L and n is smaller,Value it is bigger, i.e., the keyword of any resource is asked with resource
Ask the matching degree of the keyword of middle carrying higher, which more meets the resource requirement of user.
It should be noted that carrying out bit arithmetic by Hash operation and Bloom Filter obtains appointing for n-dimensional vector form
After the mark of one resource and the mark of resource request, then obtain the keyword carried in the keyword and resource request of any resource
Matching degree, calculating magnitude be O (| R |) i.e. to | R | it is directly proportional, wherein | R | be the number of all resources in resource sharing system
Amount.And in the prior art, if the set of keywords of the w keyword carried in resource request is combined into { k1, k2 ..., kw }, resource is total
Enjoy any resource r in the resource collection R of all resources in systemiThe set of keywords of f keyword be combined into { ki1,ki2,…,
kf, calculate separately in all resources pass identical with the keyword carried in resource request in the set of keywords of any resource
The quantity of key word, i.e., | { { k1, k2 ..., kw } ∩ ki1,ki2,…,kif| value, namely any resource in all resources is obtained respectively
Keyword and resource request in the matching degree of keyword that carries, wherein the quantity of identical keyword is more, and matching degree is got over
It is high.Since the quantity of keyword in resource requests different in the prior art is different, the pass of different resource in resource sharing system
The quantity of key word is also different, i.e., the quantity of the keyword of resource is dynamic change in resource request and resource sharing system, because
And the prior art obtain the process of matching degree calculating magnitude be O (m* | R |) i.e. to m* | R | value it is directly proportional, it is therein | R | table
Show the resource quantity of all resources in resource sharing system, m indicates the quantity of the keyword of dynamic change.As it can be seen that with existing skill
It is compared in art, the embodiment of the present invention obtains the key carried in the keyword of any resource and resource request in all resources respectively
The calculation of the matching degree of word reduces calculating magnitude, optimizes the matching way of resource recommendation.
Wherein, in the embodiment of the present invention to the specific value of s, n, f, w without limitation.
208, control device distinguishes the corresponding confidence level of owning user according to all resources and all resources are corresponding
The matching degree of the keyword carried in keyword and resource request, obtains the corresponding recommendation of all resources.
It is obtained in all users after the corresponding confidence level of any user in step 203 or 204, control device can basis
Any resource in all resources obtained in the corresponding confidence level of any resource owning user and step 207 in all resources
The matching degree of keyword and the keyword of resource request obtains the corresponding recommendation of all resources.Wherein, in all resources
The corresponding recommendation of any resource can be expressed as formula 9:
Mi=T (up)*Si；
In formula 9, MiIndicate the corresponding recommendation of any resource i, T (u in all resourcesp) indicate to appoint in all resources
The confidence level of one resource i owning user p, SiIt indicates to carry in the keyword of any resource i and resource request in all resources
The matching degree of keyword.
By formula 9 it is found that if of the keyword carried in the keyword of any resource i and resource request in all resources
With degree SiBigger, i.e., any resource more meets the resource requirement of user in all resources, then any resource i is corresponding in all resources
Recommendation MiIt is bigger；Meanwhile in all resources any resource i owning user p confidence level it is higher, then it is any in all resources
The corresponding recommendation M of resource iiIt is bigger.
209, control device is ranked up recommendation according to the size of recommendation, and according to from big to small after sequence
V recommendation before sequence is chosen, and show the corresponding resource of preceding v recommendation.
Control device is after sorting to recommendation, v recommendation before choosing according to sequence from big to small, preceding v recommendation
Being worth corresponding resource is the resource that resource sharing system is recommended, and the resource of recommendation is shown to user, so as to user
It is selected.The specific value of certain v, which can according to need, to be set, here without limitation.Illustratively, be particularly shown to
The resource name of the recommendation resource that can be tabular form or other forms of user, the resource name can be known with system
Other mark (such as resource number, the storage location etc. of resource) is associated, in order to which user recommends resource corresponding by selection
Resource name can navigate to and recommend the page of resource to sharing the recommendation resource.
Since the corresponding confidence level of any resource owning user is higher, and of its keyword and the keyword of resource request
Higher with spending, then the corresponding recommendation of the resource is bigger, so that resource sharing system can be user's preferential recommendation
The resource that user with a high credibility provides, and in embodiments of the present invention, the higher resource of quality more can be by other users institute
Trust, i.e. the confidence level of resource is higher, and the corresponding confidence level of owning user is also higher, and the resource is easier preferentially to be pushed away
It recommends, that is to say, that the higher resource of quality is easier to be preferentially recommended, thus the confidence level of introducing resource and user are corresponding
The resource quality for recommending resource can be improved in confidence level；And the user with insincere behavior is corresponding with a low credibility, it is difficult to quilt
Other users are trusted, and the quality for the resource which provides also tends to poor, do not meet the resource requirement of other users, thus
It will not be by preferential recommendation.Meanwhile resource sharing system can be enabled to be its keyword of user's preferential recommendation and user resources
The keyword match degree height carried in request more meets the resource of user resources demand.
It should be noted that in embodiments of the present invention, obtaining the corresponding confidence level of all resource difference owning users
Process (process for wherein, obtaining the corresponding confidence level of any user is step 201-204), respectively corresponds with all resources are obtained
Keyword and resource request in the process of the matching degree of keyword that carries (wherein, obtain the corresponding keyword of any resource
Process with the matching degree of the keyword carried in resource request is step 205-207) without specific precedence relationship, the present invention
Embodiment is without limitation.
In addition, be in the embodiment of the present invention in obtaining resource sharing system all resources difference owning user is corresponding can
After the matching degree of the keyword carried in reliability and the corresponding keyword of all resources and resource request, all resources are obtained
Corresponding recommendation is to carry out resource recommendation.It is of course also possible to which it is corresponding to obtain user in resource sharing system first
Confidence level, then the corresponding confidence level of user is ranked up, then obtains the resource collection of preceding g user with a high credibility
In the matching degree of keyword that carries in the corresponding keyword of all resources and resource request, and then it is high to choose matching degree
Preceding v resource carries out resource recommendation；Alternatively, can also obtain first in the corresponding keyword of all resources and resource request
The matching degree of the keyword of carrying then obtains the high corresponding confidence level of preceding x resource difference owning user of matching degree, in turn
It chooses the corresponding preceding v resource with a high credibility of owning user and carries out resource recommendation, the embodiment of the present invention is without limitation.Wherein,
The value of g, x, v can according to need and set, here without limitation.
In addition, the corresponding recommendation of any resource i can also be according to the confidence level of any resource i and any resource i
The matching degree of the keyword carried in keyword and resource request is calculated, such as the corresponding recommendation of any resource i can be with
It is expressed as Mi=T (u, ri)*Si.Wherein, MiIndicate the corresponding recommendation of any resource i, T (u, r in all resourcesi) indicate any
The confidence level of resource i, T (u, ri) acquisition, S can be calculated according to the algorithm in step 201iIndicate any resource i in all resources
Keyword and resource request in the matching degree of keyword that carries, the present invention is without limitation.
The embodiment of the present invention provides a kind of resource recommendation method, and the control device in resource sharing system receives resource
When request, the corresponding confidence level of user in resource sharing system is obtained, which can be by it to characterize the user
The degree of his users to trust, and according to resource in the corresponding confidence level of resource owning user and the resource sharing system of acquisition
The matching degree of keyword in keyword and resource request calculates recommendation, thus the money big for user's preferential recommendation recommendation
Source, i.e. owning user be corresponding with a high credibility and the matching degree height of its keyword and the keyword in resource request is to more meet
The resource of user resources demand.Due to resource sharing system can preferential recommendation more can by other users trust user provide
Resource, and the User reliability with insincere behavior is low, it is difficult to trusted by other users, the resource which provides is also past
Toward the resource requirement for not meeting other users thus will not be by preferential recommendation, it is thus possible to solve in the prior art due to user's
The low problem of the accuracy of resource recommendation caused by insincere behavior.
Referring to Fig. 5, the embodiment of the present invention provides a kind of control device 300, which may include:
Acquiring unit 301 can be used for when receiving resource request, obtain all user's difference in resource sharing system
The matching degree of the keyword carried in corresponding confidence level and the corresponding keyword of all resources and resource request, according to institute
There are the corresponding confidence level of resource difference owning user and matching degree to obtain the corresponding recommendation of all resources.
Control unit 302 can be used for being ranked up all resources according to the size of recommendation, and after sequence according to
V recommendation before sequence from big to small is chosen, and show the corresponding resource of preceding v recommendation.
Wherein, control device 300, which can be, is responsible for calculating the computer equipment with resource management in resource sharing system.
Optionally, acquiring unit 301 can be specifically used for:
Determine whether the resource quantity in all users in the resource collection of any user is 0；
If not 0, then obtain the confidence level of any resource in the resource collection of any user in all users；
Then, the ratio of resource quantity in the value and resource collection after the confidence level of each resource is cumulative in acquisition resource collection
Value, ratio are the corresponding confidence level of any user in all users；
Alternatively, then the corresponding confidence level of any user is the first preset value in all users if 0；
Wherein, the corresponding confidence level of any user indicates in all users are as follows:
T (u) indicates the corresponding confidence level of any user in all users, and T (u, r) indicates any resource in resource collection
Confidence level, RuIndicate resource collection, | Ru| indicate the quantity of resource in resource collection, q indicates the first preset value.
Optionally, acquiring unit 301 can be specifically used for:
It, and will be every according to the product of the quantity for the user for providing evaluation to resource any in resource collection and the first preset value
The product of value and the second preset value after the evaluation of estimate of a user is cumulative and, then with the first preset value value after being added, acquisition
The confidence level of any resource in resource collection, wherein the evaluation of estimate of user is the corresponding confidence level of user and user to resource set
The product for the evaluation indicated value that any resource provides in conjunction, the ratio of the quantity for the evaluation that all resources are provided with user；
Wherein, the confidence level of any resource indicates in the resource collection of any user in all users are as follows:
T (u, r)=q+q*s+ (1-q) ∑J=1,2 ..., sT(uj)*kj/tj；
T (u, r) indicates the confidence level of any resource in resource collection, and q indicates that the first preset value, s are indicated to resource collection
In any resource provide evaluation user quantity, (1-q) indicate the second preset value, T (uj) indicate to any in resource collection
Resource provides the corresponding confidence level of j-th of user, k in s user of evaluationjIndicate j-th of user to any in resource collection
The evaluation indicated value that resource provides, tjIndicate the quantity for the evaluation that j-th of user provides all resources.
Optionally, acquiring unit 301 can also be specifically used for:
The corresponding keyword of resource any in all resources is carried out in Bloom Filter after Hash operation slotting
Enter operation, obtains the mark of any resource in all resources, and by the keyword carried in resource request after Hash operation
Insertion operation is carried out, in Bloom Filter to obtain the mark of resource request；
Obtain the Hamming distance of the mark of the mark and resource request of any resource in all resources from；
Ratio of the Hamming distance from the length n with Bloom Filter is obtained, the keyword of any resource in all resources
Matching degree with the keyword carried in resource request is the difference of 1 with ratio；
Wherein, the matching degree of the keyword carried in the keyword of any resource and resource request in all resources indicates
Are as follows:
S indicates the matching degree of the keyword carried in the keyword of any resource and resource request in all resources, and L is indicated
Hamming distance is from n indicates the length of Bloom Filter.
Optionally, the corresponding recommendation of any resource indicates in all resources are as follows:
Mi=T (up)*Si；
Wherein, MiIndicate the corresponding recommendation of any resource i, T (u in all resourcesp) indicate any resource in all resources
The confidence level of i owning user p, SiIndicate the keyword carried in the keyword of any resource i and resource request in all resources
Matching degree.
The embodiment of the present invention provides a kind of control device 300, and the control device 300 in resource sharing system receives money
When source is requested, the corresponding confidence level of user in resource sharing system is obtained, which can be by characterize the user
The degree that other users are trusted, and according to resource in the corresponding confidence level of resource owning user and the resource sharing system of acquisition
Keyword and resource request in the matching degree of keyword calculate recommendation, thus for the big money of user's preferential recommendation recommendation
Source, i.e. owning user be corresponding with a high credibility and the matching degree height of its keyword and the keyword in resource request is to more meet
The resource of user resources demand.Due to resource sharing system can preferential recommendation more can by other users trust user provide
Resource, and the User reliability with insincere behavior is low, it is difficult to trusted by other users, the resource which provides is also past
Toward the resource requirement for not meeting other users thus will not be by preferential recommendation, it is thus possible to solve in the prior art due to user's
The low problem of the accuracy of resource recommendation caused by insincere behavior.
In several embodiments provided herein, it should be understood that disclosed device and method can pass through it
Its mode is realized.For example, apparatus embodiments described above are merely indicative, for example, the division of the unit, only
Only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or components can be tied
Another system is closed or is desirably integrated into, or some features can be ignored or not executed.Another point, it is shown or discussed
Mutual coupling, direct-coupling or communication connection can be through some interfaces, the INDIRECT COUPLING or logical of device or unit
Letter connection can be electrical property, mechanical or other forms.
In addition, each functional unit can integrate to be handled at one in equipment and system in each embodiment of the present invention
In unit, being also possible to the independent physics of each unit includes that can also be integrated in one unit with two or more units.
And above-mentioned each unit both can take the form of hardware realization, can also add the form of SFU software functional unit real using hardware
It is existing.
This can be accomplished by hardware associated with program instructions for all or part of the steps of realization above method embodiment, preceding
The program stated can be stored in a computer readable storage medium, and when being executed, execute includes that the above method is real to the program
The step of applying；And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (Read Only Memory, abbreviation
ROM), random access memory (Random Access Memory, abbreviation RAM), magnetic or disk etc. are various can store
The medium of program code.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. a kind of resource recommendation method characterized by comprising
When receiving resource request, the corresponding confidence level of all users in resource sharing system is obtained, and is obtained all
The matching degree of the keyword carried in the corresponding keyword of resource and the resource request, any use in all users
The quantity that evaluation user is provided according to the confidence level at family, the evaluation of estimate for each providing evaluation user obtain；
For any user in all users, the corresponding confidence level packet of any user described in resource sharing system is obtained
It includes:
Determine whether the resource quantity in all users in the resource collection of any user is 0；
If not 0, then obtain the confidence level of any resource in the resource collection of any user in all users, comprising: root
According to the quantity for the user for providing evaluation to any resource in the resource collection and the product of the first preset value, described in will be each
The product of value and the second preset value after the evaluation of estimate of user is cumulative and, then with the first preset value value after being added, obtain
Take the confidence level of any resource in the resource collection, wherein the evaluation of estimate of the user is the corresponding confidence level of the user
The product for the evaluation indicated value that any resource in the resource collection is provided with the user, with the user to described all
The ratio of the quantity for the evaluation that resource provides；
Then, obtain each resource in the resource collection confidence level it is cumulative after value and the resource collection in resource quantity
Ratio, the ratio be all users in the corresponding confidence level of any user；
Alternatively, then the corresponding confidence level of any user is the first preset value in all users if 0；
According to all resources difference corresponding confidence levels of owning user and the corresponding keyword of all resources with
The matching degree of the keyword carried in the resource request obtains the corresponding recommendation of all resources；
The recommendation is ranked up according to the size of the recommendation, and is chosen after sequence according to sequence from big to small
Preceding v recommendations, and the corresponding resource of the v recommendations before showing.
2. the method according to claim 1, wherein the corresponding confidence level table of any user in all users
It is shown as:
T (u) indicates the corresponding confidence level of any user in all users, and T (u, r) indicates any money in the resource collection
The confidence level in source, RuIndicate the resource collection, | Ru| indicate the quantity of resource in the resource collection, q indicates that described first is pre-
If value.
3. the method according to claim 1, wherein
The confidence level of any resource indicates in the resource collection of any user in all users are as follows:
T (u, r)=q+q*s+ (1-q) ∑J=1,2 ..., sT(uj)*kj/tj；
T (u, r) indicates the confidence level of any resource in the resource collection, and q indicates that first preset value, s are indicated to described
Any resource provides the quantity of the user of evaluation in resource collection, and (1-q) indicates second preset value, T (uj) indicate to institute
It states any resource in resource collection and provides the corresponding confidence level of j-th of user, k in s user of evaluationjIt indicates described j-th
The evaluation indicated value that user provides any resource in the resource collection, tjIndicate j-th of user to all moneys
The quantity for the evaluation that source provides.
4. the method according to claim 1, wherein obtaining resource for any resource in all resources
The matching degree of the keyword carried in the keyword of any resource described in shared system and the resource request includes:
The corresponding keyword of resource any in all resources is carried out in Bloom Filter after Hash operation slotting
Enter operation, obtains the mark of any resource in all resources, and the keyword carried in the resource request is passed through into institute
Insertion operation is carried out in the Bloom Filter after stating Hash operation, to obtain the mark of the resource request；
Obtain the Hamming distance of the mark of any resource in all resources and the mark of the resource request from；
Obtain ratio of the Hamming distance from the length n with the Bloom Filter, any resource in all resources
Keyword and the resource request in the matching degree of keyword that carries be 1 with the ratio difference；
Wherein, the matching degree table of the keyword carried in the keyword of any resource and the resource request in all resources
It is shown as:
S indicates the matching degree of the keyword carried in the keyword of any resource and the resource request in all resources, L
Indicate the Hamming distance from n indicates the length of the Bloom Filter.
5. method according to claim 1-4, which is characterized in that any resource is corresponding in all resources
Recommendation indicates are as follows:
Mi=T (up)*Si；
Wherein, MiIndicate the corresponding recommendation of any resource i, T (u in all resourcesp) indicate any in all resources
The confidence level of resource i owning user p, SiIt indicates in all resources in the keyword of any resource i and the resource request
The matching degree of the keyword of carrying.
6. a kind of control device characterized by comprising
Acquiring unit, it is corresponding credible for when receiving resource request, obtaining all users in resource sharing system
Degree, and obtain the matching degree of the keyword carried in the corresponding keyword of all resources and the resource request, the institute
There are the quantity for providing evaluation user in user according to the confidence level of any user, each evaluation of estimate for providing evaluation user to obtain
It takes；
Determine whether the resource quantity in all users in the resource collection of any user is 0；
If not 0, then obtain the confidence level of any resource in the resource collection of any user in all users, comprising: root
According to the quantity for the user for providing evaluation to any resource in the resource collection and the product of the first preset value, described in will be each
The product of value and the second preset value after the evaluation of estimate of user is cumulative and, then with the first preset value value after being added, obtain
Take the confidence level of any resource in the resource collection, wherein the evaluation of estimate of the user is the corresponding confidence level of the user
The product for the evaluation indicated value that any resource in the resource collection is provided with the user, with the user to described all
The ratio of the quantity for the evaluation that resource provides；
Then, obtain each resource in the resource collection confidence level it is cumulative after value and the resource collection in resource quantity
Ratio, the ratio be all users in the corresponding confidence level of any user；
Alternatively, then the corresponding confidence level of any user is the first preset value in all users if 0；
According to all resources difference corresponding confidence levels of owning user and the corresponding keyword of all resources with
The matching degree of the keyword carried in the resource request obtains the corresponding recommendation of all resources；
Control unit, for being ranked up according to the size of the recommendation to the recommendation, and according to from big after sequence
The v recommendations before being chosen to small sequence, and the corresponding resource of the v recommendations before showing.
7. control device according to claim 6, which is characterized in that any user is corresponding credible in all users
Degree indicates are as follows:
T (u) indicates the corresponding confidence level of any user in all users, and T (u, r) indicates any money in the resource collection
The confidence level in source, RuIndicate the resource collection, | Ru| indicate the quantity of resource in the resource collection, q indicates that described first is pre-
If value.
8. control device according to claim 6, which is characterized in that
The confidence level of any resource indicates in the resource collection of any user in all users are as follows:
T (u, r)=q+q*s+ (1-q) ∑J=1,2 ..., sT(uj)*kj/tj；
T (u, r) indicates the confidence level of any resource in the resource collection, and q indicates that first preset value, s are indicated to described
Any resource provides the quantity of the user of evaluation in resource collection, and (1-q) indicates second preset value, T (uj) indicate to institute
It states any resource in resource collection and provides the corresponding confidence level of j-th of user, k in s user of evaluationjIt indicates described j-th
The evaluation indicated value that user provides any resource in the resource collection, tjIndicate j-th of user to all moneys
The quantity for the evaluation that source provides.
9. control device according to claim 6, which is characterized in that the acquiring unit is specifically used for:
The corresponding keyword of resource any in all resources is carried out in Bloom Filter after Hash operation slotting
Enter operation, obtains the mark of any resource in all resources, and the keyword carried in the resource request is passed through into institute
Insertion operation is carried out in the Bloom Filter after stating Hash operation, to obtain the mark of the resource request；
Obtain the Hamming distance of the mark of any resource in all resources and the mark of the resource request from；
Obtain ratio of the Hamming distance from the length n with the Bloom Filter, any resource in all resources
Keyword and the resource request in the matching degree of keyword that carries be 1 with the ratio difference；
Wherein, the matching degree table of the keyword carried in the keyword of any resource and the resource request in all resources
It is shown as:
S indicates the matching degree of the keyword carried in the keyword of any resource and the resource request in all resources, L
Indicate the Hamming distance from n indicates the length of the Bloom Filter.
10. according to the described in any item control devices of claim 6-9, which is characterized in that any resource in all resources
Corresponding recommendation indicates are as follows:
Mi=T (up)*Si；
Wherein, MiIndicate the corresponding recommendation of any resource i, T (u in all resourcesp) indicate any in all resources
The confidence level of resource i owning user p, SiIt indicates in all resources in the keyword of any resource i and the resource request
The matching degree of the keyword of carrying.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510201816.5A CN104796478B (en) | 2015-04-24 | 2015-04-24 | A kind of resource recommendation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510201816.5A CN104796478B (en) | 2015-04-24 | 2015-04-24 | A kind of resource recommendation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104796478A CN104796478A (en) | 2015-07-22 |
| CN104796478B true CN104796478B (en) | 2019-03-15 |
Family
ID=53560986
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510201816.5A Active CN104796478B (en) | 2015-04-24 | 2015-04-24 | A kind of resource recommendation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104796478B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111199412B (en) * | 2018-11-19 | 2023-05-05 | 腾讯科技（深圳）有限公司 | Information processing method, apparatus, computer device, and storage medium |
| CN109919715A (en) * | 2019-02-12 | 2019-06-21 | 秒针信息技术有限公司 | The sharing method and device of article, storage medium and electronic device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103246677A (en) * | 2012-02-13 | 2013-08-14 | 广州淘信互联网科技有限公司 | Search method and search system on basis of social intercourse |
| CN103294677A (en) * | 2012-02-22 | 2013-09-11 | 阿里巴巴集团控股有限公司 | Searching method of electronic commerce search engine and electronic commerce search engine |
| CN104102748A (en) * | 2014-08-08 | 2014-10-15 | 中国联合网络通信集团有限公司 | Method and device for file mapping and method and device for file recommendation |
| CN104462333A (en) * | 2014-12-03 | 2015-03-25 | 上海耀肖电子商务有限公司 | Shopping search recommending and alarming method and system |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9514236B2 (en) * | 2005-08-19 | 2016-12-06 | Resource Consortium Limited | Recommendation network |
| CN103150399A (en) * | 2013-03-26 | 2013-06-12 | 常州德诗蓝电子科技有限公司 | Information recommending method based on trust in social network |
| CN103164539B (en) * | 2013-04-15 | 2016-12-28 | 中国传媒大学 | A kind of combination user evaluates and the interactive image retrieval method of mark |
| CN104298785B (en) * | 2014-11-12 | 2017-05-03 | 中南大学 | Searching method for public searching resources |
-
2015
- 2015-04-24 CN CN201510201816.5A patent/CN104796478B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103246677A (en) * | 2012-02-13 | 2013-08-14 | 广州淘信互联网科技有限公司 | Search method and search system on basis of social intercourse |
| CN103294677A (en) * | 2012-02-22 | 2013-09-11 | 阿里巴巴集团控股有限公司 | Searching method of electronic commerce search engine and electronic commerce search engine |
| CN104102748A (en) * | 2014-08-08 | 2014-10-15 | 中国联合网络通信集团有限公司 | Method and device for file mapping and method and device for file recommendation |
| CN104462333A (en) * | 2014-12-03 | 2015-03-25 | 上海耀肖电子商务有限公司 | Shopping search recommending and alarming method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104796478A (en) | 2015-07-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10042911B2 (en) | Discovery of related entities in a master data management system | |
| CN105247507B (en) | Method, system and storage medium for the influence power score for determining brand | |
| US9569499B2 (en) | Method and apparatus for recommending content on the internet by evaluating users having similar preference tendencies | |
| CN103870505B (en) | One kind inquiry words recommending method and query word commending system | |
| CN109829775A (en) | A kind of item recommendation method, device, equipment and readable storage medium storing program for executing | |
| US8260903B2 (en) | System and method for assessing the usability and accessibility of Web 2.0 features and functionalities of websites | |
| CN106233279A (en) | The content included based on digizine is to digizine server user's content recommendation | |
| CN106469169A (en) | Information processing method and device | |
| US9135307B1 (en) | Selectively generating alternative queries | |
| US10185725B1 (en) | Image annotation based on label consensus | |
| CA2869888C (en) | Discovering spam merchants using product feed similarity | |
| JP6203304B2 (en) | Information processing apparatus, information processing method, and information processing program | |
| JP2019512143A (en) | Data processing method and apparatus | |
| JP5048852B2 (en) | Search device, search method, search program, and computer-readable recording medium storing the program | |
| CN104796478B (en) | A kind of resource recommendation method and device | |
| CN106156351B (en) | Multimedia resource recommendation information generation method and device | |
| CN104123321B (en) | A kind of determining method and device for recommending picture | |
| CN110147514A (en) | A kind of resource exhibition method, device and its equipment | |
| US8949327B2 (en) | Method and device to provide trusted recommendations of websites | |
| US9110943B2 (en) | Identifying an image for an entity | |
| CN106776910A (en) | The display methods and device of a kind of Search Results | |
| CN104573132B (en) | Song lookup method and device | |
| Roszkowska et al. | Can the holistic preference elicitation be used to determine an accurate negotiation offer scoring system? A comparison of direct rating and UTASTAR techniques | |
| CN103984754A (en) | Search system and search method | |
| CN104715022B (en) | A kind of related search method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |