CN104463594B - The method and its device of computing object influence power - Google Patents
The method and its device of computing object influence power Download PDFInfo
- Publication number
- CN104463594B CN104463594B CN201310435049.5A CN201310435049A CN104463594B CN 104463594 B CN104463594 B CN 104463594B CN 201310435049 A CN201310435049 A CN 201310435049A CN 104463594 B CN104463594 B CN 104463594B
- Authority
- CN
- China
- Prior art keywords
- event
- power fraction
- time
- relevant
- fraction
- 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
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The disclosure a kind of method and its device of computing object influence power, this method include：At least one event is obtained, the event includes following information：Time To Event, object, the relevant theme of the event for issuing event；It is to distinguish distribution time influence power fraction with the relevant each event of given theme according to the sequencing of Time To Event；The definite and relevant event of object, and according to the influence power fraction of the time effects power fraction of the event calculating object.The computation complexity of computing object influence power is reduced by the application and reduces overhead.
Description
Technical field
This application involves Internet technical field more particularly to a kind of methods and its device of computing object influence power.
Background technology
In internet arena, one object of evaluation is generally required（Such as Internet user, the owner of website, interconnection
Net provides businessman of commodity or service etc.）Influence power.By the assessment to businessman's influence power, those can be found immediately following trend
, deep popular businessman.In a commercial articles searching engine, these correlations to the businessman of city's field sensitive can be expanded
Commodity show probability to promote user experience.On a C2C website, that can be found out according to the assessment of user force
A little shopping intelligents（The people being expert at very much in terms of how selecting commodity）.It, can also be according to user on a SNS website
Those users for often delivering valuable opinion are found out in the assessment of influence power.
The associated sequence of events of certain an object（Such as the object is having the behavior letter of issue different themes at different moments
Breath）, can be used for assessing the influence power of the object.
The prior art, it is necessary to first extract the information transfering relation between object, is calculated in the influence power of computing object
Go out the strength of association between object, then the influence power of computing object again.But in practice, as the time elapses, a net
Number of objects in the system of road can gradually increase, and the quantity of even object is linearly increasing, and the calculating of object relationship is opened
Pin also will be into a square times growth.In addition, when some hot spot themes occur, hot spot master is often insanely forwarded between object
Relevant message is inscribed, this computing cost for also resulting in analysis object relationship rapidly increases.
In summary, it is known that the technical solution of computing object influence power exists in the prior art opens as the time increases to calculate
The defects of pin is in geometry quantity increase, it is therefore necessary to propose that improved technological means solves the above problems.
The content of the invention
The main purpose of the application is to provide a kind of method and its device of computing object influence power, to solve existing skill
The defects of existing expense for increasing computing object influence power with the time of art is in geometry quantity increase.
In order to solve the above technical problems, according to the one side of the application, a kind of side of computing object influence power is provided
Method, including：At least one event is obtained, the event includes following information：Time To Event, issue event object,
The relevant theme of the event；According to the sequencing of Time To Event for when being distributed respectively with the relevant each event of given theme
Between influence power fraction；The definite and relevant event of object, and according to the shadow of the time effects power fraction of the event calculating object
Ring power fraction.
Wherein, the sequencing according to Time To Event is when distributing respectively with the relevant each event of given theme
Between influence fraction the step of, including：It will be with the relevant each event ordering of given theme according to the sequencing of Time To Event
Form sequence of events；Distinguishing distribution time according to previously given descending function for each event in the sequence of events influences
Power fraction, wherein, the fraction of the time effects power event to rank forefront in the sequence of events is higher than the time effects of the event to lean backward
Power fraction.
Wherein, the descending function includes one below：Geometric sequence, arithmetic sequence, consider time span etc. compare sequence
Row, the geometric sequence for considering sequence of events population size.
Wherein, if only comprising an event in the sequence of events, the time effects power fraction for event distribution is
Pre-set constant value.
Wherein, the definite and relevant event of object, and the object is calculated according to the time effects power fraction of event
Influence power fraction the step of, including：The definite and relevant multiple events of object, by the time effects power fraction of the plurality of event
It carries out average computation and obtains the influence power fraction of the object.
Wherein, the definite and relevant event of object, and the object is calculated according to the time effects power fraction of event
Influence power fraction the step of, including：The definite and relevant multiple events of object, by the time effects power fraction of the plurality of event
Average computation, which is weighted, according to Time To Event obtains the influence power fraction of the object.
Wherein, the definite and relevant event of object, and the object is calculated according to the time effects power fraction of event
Influence power fraction the step of, including：The definite and relevant multiple events of object, by the time effects power fraction of the plurality of event
Average computation, which is weighted, according to previously given event weights obtains the influence power fraction of the object.
Wherein, if an event is related to multiple themes, previously given event weights are related to one below：It is multiple
The average value of topic weights；The minimum value of multiple topic weights；The maximum of multiple topic weights；The middle position of multiple topic weights
Value.
Wherein, the definite and relevant event of object, and the object is calculated according to the time effects power fraction of event
Influence power fraction the step of, including：It will polymerize with the relevant event of the object according to theme, this is calculated respectively
Time effects power fraction of the object under each theme；It is calculated according to time effects power fraction of the object under each theme
To the influence power fraction of the object.
Wherein, in the step of being polymerize with the relevant event of the object according to theme, further include：Same subject
Under event the time effects power fraction for averagely obtaining the event under the theme is weighted according to time factor；It should in basis
In the step of influence power fraction of the object is calculated in time effects power fraction of the object under each theme, further include：Often
Time effects power fraction under a theme is weighted the shadow that the object averagely, is calculated according to previously given topic weights
Ring power fraction.
According to the another aspect of the application, a kind of device of computing object influence power is additionally provided, including：Obtain mould
Block, for obtaining at least one event, the event includes following information：Time To Event, the object for issuing event, the thing
The relevant theme of part；Event score computing module, for being related to given theme according to the sequencing of Time To Event
Each event difference distribution time influence power fraction；Object influences power fraction computing module, for determining and the relevant thing of object
Part, and according to the influence power fraction of the time effects power fraction of the event calculating object.
Wherein, the event score computing module includes：Event ordering module, for the priority according to Time To Event
Order will form sequence of events with the relevant each event ordering of given theme；Event score distribution module, for according to giving in advance
Fixed descending function distinguishes distribution time influence power fraction for each event in the sequence of events, wherein, the event sequence
The time effects power fraction of the event to rank forefront in row is higher than the time effects power fraction of the event to lean backward.
Wherein, the descending function includes one below：Geometric sequence, arithmetic sequence, consider time span etc. compare sequence
Row, the geometric sequence for considering sequence of events population size.
Wherein, if only comprising an event in the sequence of events, the event score distribution module is the event point
The time effects power fraction matched somebody with somebody is pre-set constant value.
Wherein, the object influences power fraction computing module is further used for：The definite and relevant multiple events of object, will
The time effects power fraction of the plurality of event carries out average computation and obtains the influence power fraction of the object.
Wherein, the object influences power fraction computing module is further used for：The definite and relevant multiple events of object, will
The time effects power fraction of the plurality of event is weighted average computation according to Time To Event and obtains the influence power of the object
Fraction.
Wherein, the object influences power fraction computing module is further used for：The definite and relevant multiple events of object, will
The time effects power fraction of the plurality of event is weighted average computation according to previously given event weights and obtains the object
Influence power fraction.
Wherein, if an event is related to multiple themes, previously given event weights are related to one below：It is multiple
The average value of topic weights；The minimum value of multiple topic weights；The maximum of multiple topic weights；The middle position of multiple topic weights
Value.
Wherein, the object influences power fraction computing module is further used for：Will with the relevant event of the object according to
Theme is polymerize, and time effects power fraction of the object under each theme is calculated respectively, wherein, under same subject
Event is weighted the time effects power fraction for averagely obtaining the event under the theme according to time factor；Existed according to the object
The influence power fraction of the object is calculated in time effects power fraction under each theme, wherein, the time shadow under each theme
It rings power fraction and the influence power fraction that the object averagely, is calculated is weighted according to previously given topic weights.
According to the technical solution of the application, the important of the object is evaluated by the fraction situation of the event according to object publishing
Property, it is not intended that the propagation relation between event and between object, thus reduces computation complexity and reduces system and open
Pin, so as to the influence power of computing object quickly on large-scale data.
Description of the drawings
Attached drawing described herein is used for providing further understanding of the present application, forms the part of the application, this Shen
Schematic description and description please does not form the improper restriction to the application for explaining the application.In the accompanying drawings：
Fig. 1 is the flow chart according to the method for the computing object influence power of the embodiment of the present application；
Fig. 2 is the structure diagram according to the device of the computing object influence power of the application one embodiment；
Fig. 3 is the structure diagram according to the device of the computing object influence power of another embodiment of the application.
Specific embodiment
To make the purpose, technical scheme and advantage of the application clearer, below in conjunction with drawings and the specific embodiments, to this
Application is described in further detail.
According to an embodiment of the present application, a kind of method of computing object influence power is provided.Fig. 1 is implemented according to the application
The flow chart of the method for the computing object influence power of example, as shown in Figure 1, the method comprising the steps of S102-S106.
Step S102, obtains at least one event, and the event includes following information：Time To Event issues the thing
The object of part, the relevant theme of the event.
The method of the computing object influence power of the application can be applied to business website, commodity website or social networks
（SNS）Etc. plurality of application scenes.In different application scenarios, the information such as event, object and theme can represent different
Meaning, in simple terms, object just refer to the user registered in network, and event refers to specific behavior of the object on network,
And theme is then the core of the event.
The result of the sequencing arrangement for the time that the set of the event occurred corresponding to same subject occurs according to event
The sequence of events of the theme can be formed.According to Time To Event to a certain event in the event belonging to sequence of events in
The evaluation of importance is denoted as the time effects power fraction of event herein.
The influence power of object is then the synthesis of the time effects power fraction of each event of the object publishing.
Be exemplified below this application involves some concepts.
For a business website, event can be user in the business website to the operation behavior of a certain commodity, such as
The behavior of merchandise news occurs for seller or buyer buys the behavior of commodity.In such an example, object is to make the behavior
User（That is seller or buyer）；Theme is the behavior corresponding commodity；The time that event occurs issues the commodity for the seller
Time or buyer buy the time of the commodity；The sequence of events of some theme is the commodity operation behavior that will belong to a certain commodity together
The result that set sorts according to the sequencing of issuing time；Time effects power fraction is used to evaluate certain in some sequence of events
Importance of one event in Time To Event dimension.The time effects power fraction of event and event of the event belonging to it
Sequence in sequence according to time of origin is related.The time of origin of a certain event is compared in the sequence of events belonging to the event
The time of origin of other events is more early, then the time effects power fraction of the event is higher.The influence power of object is issued for the user
All events time effects power fraction synthesis.If the time effects power fraction of each event of a certain seller's issue is always
Higher, then the influence power of the seller is generally large, illustrates that the seller is deeply popular.If each event of a certain buyer's issue
Time effects power fraction it is always higher, the influence power of the buyer is generally large, illustrates the fashion trend to commodity of the buyer
Assurance ability it is stronger.
For a SNS system, event issues article for user in SNS system（Or model）Behavior, and object is
The registered user of this article is issued, theme is the corresponding topic of this article；The time that event occurs is object publishing article（Or note
Son）Time；The sequence of events of some theme is the behavior set for the article that will belong to same topic according to the elder generation of issuing time
The result of order sequence afterwards；The time effects power fraction of event is for evaluating in the sequence of events of some topic, a certain event
With the relevant importance of time of origin；The influence power of object is comprehensive for the time effects fraction of all articles of the user's issue
It closes.If the time effects power fraction of the article of the user's issue is always higher, the influence power of the user is generally large, explanation
The user often delivers valuable opinion.
It should be noted that above-mentioned business website or the theme of commodity website are illustrated by taking " product " as an example, this Shen
It please be not limited to this, it can also be with other dimensions such as brand, material, style or standard product unit（SPU）Deng progress table
Show.
Step S104, according to the sequencing of Time To Event be and each thing in the relevant sequence of events of given theme
Part distinguishes distribution time influence power fraction.
For given theme, whole events containing the theme are selected, and it is suitable according to the priority of Time To Event
The event ordering selected is formed sequence of events by sequence.Then, it is each in sequence of events according to previously given descending function
Event distributes a fraction to get to time influence power fraction.Wherein, the time effects power point of the event to rank forefront in sequence of events
Number is always above the time effects power fraction of the event to lean backward in sequence of events.
The time effects power fraction of event is used to evaluate the importance of event, in above-mentioned processing procedure, identical theme
Event time effects power fraction（Importance）Decay at any time, thus without propagating relation between consideration event,
So as to reduce computation complexity.
Wherein, the descending function can include but not limited to one below there are many realization method：
（1）By geometric sequence F (n)=a^n, the time effects power fraction of nth event in sequence of events is drawn, wherein,
A is the constant between previously given 0 to 1, the time effects power fraction of nth event in F (n) i.e. sequences of events；
（2）By arithmetic sequence F (n)=- b*n, the time effects power fraction of nth event in sequence of events is drawn,
In, b is previously given constant, the time effects power fraction of nth event in F (n) i.e. sequences of events；
（3）By geometric sequence F (n, t, t_last)=(a^n)/(t-t_last) of certain consideration time span, draw
The time effects power fraction of nth event in sequence of events, wherein, the Time To Event of nth event is t, (n-1)th thing
The Time To Event of part is t_last, the time effects power fraction of nth event in F (n, t, t_last) i.e. sequences of events；
（4）Geometric sequence F (n, N)=(a^ (n/N)) of sequence population size is considered by certain, draws sequence of events
The time effects power fraction of middle nth event, wherein, N is the sum of whole events in sequence of events；F (n, N) i.e. sequences of events
The time effects power fraction of middle nth event.
In above processing, when only including an event in sequence of events, the time effects power of the event is scored at
A previously given constant value.It should be noted that in practical applications, corresponding descending letter should be selected according to actual conditions
Number is calculated, and the application is not limited this.
Step S106, the definite and relevant event of object, and the shadow of the time effects power fraction computing object according to event
Ring power fraction.For each object, according to time effects power fraction of the relevant each event of the object under relevant each theme,
Calculate the influence power fraction of the object.That is, calculating the influence power fraction with the relevant each event of object first, then integrate
The time effects power fraction of the corresponding all events of the object is calculated.
Specifically, the calculation of the influence power fraction of the object of the application includes but not limited in the following manner（1）-（4）
One of.
（1）The time effects power fraction of the plurality of event is carried out average computation by the definite and relevant multiple events of object
Obtain the influence power fraction of the object；
（2）The definite and relevant multiple events of object, the time effects power fraction of the plurality of event is occurred according to event
Time is weighted average computation and obtains the influence power fraction of the object；
（3）The definite and relevant multiple events of object, by the time effects power fraction of the plurality of event according to previously given
Event weights be weighted average computation and obtain the influence power fraction of the object, wherein it is possible to be thing according to different themes
Part sets weight；
Wherein, if an event is related to multiple themes, previously given event weights are related to one below：
The average value of multiple topic weights；
The minimum value of multiple topic weights；
The maximum of multiple topic weights；
The median of multiple topic weights.
（4）It will polymerize first with the relevant event of object according to theme, the object is calculated respectively in each master
Time effects power fraction under topic, further, the event under same subject is weighted according to time factor is averagely somebody's turn to do
Time effects power fraction of the event under the theme；Then calculated according to time effects power fraction of the object under each theme
Time effects power fraction of the object under all themes is obtained, further, the time effects power fraction under different themes is pressed
The influence power fraction that the object averagely, is calculated is weighted according to previously given topic weights.
It is above-mentioned（4）Be than more preferably method, when an object is related to a large amount of events, and event is related to a large amount of themes,
If using（4）Method, can eaily realize more complicated weighting logic, and convenient in similar map-reduce shapes
It is realized in the distributed system of formula, speed up processing.
In practice, corresponding calculation can be selected according to actual needs, and the application is without limiting.
According to an embodiment of the present application, the importance of the object is evaluated by the fraction situation of the event of object publishing（Such as
The score of one relevant event of object of fruit is generally high, then the influence power of the object is inevitable high）, it is not intended that the biography between object
Relation is broadcast, thus reduces computation complexity, so as to fulfill the influence power of computing object quickly on large-scale data.
The process of the computing object influence power of the application is described with reference to specific example.
Example one, the influence power for calculating certain shopping platform seller, by seller issue the new product time be used as weigh he
Influence power in this market.
In this example, event issues new product for seller in certain shopping platform, and object is seller, and theme is product.Tool
Body includes following 8 events altogether：（1）Seller's first was on 03 01st, 2013 release product A；（2）Seller's second was in 2013 03
Moon release product B on the 13rd；（3）Seller's second was on 03 15th, 2013 release product A；（4）Seller's second was on 03 21st, 2013
Release product C；（5）Seller third was on 04 01st, 2013 release product C；（6）Seller's first was in issue production on 04 02nd, 2013
Product B；（7）Seller's first was on 04 05th, 2013 release product C；（8）The retainer of a big family is sold on 04 16th, 2013 release product A.
（1）First, the information of event is obtained, in chronological sequence sort the relevant event summary of identical product and sequentially shape
Into sequence of events：
Product A events 1->Event 3->Event 8
Product B events 2->Event 6
Products C event 4->Event 5->Event 7
（2）Time-sequencing of certain event on its affiliated product is denoted as n, and n is the natural number more than or equal to 1.Using etc.
It is the relevant each event distribution time influence power fraction of the product than sequence, i.e., by the relevant each event of the product
Time effects power fraction forms geometric sequence.The geometric sequence can be expressed as S (x)=2^ (1-n), S (x) be event when
Between influence power fraction.According to the above-mentioned mode for event distribution time influence power fraction, determine each product it is relevant each
The score of event.
Therefore, the time effects power of the associated events 1 of product A is scored at 1, the time shadow of the associated events 3 of product A
It rings power and is scored at 1/2, the time effects power of the associated events 8 of product A is scored at 1/4.
The time effects power of the associated events 2 of product B is scored at 1, the time effects power of the associated events 6 of product B
It is scored at 1/2.
The time effects power of the associated event 4 of products C is scored at 1, the time effects power of the associated event 5 of products C
1/2 is scored at, the time effects power of the associated event 7 of products C is scored at 1/4.
（3）The influence power of seller is scored at the average mark of the time effects power score of all events involved by the seller,
Then：
Seller's first is related to event 1, event 6 and event 7 totally 3 events, and the influence power of seller's first is scored at above-mentioned 3 events
The sum of time effects power score average mark, i.e.,：(1+1/2+1/4)/3=0.583；
Seller's second is related to event 2, event 3 and event 4 totally 3 events, and the influence power of seller's second is scored at above-mentioned 3 events
The sum of time effects power score average mark, i.e.,：(1+1/2+1)/3=0.833；
Seller third pertains only to event 5, and the influence power score of seller third is the score of the time effects power of event 5：0.5；
It sells the retainer of a big family and pertains only to event 8, the influence power score for selling the retainer of a big family is the score of the time effects power of event 8：0.25.
The influence power of seller is calculated with simple calculation by the above process.The influence power of seller's second from fraction
Highest scoring, this is consistent with the actual conditions that data are embodied, because the time ratio of seller's second release product on the whole
Other seller is early.
Example two, the influence power for calculating certain shopping platform seller by seller are used as and weigh him at this issue new product time
Influence power in a market.
In this example, event issues different product for seller in certain shopping platform（Belong to different in product classification tree
Classification）, object is seller, and theme is the product classification belonging to product.Specifically include following 6 events altogether：（1）Seller's first in
On 04 21st, 2013 release products, the affiliated classification of product are A1/B2；（2）Seller's second on 04 22nd, 2013 release products,
The affiliated classification of product is A1/B2；（3）For seller's second on 04 24th, 2013 release products, the affiliated classification of product was A2/B6；（4）
For seller's first on 04 25th, 2013 release products, the affiliated classification of product was A2/B3；（5）Seller's second was on 04 27th, 2013
Release product, the affiliated classification of product are A1/B4；（6）Seller's second is on 04 29th, 2013 release products, the affiliated classification of product
A2/B5。
（1）The information of acquisition event is in chronological sequence arranged the relevant event summary of the affiliated classification of identical product and sequentially
Sequence forms sequence of events：
Product classification A1 events 1 (score 1)->Event 2 (score 1/2)->Event 5 (score 1/4)
Product classification A2 events 3 (score 1)->Event 4 (score 1/2)->Event 6 (score 1/4)
Product classification B2 events 1 (score 1)->Event 2 (score 1/2)
Product classification B3 events 4 (score 1)
Product classification B4 events 5 (score 1)
Product classification B5 events 6 (score 1)
Product classification B6 events 3 (score 1)
（2）Some the product classification of certain event belonging to it（It is denoted as c）On time-sequencing be denoted as n, n be more than or equal to
1 natural number.Geometric sequence event is used as the relevant each event distribution time influence power fraction of the product classification, it will
The time effects power fraction of the relevant each event of product classification forms geometric sequence.The geometric sequence can be expressed as
S (x, c)=2^ (1-n), S (x, c) are the time effects power fraction of event.According to above-mentioned for event distribution time influence power fraction
Mode, determine the time effects power score of each each relevant event of product classification.The seller is set in this example
Time effects power of the seller in the product class involved event now is scored in the time effects power of the product class now
Best result.
Therefore, the time effects power of the associated events 1 of product classification A1 is scored at 1, the associated things of product classification A1
The time effects power of part 2 is scored at 1/2, and the time effects power of the associated events 5 of product classification A1 is scored at 1/4.
The time effects power of the associated events 3 of product classification A2 is scored at 1, the associated events 4 of product classification A2
Time effects power is scored at 1/2, and the time effects power of the associated events 6 of product classification A2 is scored at 1/4.
The time effects power of the associated events 1 of product classification B2 is scored at 1, the associated events 2 of product classification B2
Time effects power is scored at 1/2.
The time effects power of the associated events 4 of product classification B3 is scored at 1.
The time effects power of the associated events 5 of product classification B4 is scored at 1.
The time effects power of the associated events 6 of product classification B5 is scored at 1.
The time effects power of the associated events 3 of product classification B6 is scored at 1.
（3）The influence power of seller is scored at being averaged for the score of the time effects power of all events involved by the seller
Point, then：
Seller's first is related to event 1 and event 4 totally 2 events and A1, A2, B2 and B3 totally 4 product classifications, seller's first
Influence power is scored at the average mark of the sum of the time effects power score of the corresponding event of above-mentioned 4 product classifications, i.e.,：(1+1/2+1
+1)/7=0.5；
Seller's second is related to event 2, event 3, event 5 and event 6 totally 4 events and A1, A2, B2, B4, B5 and B6 totally 6
A product classification（Only calculate best result of the seller in the product class involved event now）, the influence power of seller's second is scored at
The average mark of the sum of the time effects power score of the corresponding event of above-mentioned 6 product classifications, i.e. (1/2+1+1/2+1+1+1)/7=
0.714。
The influence power of seller is calculated with simple calculation by the above process.The influence power of seller's second from fraction
Score is higher than the influence power score of seller's first, this is consistent with the actual conditions that data are embodied, because several product classes
Mesh only has the product of seller's second.
It is to be appreciated that examples detailed above one and example two are merely illustrative, the protection domain of the application is not intended to limit.
According to an embodiment of the present application, a kind of device of computing object influence power is additionally provided, with reference to figure 2, described device
It includes at least：Acquisition module 22, event score computing module 24 and object influences power fraction computing module 26, are described below in detail
The 26S Proteasome Structure and Function of above-mentioned each module.
Acquisition module 22, for obtaining at least one event, the event includes following information：Time To Event, hair
The object of cloth event, the relevant theme of the event.
Event score computing module 24, for being relevant each with given theme according to the sequencing of Time To Event
Event distinguishes distribution time influence power fraction.
With reference to figure 3, the event score computing module 24 further comprises：Event ordering module 242 and event score point
With module 244.Event ordering module 242 is used for will be with the relevant each thing of given theme according to the sequencing of Time To Event
Part sorts to form sequence of events.Event score distribution module 244 and 242 phase of event ordering module couple, for according to giving in advance
Fixed descending function distinguishes distribution time influence power fraction for each event in the sequence of events, wherein, the event sequence
The time effects power fraction of the event to rank forefront in row is higher than the time effects power fraction of the event to lean backward.Further, the drop
Order function includes but not limited to one below：Geometric sequence, arithmetic sequence, the geometric sequence for considering time span, consideration event
The geometric sequence of sequence population size.
If in addition, only comprising an event in the sequence of events, the event score distribution module is the event point
The time effects power fraction matched somebody with somebody is pre-set constant value.
Object influences power fraction computing module 26 is coupled with 24 phase of event score computing module, related to object for determining
Event, and calculate according to the time effects power fraction of event the influence power fraction of the object.
In practical applications, the influence power of in the following manner computing object may be employed in object influences power fraction computing module 26
Fraction：
（1）The time effects power fraction of the plurality of event is carried out average computation by the definite and relevant multiple events of object
Obtain the influence power fraction of the object；
（2）The definite and relevant multiple events of object, the time effects power fraction of the plurality of event is occurred according to event
Time is weighted average computation and obtains the influence power fraction of the object；
（3）The definite and relevant multiple events of object, by the time effects power fraction of the plurality of event according to previously given
Event weights be weighted average computation and obtain the influence power fraction of the object, wherein it is possible to be thing according to different themes
Part sets weight；
Further, if an event is related to multiple themes, previously given event weights are related to one below：
The average value of multiple topic weights；
The minimum value of multiple topic weights；
The maximum of multiple topic weights；
The median of multiple topic weights.
（4）It will polymerize first with the relevant event of object according to theme, the object is calculated respectively in each master
Time effects power fraction under topic, further, the event under same subject is weighted according to time factor is averagely somebody's turn to do
Time effects power fraction of the event under the theme；Then calculated according to time effects power fraction of the object under each theme
The influence power fraction of the object is obtained, further, the time effects power fraction under each theme is according to previously given theme
Weight is weighted the influence power fraction that the object averagely, is calculated.
It is above-mentioned（4）Be than more preferably method, when an object is related to a large amount of events, and event is related to a large amount of themes,
If using（4）Method, can more conveniently realize more complicated weighting logic, and convenient in similar map-reduce forms
Distributed system on realize, speed up processing.
The operating procedure of the present processes is corresponding with the structure feature of device, can be cross-referenced, no longer repeats one by one.
In conclusion according to the above-mentioned technical proposal of the application, commented by the fraction situation of the event according to object publishing
The importance of the valency object, it is not intended that the propagation relation between object, thus reduces computation complexity and reduce system
Expense, so as to the influence power of computing object quickly on large-scale data.
In a typical configuration, computing device includes one or more processors (CPU), input/output interface, net
Network interface and memory.
Memory may include computer-readable medium in volatile memory, random access memory (RAM) and/or
The forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flashRAM).Memory is showing for computer-readable medium
Example.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology come realize information store.Information can be computer-readable instruction, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase transition internal memory (PRAM), static RAM (SRAM), moves
State random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electric erasable
Programmable read only memory (EEPROM), fast flash memory bank or other memory techniques, read-only optical disc read-only memory (CD-ROM),
Digital versatile disc (DVD) or other optical storages, magnetic tape cassette, the storage of tape magnetic rigid disk or other magnetic storage apparatus
Or any other non-transmission medium, the information that can be accessed by a computing device available for storage.It defines, calculates according to herein
Machine readable medium does not include temporary computer readable media (transitory media), such as data-signal and carrier wave of modulation.
It should also be noted that, term " comprising ", "comprising" or its any other variant are intended to nonexcludability
Comprising so that process, method, commodity or equipment including a series of elements are not only including those elements, but also wrap
Include other elements that are not explicitly listed or further include for this process, method, commodity or equipment it is intrinsic will
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that wanted including described
Also there are other identical elements in the process of element, method, commodity or equipment.
It will be understood by those skilled in the art that embodiments herein can be provided as method, system or computer program product.
Therefore, complete hardware embodiment, complete software embodiment or the embodiment in terms of combining software and hardware can be used in the application
Form.It is deposited moreover, the application can be used to can use in one or more computers for wherein including computer usable program code
Storage media（Including but not limited to magnetic disk storage, CD-ROM, optical memory etc.）The shape of the computer program product of upper implementation
Formula.
The foregoing is merely embodiments herein, are not limited to the application.For those skilled in the art
For, the application can have various modifications and variations.All any modifications made within spirit herein and principle are equal
Replace, improve etc., it should be included within the scope of claims hereof.
Claims (17)
- A kind of 1. method of computing object influence power, which is characterized in that including：At least one event is obtained, the event includes following information：Time To Event, the object for issuing event, the event Relevant theme；It is to distinguish distribution time influence power fraction with the relevant each event of given theme according to the sequencing of Time To Event；The definite and relevant event of object, and according to the influence power fraction of the time effects power fraction of the event calculating object；Wherein, the sequencing according to Time To Event is to distinguish distribution time shadow with the relevant each event of given theme The step of ringing power fraction, including：Sequence of events will be formed with the relevant each event ordering of given theme according to the sequencing of Time To Event；Distinguish distribution time influence power fraction according to previously given descending function for each event in the sequence of events, In, the time effects power fraction of the event to rank forefront in the sequence of events is higher than the time effects power fraction of the event to lean backward.
- 2. according to the method described in claim 1, it is characterized in that, the descending function includes one below：Geometric sequence, etc. Difference sequence, the geometric sequence for considering time span, the geometric sequence for considering sequence of events population size.
- If 3. according to the method described in claim 1, it is characterized in that, only comprising an event in the sequence of events, for The time effects power fraction of event distribution is pre-set constant value.
- 4. according to the method described in claim 1, it is characterized in that, it is described determine with the relevant event of object, and according to event Time effects power fraction the step of calculating the influence power fraction of the object, including：It is right to be obtained this by the definite and relevant multiple events of object for the time effects power fraction progress average computation of the plurality of event The influence power fraction of elephant.
- 5. according to the method described in claim 1, it is characterized in that, it is described determine with the relevant event of object, and according to event Time effects power fraction the step of calculating the influence power fraction of the object, including：The definite and relevant multiple events of object, the time effects power fraction of the plurality of event is carried out according to Time To Event Weighted average calculation obtains the influence power fraction of the object.
- 6. according to the method described in claim 1, it is characterized in that, it is described determine with the relevant event of object, and according to event Time effects power fraction the step of calculating the influence power fraction of the object, including：The definite and relevant multiple events of object weigh the time effects power fraction of the plurality of event according to previously given event Average computation is weighted again obtains the influence power fraction of the object.
- 7. it is if previously given according to the method described in claim 6, it is characterized in that, an event is related to multiple themes Event weights it is related to one below：The average value of multiple topic weights；The minimum value of multiple topic weights；The maximum of multiple topic weights；The median of multiple topic weights.
- 8. according to the method described in claim 1, it is characterized in that, it is described determine with the relevant event of object, and according to event Time effects power fraction the step of calculating the influence power fraction of the object, including：To polymerize with the relevant event of the object according to theme, be calculated respectively the object under each theme when Between influence power fraction；The influence power fraction of the object is calculated according to time effects power fraction of the object under each theme.
- 9. according to the method described in claim 8, it is characterized in that,In the step of being polymerize with the relevant event of the object according to theme, further include：Event under same subject The time effects power fraction for averagely obtaining the event under the theme is weighted according to time factor；The influence power fraction of the object is calculated according to time effects power fraction of the object under each theme the step of In, it further includes：Time effects power fraction under each theme is weighted averagely according to previously given topic weights, is calculated To the influence power fraction of the object.
- 10. a kind of device of computing object influence power, which is characterized in that including：Acquisition module, for obtaining at least one event, the event includes following information：Time To Event, issue event Object, the relevant theme of the event；Event score computing module, for being and given theme relevant each event point according to the sequencing of Time To Event Other distribution time influence power fraction；Object influences power fraction computing module, for the definite and relevant event of object, and according to the time effects power point of event Number calculates the influence power fraction of the object；Wherein, the event score computing module includes：Event ordering module, will be with the relevant each event ordering shape of given theme for the sequencing according to Time To Event Into sequence of events；Event score distribution module, for distinguishing according to previously given descending function for each event in the sequence of events Distribution time influence power fraction, wherein, the time effects power fraction of the event to rank forefront in the sequence of events is higher than the thing to lean backward The time effects power fraction of part.
- 11. device according to claim 10, which is characterized in that the descending function includes one below：Geometric sequence, Arithmetic sequence, the geometric sequence for considering time span, the geometric sequence for considering sequence of events population size.
- 12. device according to claim 10, which is characterized in that if only including an event in the sequence of events, The event score distribution module is that the time effects power fraction of event distribution is pre-set constant value.
- 13. device according to claim 10, which is characterized in that the object influences power fraction computing module is further used In：It is right to be obtained this by the definite and relevant multiple events of object for the time effects power fraction progress average computation of the plurality of event The influence power fraction of elephant.
- 14. device according to claim 10, which is characterized in that the object influences power fraction computing module is further used In：The definite and relevant multiple events of object, the time effects power fraction of the plurality of event is carried out according to Time To Event Weighted average calculation obtains the influence power fraction of the object.
- 15. device according to claim 10, which is characterized in that the object influences power fraction computing module is further used In：The definite and relevant multiple events of object weigh the time effects power fraction of the plurality of event according to previously given event Average computation is weighted again obtains the influence power fraction of the object.
- 16. device according to claim 15 a, which is characterized in that if event is related to multiple themes, give in advance Fixed event weights are related to one below：The average value of multiple topic weights；The minimum value of multiple topic weights；The maximum of multiple topic weights；The median of multiple topic weights.
- 17. device according to claim 10, which is characterized in that the object influences power fraction computing module is further used In：To polymerize with the relevant event of the object according to theme, be calculated respectively the object under each theme when Between influence power fraction, wherein, the event under same subject is weighted according to time factor averagely obtains the event in the theme Under time effects power fraction；The influence power fraction of the object is calculated according to time effects power fraction of the object under each theme, wherein, often Time effects power fraction under a theme is weighted the shadow that the object averagely, is calculated according to previously given topic weights Ring power fraction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310435049.5A CN104463594B (en) | 2013-09-23 | 2013-09-23 | The method and its device of computing object influence power |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310435049.5A CN104463594B (en) | 2013-09-23 | 2013-09-23 | The method and its device of computing object influence power |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104463594A CN104463594A (en) | 2015-03-25 |
| CN104463594B true CN104463594B (en) | 2018-05-29 |
Family
ID=52909586
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310435049.5A Active CN104463594B (en) | 2013-09-23 | 2013-09-23 | The method and its device of computing object influence power |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104463594B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106611339B (en) * | 2015-10-21 | 2020-12-29 | 创新先进技术有限公司 | Seed user screening method, and product user influence evaluation method and device |
| CN105760618B (en) * | 2016-03-08 | 2019-02-19 | 中国人民解放军总参谋部第五十四研究所 | Target situation methods of exhibiting based on Virtual process geography GIS |
| CN108959249A (en) * | 2018-06-21 | 2018-12-07 | 中国农业科学院农田灌溉研究所 | A kind of evaluation method of periodical best paper |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101251850A (en) * | 2008-01-04 | 2008-08-27 | 杨虡 | Internet topics ranking system and method based on user prestige |
| CN102426610A (en) * | 2012-01-13 | 2012-04-25 | 中国科学院计算技术研究所 | Microblog rank searching method and microblog searching engine |
| CN102945279A (en) * | 2012-11-14 | 2013-02-27 | 清华大学 | Evaluating method and device of influence effect of microblog users |
-
2013
- 2013-09-23 CN CN201310435049.5A patent/CN104463594B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101251850A (en) * | 2008-01-04 | 2008-08-27 | 杨虡 | Internet topics ranking system and method based on user prestige |
| CN102426610A (en) * | 2012-01-13 | 2012-04-25 | 中国科学院计算技术研究所 | Microblog rank searching method and microblog searching engine |
| CN102945279A (en) * | 2012-11-14 | 2013-02-27 | 清华大学 | Evaluating method and device of influence effect of microblog users |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104463594A (en) | 2015-03-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103823908B (en) | Content recommendation method and server based on user preference | |
| CN103678672B (en) | Method for recommending information | |
| CN108121737B (en) | Method, device and system for generating business object attribute identifier | |
| CN109492180A (en) | Resource recommendation method, device, computer equipment and computer readable storage medium | |
| CN107632984A (en) | A kind of cluster data table shows methods, devices and systems | |
| CN106886518A (en) | A kind of method of microblog account classification | |
| CN110941762B (en) | Method for improving sequencing result of personalized recommendation system | |
| CN109658206A (en) | Information recommendation method and device | |
| CN103970850B (en) | Site information recommends method and system | |
| Cho et al. | Implementation of Personalized recommendation System using k-means Clustering of Item Category based on RFM | |
| CN102841929A (en) | Recommending method integrating user and project rating and characteristic factors | |
| CN110335123B (en) | Commodity recommendation method, system, computer readable medium and device based on social e-commerce platform | |
| CN107256513A (en) | Method and device is recommended in a kind of collocation of object | |
| KR20140056731A (en) | Purchase recommendation service system and method | |
| CN104751354B (en) | A kind of advertisement crowd screening technique | |
| CN110197404A (en) | The personalized long-tail Method of Commodity Recommendation and system of popularity deviation can be reduced | |
| CN110727859B (en) | Recommendation information pushing method and device | |
| CN104616178A (en) | Recommendation method of E-commerce goods based on big-data multi-label classification method | |
| CN104463594B (en) | The method and its device of computing object influence power | |
| Tsuji et al. | Book recommendation using machine learning methods based on library loan records and bibliographic information | |
| CN110674178A (en) | Method and system for constructing user portrait label | |
| CN116188120B (en) | Method, device and system for recommending audio books and storage medium | |
| CN107944946A (en) | Commercial goods labels generation method and device | |
| CN106897412A (en) | A kind of method and apparatus for recommending associated application based on intended application | |
| CN109117434A (en) | Judgement document's search method, device, storage medium and processor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |