CN104092567A

CN104092567A - Method and device for confirming influence sequencing of users

Info

Publication number: CN104092567A
Application number: CN201410294986.8A
Authority: CN
Inventors: 陈凯; 周异; 周曲
Original assignee: Huawei Technologies Co Ltd
Current assignee: Changzhou Hengtang Technology Industry Co ltd
Priority date: 2014-06-26
Filing date: 2014-06-26
Publication date: 2014-10-08
Anticipated expiration: 2034-06-26
Also published as: CN104092567B

Abstract

An embodiment of the invention provides a method and a device for confirming influence sequencing of users. The method comprises the steps of obtaining first information and second information, wherein the first information is used for indicating the mutual focusing relation of N users in a social network, and the second information is used for indicating the message number sent by the user in N users and schemes of messages sent by the users; confirming N-dimensional weight vectors corresponding to target schemes according to the first information and the second information; obtaining third information, wherein the third information is used for indicating the number of messages forwarded by the users having message forwarding relation and the schemes of the forwarded messages; confirming transition probability matrixes corresponding to the target schemes according to the third information; utilizing a Pecs sorting algorithm to confirm the influence sequencing of N users in the social network in the target scheme field. The actual relevancy of the users in the social network is reflected, so that the accuracy of an influence sequencing result is effectively improved.

Description

Determine the method and apparatus of user's influence power sequence

Technical field

The present invention relates to computer network field, and relate in particular to the method and apparatus of a kind of definite user's influence power sequence.

Background technology

Along with popularizing of fast wireless network, the network media is being played the part of more and more important role in social information's propagation and commercial field.Such as integrated information portal website and social networks progressively replace the most important platform that traditional media becomes Information Communication, the website that on these network informations and social platform, influence power is larger and user play especially the effect of promotion in the propagation of event.

The network node of material impact from finding relational network intricate and that user is huge (can be for example website, cellphone subscriber, microblogging or micro-credit household, operator's gateway etc.) is the key of product promotion or Information Communication monitoring.Therefore finding out influential network node and carrying out influence power rank is a kind of effective method that enterprise carries out product promotion or Information Communication monitoring.

PageRank (Page sequence) algorithm is the general-purpose algorithm of network node influence power rank, and it is proposed by Google at first.Page sort algorithm has defined random surfer's model, imagine and in network, exist many random surfers, they from certain network node (for example, webpage) set out, according to transition probability matrix migration between network node, so random surfer's position probability distribution has represented the importance of webpage.

Page sort algorithm has Page sequence (Topic Sensitive PageRank) version of prototype version and subject-oriented.Wherein, the Page sort algorithm of subject-oriented can be applicable in social networks, for determining that each user of social networks is in the influence power of a certain target topic.This algorithm adopts following formula iteration to obtain each user's influence power sequence:

p _i+1＝βMp _i+(1-β)e _s/│S│

Wherein, p _i+1and p _ibe illustrated respectively in the i+1 time and during the i time iterative computation, the vector that each user's Page ranking value forms, this vectorial dimension n represents user's to be sorted number; β is the transfer constant between 0 to 1, conventionally value 0.15; M is transition probability matrix, is that the mutual concern relation based between user is set up; E in formula _s/ │ S │ is weight vectors, and wherein, S is the set of the node that target topic is relevant, e _sin vector, the value of each element depends on whether the user that this element is corresponding belongs to S set, if so, gets 1, otherwise, get 0.

But, in social networks, there is following shortcoming in the Page sort algorithm of above-mentioned subject-oriented: the selection of weight vectors is simple with 0 or 1, i.e. relevant or uncorrelated expression, and in reality, between the user relevant from target topic, also have the different of degree of correlation, the expression of above-mentioned weight vector is not accurate enough.In addition, in transition probability matrix, each element is only determined by the concern relation between user.Can not reflect in actual social networks, thereby make the accuracy of influence power ranking results lower.

Summary of the invention

Embodiments of the invention provide the method and apparatus of a kind of definite user's influence power, can effectively improve the accuracy of influence power ranking results.

First aspect, a kind of method is provided, the method comprises: obtain the first information and the second information, the first information is used to indicate the mutual concern relation of N user in social networks, the second information is used to indicate the message count of each user's issue in N user, and the theme of the message of each user's issue; According to the first information and the second information, determine the N dimensional weight vector that target topic is corresponding, i element in weight vectors is determined based on the first proportion and the second proportion, wherein, the first proportion is: in the message of i user's issue, about the shared proportion of the message of target topic, the second proportion is: in the user that i user pays close attention to, the message of issue comprises the shared proportion of user about the message of target topic, and i is the arbitrary integer in 1 to N; Obtain the 3rd information, the 3rd information is used to indicate the quantity of forwarding messages and the theme of forwarding messages between the user in N user with message forwarding relation; According to the 3rd information, determine the transition probability matrix that target topic is corresponding, (the j of transition probability matrix, k) element based on: k user is forwarded by j user, quantity about the message of target topic, the quantity of the forwarding associated user that j user is corresponding, and forward that magnitude relationship between the quantity of the message about target topic of associated user issue determines, wherein, forwarding associated user is in N user, the user that the message of issue was forwarded by j user, j, k are the arbitrary integer in 1 to N; According to transition probability matrix and weight vector, utilize Page sort algorithm to determine in social networks, N user is in the sequence of the influence power in target topic field.

In conjunction with first aspect, in the possible implementation of the first of first aspect, the i element in weight vectors is determined by following formula: A _i=x (a/b)+(1-x) (c/d), wherein, A _iit is the i element in weight vectors; X is predefined real number, and 0≤x≤1; A is the quantity of the message about target topic of i user's issue; B is the message sum of i user's issue; C is that in the user of i user's concern, the message of issue comprises the quantity about the user of the message of target topic; D is the total number of users that i user pays close attention to.

In conjunction with the possible implementation of the first of first aspect or first aspect, in the possible implementation of the second of first aspect, (j, k) element of transition probability matrix is determined by following formula: B _jk=max (e, 1)/max (f, g), wherein, B _jk(j, k) element for transition probability matrix; To be k user forwarded by j user e, about the quantity of the message of target topic; F is for forwarding the quantity of the message about target topic of associated user issue; G is the quantity of forwarding associated user corresponding to j user.

The first or the possible implementation of the second in conjunction with first aspect, first aspect, in the third possible implementation of first aspect, according to transition probability matrix and weight vector, utilize Page sort algorithm to determine the sequence of N user's influence power, comprise: according to transition probability matrix and weight vector, each self-corresponding Page ranking value of N user is carried out to iteration for the first time, and iteration is the predefined limited number of time iteration of iterations for the first time;

According to after iteration for the first time, N each self-corresponding Page ranking value of user, is divided into the first set and the second set by N user, and wherein, Page ranking value corresponding to user in the second set is less than Page ranking value corresponding to user in the first set; According to transition probability matrix and weight vector, each self-corresponding Page ranking value of N user is carried out to iteration for the second time, in iteration for the second time, the iteration stopping threshold value of the Page ranking value that user in the first set is corresponding is less than the iteration stopping threshold value of Page ranking value corresponding to user in the second set; After iteration, the sequence of the size of N each self-corresponding Page ranking value of user is defined as N user in the sequence of the influence power in target topic field for the second time.

Second aspect, provides a kind of method, and the method comprises: each self-corresponding Page ranking value of N network node is carried out to iteration for the first time, and iteration is the predefined limited number of time iteration of iterations for the first time; According to after iteration for the first time, N each self-corresponding Page ranking value of network node, N network node is divided into the first set and the second set, and wherein, Page ranking value corresponding to network node in the second set is less than Page ranking value corresponding to network node in the first set; Each self-corresponding Page ranking value of N network node is carried out to iteration for the second time, in iteration for the second time, the iteration stopping threshold value of the Page ranking value that network node in the first set is corresponding is less than the iteration stopping threshold value of the Page ranking value that network node in the second set is corresponding; After iteration, the sequence of the size of N each self-corresponding Page ranking value of network node is defined as the sequence of the influence power of N network node for the second time.

The third aspect, a kind of device is provided, this device comprises: the first acquisition module, be used for obtaining the first information and the second information, the first information is used to indicate the mutual concern relation of N user in social networks, the second information is used to indicate the message count of each user's issue in N user, and the theme of the message of each user's issue, the first determination module, for the first information and the second information of obtaining according to the first acquisition module, determine the N dimensional weight vector that target topic is corresponding, i element in weight vectors is determined based on the first proportion and the second proportion, wherein, the first proportion is: in the message of i user's issue, about the shared proportion of the message of target topic, the second proportion is: in the user that i user pays close attention to, the message of issue comprises the shared proportion of user about the message of target topic, and i is the arbitrary integer in 1 to N, the second acquisition module, for obtaining the 3rd information, the 3rd information is used to indicate the quantity of forwarding messages and the theme of forwarding messages between the user in N user with message forwarding relation, the second determination module, for the 3rd information of obtaining according to the second acquisition module, determine the transition probability matrix that target topic is corresponding, (the j of transition probability matrix, k) element based on: k user is forwarded by j user, quantity about the message of target topic, the quantity of the forwarding associated user that j user is corresponding, and forward that magnitude relationship between the quantity of the message about target topic of associated user issue determines, wherein, forwarding associated user is in N user, the user that the message of issue was forwarded by j user, j, k is the arbitrary integer in 1 to N, the 3rd determination module, for according to transition probability matrix and weight vector, utilizes Page sort algorithm to determine in social networks, and N user is in the sequence of the influence power in target topic field.

In conjunction with the third aspect, in the possible implementation of the first of the third aspect, the i element in weight vectors is determined by following formula: A _i=x (a/b)+(1-x) (c/d), wherein, A _iit is the i element in weight vectors; X is predefined real number, and 0≤x≤1; A is the quantity of the message about target topic of i user's issue; B is the message sum of i user's issue; C is that in the user of i user's concern, the message of issue comprises the quantity about the user of the message of target topic; D is the total number of users that i user pays close attention to.

In conjunction with the possible implementation of the first of the third aspect or the third aspect, in the possible implementation of the second of the third aspect, (j, k) element of transition probability matrix is determined by following formula: B _jk=max (e, 1)/max (f, g), wherein, B _jk(j, k) element for transition probability matrix; To be k user forwarded by j user e, about the quantity of the message of target topic; F is for forwarding the quantity of the message about target topic of associated user issue; G is the quantity of forwarding associated user corresponding to j user.

The first or the possible implementation of the second in conjunction with the third aspect, the third aspect, in the third possible implementation of the third aspect, the 3rd determining unit is specifically for according to transition probability matrix and weight vector, each self-corresponding Page ranking value of N user is carried out to iteration for the first time, and iteration is the predefined limited number of time iteration of iterations for the first time; According to after iteration for the first time, N each self-corresponding Page ranking value of user, N user is divided into the first set and the second set, and wherein, Page ranking value corresponding to user in the second set is less than the iteration stopping threshold value of Page ranking value corresponding to user in the first set; According to transition probability matrix and weight vector, each self-corresponding Page ranking value of N user is carried out to iteration for the second time, in iteration for the second time, the iteration stopping threshold value of the Page ranking value that user in the first set is corresponding is less than the iteration stopping threshold value of Page ranking value corresponding to user in the second set; After iteration, the sequence of the size of N each self-corresponding Page ranking value of user is defined as N user in the sequence of the influence power in target topic field for the second time.

Fourth aspect, provides a kind of device, and this device comprises: the first iteration module, each self-corresponding Page ranking value of N network node is carried out to iteration for the first time, and iteration is the predefined limited number of time iteration of iterations for the first time; Determination module, for each the self-corresponding Page ranking value of N network node obtaining according to the first iteration module, N network node is divided into the first set and the second set, wherein, Page ranking value corresponding to network node in the second set is less than Page ranking value corresponding to network node in the first set; Secondary iteration module, for each self-corresponding Page ranking value of N network node that the first iteration module is obtained, carry out iteration for the second time, in iteration for the second time, the iteration stopping threshold value of the Page ranking value that network node in the first set that determination module is determined is corresponding is less than the iteration stopping threshold value of the Page ranking value that network node in the second set is corresponding; Determination module is also for will be for the second time after iteration, the sort sequence of the influence power that is defined as N network node of the size of N each self-corresponding Page ranking value of network node.

Embodiments of the invention have adopted definite transition probability matrix and the weight vectors of social networks (for example forwarding relation, concern relation and target topic) based on actual between user to carry out influence power sequence.Because actual social networks between user has reflected in social networks the degree of correlation actual between user, therefore effectively improved the accuracy of influence power ranking results.

Accompanying drawing explanation

In order to be illustrated more clearly in the technical scheme of the embodiment of the present invention, to the accompanying drawing of required use in the embodiment of the present invention be briefly described below, apparently, below described accompanying drawing be only some embodiments of the present invention, for those of ordinary skills, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.

Fig. 1 is according to the indicative flowchart of the method for definite user's of the embodiment of the present invention influence power.

Fig. 2 is the schematic diagram of device of determining according to another embodiment of the present invention user's influence power.

Fig. 3 is the indicative flowchart of method of determining according to another embodiment of the present invention user's influence power.

Fig. 4 is according to the schematic diagram of the device of definite user's of further embodiment of this invention influence power.

Fig. 5 is the indicative flowchart of method of determining according to another embodiment of the present invention the influence power of network node.

Fig. 6 is the schematic diagram of device of determining according to another embodiment of the present invention the influence power of network node.

Fig. 7 is the indicative flowchart of method of determining according to another embodiment of the present invention the influence power of network node.

Fig. 8 is the schematic diagram of device of determining according to another embodiment of the present invention the influence power of network node.

Embodiment

Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is a part of embodiment of the present invention, rather than whole embodiment.Embodiment based in the present invention, the every other embodiment that those of ordinary skills obtain under the prerequisite of not making creative work, should belong to the scope of protection of the invention.

Fig. 1 is according to the indicative flowchart of the method for definite user's of the embodiment of the present invention influence power.The method of Fig. 1 is carried out by the device of determining user's influence power.The method of Fig. 1 comprises:

110, obtain the first information and the second information, the first information is used to indicate the mutual concern relation of N user in social networks, and the second information is used to indicate the message count of each user's issue in N user, and the theme of the message of each user's issue;

120, according to the first information and the second information, determine the N dimensional weight vector that target topic is corresponding, i element in weight vectors is determined based on the first proportion and the second proportion, wherein, the first proportion is: in the message of i user's issue, about the shared proportion of the message of target topic, the second proportion is: in the user that i user pays close attention to, the message of issue comprises the shared proportion of user about the message of target topic, and i is the arbitrary integer in 1 to N;

130, obtain the 3rd information, the 3rd information is used to indicate the quantity of forwarding messages and the theme of forwarding messages between the user in N user with message forwarding relation;

140, according to the 3rd information, determine the transition probability matrix that target topic is corresponding, (the j of transition probability matrix, k) element based on: k user is forwarded by j user, quantity about the message of target topic, the quantity of the forwarding associated user that j user is corresponding, and forward that magnitude relationship between the quantity of the message about target topic of associated user issue determines, wherein, forwarding associated user is in N user, the user that the message of issue was forwarded by j user, j, k are the arbitrary integer in 1 to N;

150, according to transition probability matrix and weight vector, utilize Page sort algorithm to determine in social networks, N user is in the sequence of the influence power in target topic field.

Second information that should be understood that can also be indicated the message sum of each user's issue.The theme of the message of each user issue can be such as one or more in society, education, economy, military affairs, amusement etc.

According to embodiments of the invention, the i element in weight vectors is determined by following formula: A _i=x (a/b)+(1-x) (c/d), wherein, A _iit is the i element in weight vectors; X is predefined real number, and 0≤x≤1; A is the quantity of the message about target topic of i user's issue; B is the message sum of i user's issue; C is that in the user of i user's concern, the message of issue comprises the quantity about the user of the message of target topic; D is the total number of users that i user pays close attention to.

Should be understood that the element of determining in weight vectors also can adopt other the formula of determining based on a/b and c/d.For example simply be out of shape.

According to embodiments of the invention, (j, k) element of transition probability matrix is determined by following formula: B _jk=max (e, 1)/max (f, g), wherein, B _jk(j, k) element for transition probability matrix; To be k user forwarded by j user e, about the quantity of the message of target topic; F is for forwarding the quantity of the message about target topic of associated user issue; G is the quantity of forwarding associated user corresponding to j user.

Should be understood that the element of determining in transition probability matrix also can adopt other the formula of determining based on e, f and g.

According to embodiments of the invention, according to transition probability matrix and weight vector, utilize Page sort algorithm to determine the sequence of N user's influence power, comprise: according to transition probability matrix and weight vector, each self-corresponding Page ranking value of N user is carried out to iteration for the first time, and iteration is the predefined limited number of time iteration of iterations for the first time; According to after iteration for the first time, N each self-corresponding Page ranking value of user, is divided into the first set and the second set by N user, and wherein, Page ranking value corresponding to user in the second set is less than Page ranking value corresponding to user in the first set; According to transition probability matrix and weight vector, each self-corresponding Page ranking value of N user is carried out to iteration for the second time, in iteration for the second time, the iteration stopping threshold value of the Page ranking value that user in the first set is corresponding is less than the iteration stopping threshold value of Page ranking value corresponding to user in the second set; After iteration, the sequence of the size of N each self-corresponding Page ranking value of user is defined as N user in the sequence of the influence power in target topic field for the second time.

Should be understood that the predefined limited number of time iteration of iterations, can be according to the iterations of Six Degrees theory setting, for example 5 times.Also can be other iterations, for example, be less than 10 times.In other words, iteration is the predefined limited number of time iteration of iterations for the first time, so that according to the ranking value of primary iteration result, carry out the preliminary sequence of influence power, and as required the user of the first set is sorted more accurately.The method of Fig. 1 can also comprise: in iteration for the first time, N user's Page ranking value is set to convergence outage threshold.According to after iteration for the first time, N each self-corresponding Page ranking value of user, N user is divided into the first set and the second set, can be, according to power-law distribution rule, determine two set, for example, for the first time in the Page ranking value after iteration, according to size order, front 20% user is the first set, and rear 80% user is the second set.Also can divide as required different set.

Because the convergence outage threshold that the first set is set is less than the convergence outage threshold that the second set is set, therefore when influence power sort, the computing that the second set has been carried out still less time, thus reduced the expense of calculating.

In other words, social network data has worldlet and large data attribute, and the influence power ranking value of user node is power-law distribution, and the influence power value of most nodes is all very low, only has the influence power of small part node very high.In actual node influence power rank application, do not need all nodes to carry out rank, only need carry out rank to the large node of a small amount of influence power.The little node of most of influence power, after recursive calculation 5 times, just can be ignored the variation of these nodes, no longer needs calculating, thereby saves computing time in a large number.

Described above according to the method for definite user's of the embodiment of the present invention influence power, below in conjunction with accompanying drawing, described according to the device of definite user's of the embodiment of the present invention influence power.

Fig. 2 is the schematic diagram of device of determining according to another embodiment of the present invention user's influence power.The device 200 of Fig. 2 comprises: the first acquisition module 210, be used for obtaining the first information and the second information, the first information is used to indicate the mutual concern relation of N user in social networks, the second information is used to indicate the message count of each user's issue in N user, and the theme of the message of each user's issue, the first determination module 220, for the first information and the second information of obtaining according to the first acquisition module 210, determine the N dimensional weight vector that target topic is corresponding, i element in weight vectors is determined based on the first proportion and the second proportion, wherein, the first proportion is: in the message of i user's issue, about the shared proportion of the message of target topic, the second proportion is: in the user that i user pays close attention to, the message of issue comprises the shared proportion of user about the message of target topic, and i is the arbitrary integer in 1 to N, the second acquisition module 230, for obtaining the 3rd information, the 3rd information is used to indicate the quantity of forwarding messages and the theme of forwarding messages between the user in N user with message forwarding relation, the second determination module 240, for the 3rd information of obtaining according to the second acquisition module 230, determine the transition probability matrix that target topic is corresponding, (the j of transition probability matrix, k) element based on: k user is forwarded by j user, quantity about the message of target topic, the quantity of the forwarding associated user that j user is corresponding, and forward that magnitude relationship between the quantity of the message about target topic of associated user issue determines, wherein, forwarding associated user is in N user, the user that the message of issue was forwarded by j user, j, k is the arbitrary integer in 1 to N, the 3rd determination module 250, for according to transition probability matrix and weight vector, utilizes Page sort algorithm to determine in social networks, and N user is in the sequence of the influence power in target topic field.

According to embodiments of the invention, the 3rd determining unit, specifically for according to transition probability matrix and weight vector, is carried out iteration for the first time to each self-corresponding Page ranking value of N user, and iteration is the predefined limited number of time iteration of iterations for the first time; According to after iteration for the first time, N each self-corresponding Page ranking value of user, N user is divided into the first set and the second set, and wherein, Page ranking value corresponding to user in the second set is less than the iteration stopping threshold value of Page ranking value corresponding to user in the first set; According to transition probability matrix and weight vector, each self-corresponding Page ranking value of N user is carried out to iteration for the second time, in iteration for the second time, the iteration stopping threshold value of the Page ranking value that user in the first set is corresponding is less than the iteration stopping threshold value of Page ranking value corresponding to user in the second set; After iteration, the sequence of the size of N each self-corresponding Page ranking value of user is defined as N user in the sequence of the influence power in target topic field for the second time.

Operation and the function of modules of device 200 of determining user's influence power can, with reference to the method for above-mentioned Fig. 1, for fear of repetition, not repeat them here.Below in conjunction with object lesson, embodiments of the invention are described in further detail.

Fig. 3 is the indicative flowchart of method of determining according to another embodiment of the present invention user's influence power.The method of Fig. 3 is the example of the method for Fig. 1.User in the embodiment of the user's corresponding diagram 1 in the set Z in the present embodiment second set, does not belong to the user in the first set in the embodiment that gathers the user's corresponding diagram 1 in Z in set Y; The iteration stopping threshold value of the Page ranking value that user in corresponding the first set of first threshold is corresponding, the iteration stopping threshold value of the Page ranking value that user in corresponding the first set of Second Threshold is corresponding.The first ranking value is for the initialization of ranking value vector, and wherein the first ranking value of each user is the element of ranking value vector; N the vector that each self-corresponding Page ranking value of user forms after the corresponding iteration for the first time of the second ranking value vector.Concrete implementation is as follows.

310, set iterations.

For example theoretical based on Six Degrees, iterations is set as to 5, in other words, node says that the transmission of influence power is no more than at most 5 times and just can arrives all nodes.

320, the Page ranking value assignment to all users.

Particularly, all users are set to the first identical ranking value, for example will have user's assignment is 1 more.

330, set first threshold.

For the fluctuation ratio with user's influence power ranking value, the fluctuation of influence power ranking value is defined as R=│ p to this first threshold _k-p _k+1│/p _k.

340, according to the iterations of setting, calculate, and according to first threshold determining section user's influence power ranking value.

For example, can adopt formula 3.1:p _k+1=β Bp _k+ (1-β) A (3.1)

Wherein, B is the transition probability matrix of social class theme, and this matrix is N * N matrix; A is the weight vectors of social class theme, and this vector is a corresponding N user's N dimensional vector; β is constant, for example, can be 0.15; p _kand p _k+1the ranking value that is respectively user is through the ranking value vector after the k time and the k+1 time iteration, and this vector is a corresponding N user's N dimensional vector.I element in weight vectors A is determined by following formula: A _i=x (a/b)+(1-x) (c/d), wherein, A _iit is the i element in weight vectors; X is predefined real number, and 0≤x≤1; A is the quantity of the message about target topic of i user's issue; B is the message sum of i user's issue; C is that in the user of i user's concern, the message of issue comprises the quantity about the user of the message of target topic; D is the total number of users that i user pays close attention to.(j, k) element of transition probability matrix B is determined by following formula: B _jk=max (e, 1)/max (f, g), wherein, B _jk(j, k) element for transition probability matrix; To be k user forwarded by j user e, about the quantity of the message of target topic; F is for forwarding the quantity of the message about target topic of associated user issue; G is the quantity of forwarding associated user corresponding to j user.

Particularly, utilize the weight vectors A of transition probability matrix B and theme to carry out iterative computation to the first ranking value vector of user, and after each iterative computation, calculate the undulating value R of each user's ranking value.By this undulating value R and first threshold comparison: if the undulating value in 5 iterative computation is not more than first threshold, stop the calculating to this user.This user's ranking value is defined as to influence power ranking value simultaneously, and this user is defined as gathering the element in X.If after 5 iterative computation, the undulating value of user's ranking value is greater than first threshold, this user is defined as gathering element in Y, and carries out 350.

350, determine the second ranking value vector of user.

Particularly, by a plurality of the first ranking value vectors to transition probability matrix and user, carry out iterative computation 5 times, determine the second ranking value vector of user, the second ranking value that wherein element in the second ranking value vector is each user.

360, determine the set Z of low influence power user in set Y.

Particularly, according to the size of the second ranking value of each user in set Y, tentatively sort, and according to ranking results, determine low influence power user's set Z.For example, according to power-law distribution, the user who sorts rear 80% is defined as to low influence power user.

For not belonging to the user in set Z in set Y, similar to 340, according to first threshold and formula 3.1 definite these users' influence power ranking value.User in pair set Z carries out 370.

370, for the user in set Z, set Second Threshold.

Particularly, set the Second Threshold that is greater than first threshold.Because the user who sorts rear 80% is low influence power user, therefore make Second Threshold be greater than the accuracy that first threshold can not reduce final influence power sequence substantially.

380, according to Second Threshold, determine the influence power ranking value of user in set Z.

For example, according to formula 3.1, utilize the weight vectors A of transition probability matrix B and theme to carry out iterative computation to the second ranking value vector of user, and after each iterative computation the undulating value of each user's ranking value in set of computations Z.By this undulating value and Second Threshold comparison: if this undulating value is not more than Second Threshold, stop the calculating to this user.This user's ranking value is defined as gathering the influence power ranking value of user in Z simultaneously.

390, according to all users' influence power ranking value, determine all users' influence power.

Particularly, because the large user's of influence power ranking value influence power is high, user can be sorted from big to small and determines all users' influence power according to influence power ranking value.

Fig. 4 is the indicative flowchart of device of determining according to another embodiment of the present invention user's influence power.The device 400 of Fig. 4 comprises: processor 410, memory 420, communication bus 430.Processor 410 calls the code in memory 420 by communication bus 430, and for: obtain the first information and the second information, the first information is used to indicate the mutual concern relation of N user in social networks, the second information is used to indicate the message count of each user's issue in N user, and the theme of the message of each user's issue; According to the first information and the second information, determine the N dimensional weight vector that target topic is corresponding, i element in weight vectors is determined based on the first proportion and the second proportion, wherein, the first proportion is: in the message of i user's issue, about the shared proportion of the message of target topic, the second proportion is: in the user that i user pays close attention to, the message of issue comprises the shared proportion of user about the message of target topic, and i is the arbitrary integer in 1 to N; Obtain the 3rd information, the 3rd information is used to indicate the quantity of forwarding messages and the theme of forwarding messages between the user in N user with message forwarding relation; According to the 3rd information, determine the transition probability matrix that target topic is corresponding, (the j of transition probability matrix, k) element based on: k user is forwarded by j user, quantity about the message of target topic, the quantity of the forwarding associated user that j user is corresponding, and forward that magnitude relationship between the quantity of the message about target topic of associated user issue determines, wherein, forwarding associated user is in N user, the user that the message of issue was forwarded by j user, j, k are the arbitrary integer in 1 to N; According to transition probability matrix and weight vector, utilize Page sort algorithm to determine in social networks, N user is in the sequence of the influence power in target topic field.

According to embodiments of the invention, processor 410 specifically for: according to transition probability matrix and weight vector, each self-corresponding Page ranking value of N user is carried out to iteration for the first time, iteration is the predefined limited number of time iteration of iterations for the first time; According to after iteration for the first time, N each self-corresponding Page ranking value of user, is divided into the first set and the second set by N user, and wherein, Page ranking value corresponding to user in the second set is less than Page ranking value corresponding to user in the first set; According to transition probability matrix and weight vector, each self-corresponding Page ranking value of N user is carried out to iteration for the second time, in iteration for the second time, the iteration stopping threshold value of the Page ranking value that user in the first set is corresponding is less than the iteration stopping threshold value of Page ranking value corresponding to user in the second set; After iteration, the sequence of the size of N each self-corresponding Page ranking value of user is defined as N user in the sequence of the influence power in target topic field for the second time.

Operation and the function of modules of device 400 of determining user's influence power can, with reference to the method for above-mentioned Fig. 1, for fear of repetition, not repeat them here.

In addition, the Page sort method that prior art adopts or the Page sort method of subject-oriented carry out the influence power sequence of network node, utilize random surfer's model, and the influence power that the threshold value that makes ranking value converge to setting by iterative computation is carried out sort.Yet because the threshold value that adopts iterative computation to make ranking value converge to setting need to repeatedly be calculated, when network node number is too much, computing cost is large.The method of the influence power of definite network node of another embodiment of the present invention, can, when network node number is too much, reduce the expense of calculating.

Fig. 5 is the indicative flowchart of method of determining according to another embodiment of the present invention the influence power of network node.The method of Fig. 5 is carried out by the device of determining the influence power of network node.Comprise:

510, each self-corresponding Page ranking value of N network node is carried out to iteration for the first time, iteration is the predefined limited number of time iteration of iterations for the first time;

520, according to after iteration for the first time, N each self-corresponding Page ranking value of network node, N network node is divided into the first set and the second set, and wherein, Page ranking value corresponding to network node in the second set is less than Page ranking value corresponding to network node in the first set;

530, each self-corresponding Page ranking value of N network node is carried out to iteration for the second time, in iteration for the second time, the iteration stopping threshold value of the Page ranking value that network node in the first set is corresponding is less than the iteration stopping threshold value of the Page ranking value that network node in the second set is corresponding;

540, will be for the second time after iteration, the sort sequence of the influence power that is defined as N network node of the size of N each self-corresponding Page ranking value of network node.

Should be understood that network node can be website, cellphone subscriber, microblogging or micro-credit household, operator's gateway etc.The predefined limited number of time iteration of iterations, can be according to the iterations of Six Degrees theory setting, for example 5 times.Also can be other iterations, for example, be less than 10 times.In other words, iteration is the predefined limited number of time iteration of iterations for the first time, so that according to the ranking value of primary iteration result, carry out the preliminary sequence of influence power, and as required the network node of the first set is sorted more accurately.The method of Fig. 1 can also comprise: in iteration for the first time, the Page ranking value of N network node is set to convergence outage threshold.According to after iteration for the first time, N each self-corresponding Page ranking value of network node, N network node is divided into the first set and the second set, can be, according to power-law distribution rule, determine two set, for example, for the first time in the Page ranking value after iteration, according to size order, front 20% network node is the first set, and rear 80% network node is the second set.Also can divide as required different set.

Embodiments of the invention can be set different convergence outage thresholds for different network nodes as required, owing to can carrying out the still less iterative computation of number of times for the little network node of convergence outage threshold, have therefore reduced the expense of calculating.

Described above according to the method for the influence power of definite network node of the embodiment of the present invention, below in conjunction with accompanying drawing, described according to the device of the influence power of definite network node of the embodiment of the present invention.

Fig. 6 is the schematic diagram of device of determining according to another embodiment of the present invention the influence power of network node.The device 600 of Fig. 6 comprises: the first iteration module 610, and for each self-corresponding Page ranking value of N network node is carried out to iteration for the first time, iteration is the predefined limited number of time iteration of iterations for the first time; Determination module 620, for each the self-corresponding Page ranking value of N network node obtaining according to the first iteration module, N network node is divided into the first set and the second set, wherein, Page ranking value corresponding to network node in the second set is less than Page ranking value corresponding to network node in the first set; Secondary iteration module 630, for each self-corresponding Page ranking value of N network node that the first iteration module is obtained, carry out iteration for the second time, in iteration for the second time, the iteration stopping threshold value of the Page ranking value that network node in the first set that determination module is determined is corresponding is less than the iteration stopping threshold value of the Page ranking value that network node in the second set is corresponding; Determination module 620 is also for will be for the second time after iteration, the sort sequence of the influence power that is defined as N network node of the size of N each self-corresponding Page ranking value of network node.

Operation and the function of modules of device 600 of determining the influence power of network node can, with reference to the method for above-mentioned Fig. 5, for fear of repetition, not repeat them here.Below in conjunction with object lesson, embodiments of the invention are described in further detail.

Fig. 7 is the indicative flowchart of method of determining according to another embodiment of the present invention the influence power of network node.The method of Fig. 7 is the example of the method for Fig. 5.Network node in the embodiment of the network node corresponding diagram 5 in the set Z in the present embodiment second set, does not belong to the network node in the first set in the embodiment that gathers the network node corresponding diagram 5 in Z in set Y; The iteration stopping threshold value of the Page ranking value that network node in corresponding the first set of first threshold is corresponding, the iteration stopping threshold value of the Page ranking value that network node in corresponding the first set of Second Threshold is corresponding.The first ranking value is for the initialization of ranking value vector, and wherein the first ranking value of each network node is the element of ranking value vector; N the vector that each self-corresponding Page ranking value of network node forms after the corresponding iteration for the first time of the second ranking value vector.Concrete implementation is as follows.

710, set iterations.

For example, take social networks as example, according to Six Degrees theory, iterations is set as to 5, in other words, node says that the transmission of influence power is no more than at most 5 times and just can arrives all nodes.

720, the Page ranking value assignment to all-network node.

Particularly, all network nodes are set to the first identical ranking value, for example will have network node assignment is 1 more.

730, set first threshold.

For the fluctuation ratio with the influence power ranking value of network node, the fluctuation of influence power ranking value is defined as R=│ p to this first threshold _k-p _k+1│/p _k.

740, according to the influence power ranking value of first threshold determining section network node.

For example, adopt " model of levying a tax " formula 7.1:p _k+1=β Mp _k+ (1-β) e/S (7.1)

Wherein, M is transition probability matrix, and this matrix is N * N matrix; E is weight vectors, and this vector is the N dimensional vector of a corresponding N network node; β is constant, for example, can be 0.15; p _kand p _k+1the ranking value that is respectively network node is through the ranking value vector after the k time and the k+1 time iteration, and this vector is the N dimensional vector of a corresponding N network node.S is the number of network node relevant to target topic in N network node.Element in weight vectors e depends on that whether this node is relevant to this target topic, is 1, otherwise is 0 if relevant.Particularly, utilize transition probability matrix M and weight vectors e to carry out iterative computation to the first ranking value vector of network node, and after each iterative computation, calculate the undulating value R of the ranking value of each network node.By this undulating value R and first threshold comparison: if the undulating value in 5 iterative computation is not more than first threshold, stop the calculating to this network node.The ranking value of this network node is defined as to influence power ranking value simultaneously, and this network node is defined as gathering the element in X.If after 5 iterative computation, the undulating value of the ranking value of network node is greater than first threshold, this network node is defined as gathering element in Y, and carries out 750.

750, determine the second ranking value vector of network node.

Particularly, by a plurality of the first ranking value vectors to transition probability matrix and network node, carry out iterative computation 5 times, determine the second ranking value vector of network node, the second ranking value that wherein element in the second ranking value vector is each network node.

760, determine the set Z of low influence power network node in set Y.

Particularly, according to the size of the second ranking value of each network node in set Y, tentatively sort, and according to ranking results, determine the set Z of low influence power network node.For example, according to power-law distribution, the network node sorting rear 80% is defined as to low influence power network node.

For not belonging to the network node in set Z in set Y, similar to 740, according to the influence power ranking value of first threshold and formula 7.1 definite these network nodes.Network node in pair set Z carries out 770.

770, for the network node in set Z, set Second Threshold.

Particularly, set the Second Threshold that is greater than first threshold.Because the network node sorting rear 80% is low influence power network node, therefore make Second Threshold be greater than the accuracy that first threshold can not reduce final influence power sequence substantially.

780, according to Second Threshold, determine the influence power ranking value of network node in set Z.

For example, according to formula 7.1, utilize transition probability matrix M and weight vectors e to carry out iterative computation to the second ranking value vector of network node, and after each iterative computation the undulating value of the ranking value of each network node in set of computations Z.By this undulating value and Second Threshold comparison: if this undulating value is not more than Second Threshold, stop the calculating to this network node.The ranking value of this network node is defined as gathering the influence power ranking value of network node in Z simultaneously.

790, according to the influence power ranking value of all-network node, determine the influence power of all-network node.

Particularly, because the influence power of the large network node of influence power ranking value is high, network node can be sorted from big to small and determines the influence power of all-network node according to influence power ranking value.

The device 800 of Fig. 8 comprises: processor 810, memory 820, communication bus 830.Processor 810 calls the code in memory 820 by communication bus 830, and for: each self-corresponding Page ranking value of N network node is carried out to iteration for the first time, and iteration is the predefined limited number of time iteration of iterations for the first time; According to after iteration for the first time, N each self-corresponding Page ranking value of network node, N network node is divided into the first set and the second set, and wherein, Page ranking value corresponding to network node in the second set is less than Page ranking value corresponding to network node in the first set; Each self-corresponding Page ranking value of N network node is carried out to iteration for the second time, in iteration for the second time, the iteration stopping threshold value of the Page ranking value that network node in the first set is corresponding is less than the iteration stopping threshold value of the Page ranking value that network node in the second set is corresponding; After iteration, the sequence of the size of N each self-corresponding Page ranking value of network node is defined as the sequence of the influence power of N network node for the second time.

Operation and the function of modules of device 800 of determining the influence power of network node can, with reference to the method for above-mentioned Fig. 5, for fear of repetition, not repeat them here.

In addition, term " system " and " network " are often used interchangeably in this article herein.Term "and/or", is only a kind of incidence relation of describing affiliated partner herein, and expression can exist three kinds of relations, and for example, A and/or B, can represent: individualism A exists A and B, these three kinds of situations of individualism B simultaneously.In addition, character "/", generally represents that forward-backward correlation is to liking a kind of relation of "or" herein.

Should be understood that in embodiments of the present invention, " with the corresponding B of A " represents that B is associated with A, according to A, can determine B.But should also be understood that and according to A, determine B and only do not mean that and determine B according to A, can also determine B according to A and/or out of Memory.

Those of ordinary skills can recognize, unit and the algorithm steps of each example of describing in conjunction with embodiment disclosed herein, can realize with electronic hardware, computer software or the combination of the two, for the interchangeability of hardware and software is clearly described, composition and the step of each example described according to function in the above description in general manner.These functions are carried out with hardware or software mode actually, depend on application-specific and the design constraint of technical scheme.Professional and technical personnel can specifically should be used for realizing described function with distinct methods to each, but this realization should not thought and exceeds scope of the present invention.

Those skilled in the art can be well understood to, and with succinct, the specific works process of the system of foregoing description, device and unit, can, with reference to the corresponding process in preceding method embodiment, not repeat them here for convenience of description.

In the several embodiment that provide in the application, should be understood that disclosed system, apparatus and method can realize by another way.For example, device embodiment described above is only schematic, for example, the division of described unit, be only that a kind of logic function is divided, during actual realization, can have other dividing mode, for example a plurality of unit or assembly can in conjunction with or can be integrated into another system, or some features can ignore, or do not carry out.In addition, shown or discussed coupling each other or direct-coupling or communication connection can be indirect coupling or the communication connections by some interfaces, device or unit, can be also electric, machinery or other form connect.

The described unit as separating component explanation can or can not be also physically to separate, and the parts that show as unit can be or can not be also physical locations, can be positioned at a place, or also can be distributed in a plurality of network element.Can select according to the actual needs some or all of unit wherein to realize the object of embodiment of the present invention scheme.

In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, can be also that the independent physics of unit exists, and can be also that two or more unit are integrated in a unit.Above-mentioned integrated unit both can adopt the form of hardware to realize, and also can adopt the form of SFU software functional unit to realize.

Through the above description of the embodiments, those skilled in the art can be well understood to the present invention and can realize with hardware, or firmware realization, or their compound mode realizes.When using software to realize, one or more instructions or the code that above-mentioned functions can be stored in computer-readable medium or on computer-readable medium transmit.Computer-readable medium comprises computer-readable storage medium and communication media, and wherein communication media comprises any medium of being convenient to transmit from a place to another place computer program.Storage medium can be any usable medium that computer can access.As example but be not limited to: computer-readable medium can comprise RAM, ROM, EEPROM, CD-ROM or other optical disc storage, magnetic disk storage medium or other magnetic storage apparatus or can be used in carry or store the expectation with instruction or data structure form program code and can be by any other medium of computer access.In addition.Any connection can be suitable become computer-readable medium.For example, if software be use coaxial cable, optical fiber cable, twisted-pair feeder, Digital Subscriber Line (DSL) or the wireless technology such as infrared ray, radio and microwave from website, server or the transmission of other remote source, so coaxial cable, optical fiber cable, twisted-pair feeder, DSL or the wireless technology such as infrared ray, wireless and microwave be included under in the photographic fixing of medium.As used in the present invention, dish (Disk) and dish (disc) comprise compression laser disc (CD), laser dish, laser disc, digital universal laser disc (DVD), floppy disk and Blu-ray Disc, the copy data of the common magnetic of its mid-game, dish carrys out the copy data of optics with laser.Within combination above also should be included in the protection range of computer-readable medium.

In a word, the foregoing is only the preferred embodiment of technical solution of the present invention, be not intended to limit protection scope of the present invention.Within the spirit and principles in the present invention all, any modification of doing, be equal to replacement, improvement etc., within all should being included in protection scope of the present invention.

Claims

1. a method for definite user's influence power sequence, is characterized in that, comprising:

Obtain the first information and the second information, the described first information is used to indicate the mutual concern relation of N user in social networks, described the second information is used to indicate the message count of each user's issue in a described N user, and the theme of the message of described each user's issue;

According to the described first information and described the second information, determine the N dimensional weight vector that described target topic is corresponding, i element in described weight vectors is determined based on the first proportion and the second proportion, wherein, described the first proportion is: in the message of i user's issue, about the shared proportion of the message of described target topic, described the second proportion is: in the user that described i user pays close attention to, the message of issue comprises the shared proportion of user about the message of described target topic, and i is the arbitrary integer in 1 to N;

Obtain the 3rd information, described the 3rd information is used to indicate the quantity of forwarding messages and the theme of described forwarding messages between the user in a described N user with message forwarding relation;

According to described the 3rd information, determine the transition probability matrix that described target topic is corresponding, (the j of described transition probability matrix, k) element based on: k user is forwarded by j user, quantity about the message of described target topic, the quantity of the forwarding associated user that described j user is corresponding, and magnitude relationship between the quantity of the message about described target topic of described forwarding associated user issue is determined, wherein, described forwarding associated user is in a described N user, the user that the message of issue was forwarded by described j user, j, k is the arbitrary integer in 1 to N,

According to described transition probability matrix and described weight vector, utilize Page sort algorithm to determine in described social networks, a described N user is in the sequence of the influence power in described target topic field.

2. method according to claim 1, is characterized in that, the i element in described weight vectors is determined by following formula:

A _i＝x(a/b)+(1-x)(c/d)

Wherein, A _iit is the i element in described weight vectors; X is predefined real number, and 0≤x≤1; A is the quantity of the message about described target topic of i user's issue; B is the message sum of described i user's issue; C is that in the user of described i user's concern, the message of issue comprises the quantity about the user of the message of described target topic; D is the total number of users that described i user pays close attention to.

3. method according to claim 1 and 2, is characterized in that, (j, k) element of described transition probability matrix is determined by following formula:

B _jk＝max(e,1)/max(f,g)

Wherein, B _jk(j, k) element for described transition probability matrix; To be k user forwarded by j user e, about the quantity of the message of described target topic; F is the quantity of the message about described target topic of described forwarding associated user issue; G is the quantity of forwarding associated user corresponding to described j user.

4. according to the method described in any one in claim 1-3, it is characterized in that, described according to described transition probability matrix and described weight vector, utilize Page sort algorithm to determine the sequence of a described N user's influence power, comprising:

According to described transition probability matrix and described weight vector, each self-corresponding Page ranking value of a described N user is carried out to iteration for the first time, described iteration is for the first time the predefined limited number of time iteration of iterations;

According to after described iteration for the first time, each self-corresponding Page ranking value of a described N user, a described N user is divided into the first set and the second set, and wherein, Page ranking value corresponding to user in described the second set is less than Page ranking value corresponding to user in described the first set;

According to described transition probability matrix and described weight vector, each self-corresponding Page ranking value of a described N user is carried out to iteration for the second time, in described iteration for the second time, the iteration stopping threshold value of Page ranking value corresponding to user in described the first set is less than the iteration stopping threshold value of Page ranking value corresponding to user in described the second set;

By after described iteration for the second time, the sequence of the size of each self-corresponding Page ranking value of a described N user is defined as a described N user in the sequence of the influence power in described target topic field.

5. a method for the influence power of definite network node sequence, is characterized in that, comprising:

Each self-corresponding Page ranking value of N network node is carried out to iteration for the first time, and described iteration is for the first time the predefined limited number of time iteration of iterations;

According to after described iteration for the first time, each self-corresponding Page ranking value of a described N network node, a described N network node is divided into the first set and the second set, wherein, Page ranking value corresponding to network node in described the second set is less than Page ranking value corresponding to network node in described the first set;

Each self-corresponding Page ranking value of a described N network node is carried out to iteration for the second time, in described iteration for the second time, the iteration stopping threshold value of Page ranking value corresponding to network node in described the first set is less than the iteration stopping threshold value of the Page ranking value that network node in described the second set is corresponding;

By after described iteration for the second time, the sequence of the size of each self-corresponding Page ranking value of a described N network node is defined as the sequence of the influence power of a described N network node.

6. a device for definite user's influence power sequence, is characterized in that, comprising:

The first acquisition module, be used for obtaining the first information and the second information, the described first information is used to indicate the mutual concern relation of N user in social networks, and described the second information is used to indicate the message count of each user's issue in a described N user, and the theme of the message of described each user's issue;

The first determination module, for the described first information and described the second information of obtaining according to described the first acquisition module, determine the N dimensional weight vector that described target topic is corresponding, i element in described weight vectors is determined based on the first proportion and the second proportion, wherein, described the first proportion is: in the message of i user's issue, about the shared proportion of the message of described target topic, described the second proportion is: in the user that described i user pays close attention to, the message of issue comprises the shared proportion of user about the message of described target topic, i is the arbitrary integer in 1 to N,

The second acquisition module, for obtaining the 3rd information, described the 3rd information is used to indicate the quantity of forwarding messages and the theme of described forwarding messages between the user in a described N user with message forwarding relation;

The second determination module, for described the 3rd information of obtaining according to described the second acquisition module, determine the transition probability matrix that described target topic is corresponding, (the j of described transition probability matrix, k) element based on: k user is forwarded by j user, quantity about the message of described target topic, the quantity of the forwarding associated user that described j user is corresponding, and magnitude relationship between the quantity of the message about described target topic of described forwarding associated user issue is determined, wherein, described forwarding associated user is in a described N user, the user that the message of issue was forwarded by described j user, j, k is the arbitrary integer in 1 to N,

The 3rd determination module, for according to described transition probability matrix and described weight vector, utilizes Page sort algorithm to determine in described social networks, and a described N user is in the sequence of the influence power in described target topic field.

7. device according to claim 6, is characterized in that, the i element in described weight vectors is determined by following formula:

A _i＝x(a/b)+(1-x)(c/d)

8. according to the device described in claim 6 or 7, it is characterized in that, (j, k) element of described transition probability matrix is determined by following formula:

B _jk＝max(e,1)/max(f,g)

9. according to the device described in any one in claim 6-8, it is characterized in that, described the 3rd determining unit is specifically for according to described transition probability matrix and described weight vector, each self-corresponding Page ranking value of a described N user is carried out to iteration for the first time, and described iteration is for the first time the predefined limited number of time iteration of iterations; According to after described iteration for the first time, each self-corresponding Page ranking value of a described N user, a described N user is divided into the first set and the second set, wherein, Page ranking value corresponding to user in described the second set is less than the iteration stopping threshold value of Page ranking value corresponding to user in described the first set; According to described transition probability matrix and described weight vector, each self-corresponding Page ranking value of a described N user is carried out to iteration for the second time, in described iteration for the second time, the iteration stopping threshold value of Page ranking value corresponding to user in described the first set is less than the iteration stopping threshold value of Page ranking value corresponding to user in described the second set; By after described iteration for the second time, the sequence of the size of each self-corresponding Page ranking value of a described N user is defined as a described N user in the sequence of the influence power in described target topic field.

10. a device for the influence power of definite network node sequence, is characterized in that, comprising:

The first iteration module, carries out iteration for the first time to each self-corresponding Page ranking value of N network node, and described iteration is for the first time the predefined limited number of time iteration of iterations;

Determination module, for each the self-corresponding Page ranking value of described N network node obtaining according to described the first iteration module, a described N network node is divided into the first set and the second set, wherein, Page ranking value corresponding to network node in described the second set is less than Page ranking value corresponding to network node in described the first set;

Secondary iteration module, for each self-corresponding Page ranking value of described N network node that the first iteration module is obtained, carry out iteration for the second time, in described iteration for the second time, the iteration stopping threshold value of the Page ranking value that network node in described the first set that described determination module is determined is corresponding is less than the iteration stopping threshold value of the Page ranking value that network node in described the second set is corresponding;

Described determination module is also for by after described iteration for the second time, and the sequence of the size of each self-corresponding Page ranking value of a described N network node is defined as the sequence of the influence power of a described N network node.