CN108647247A - Key node recognition methods is propagated based on the micro-blog information for improving PageRank algorithms - Google Patents
Key node recognition methods is propagated based on the micro-blog information for improving PageRank algorithms Download PDFInfo
- Publication number
- CN108647247A CN108647247A CN201810338554.0A CN201810338554A CN108647247A CN 108647247 A CN108647247 A CN 108647247A CN 201810338554 A CN201810338554 A CN 201810338554A CN 108647247 A CN108647247 A CN 108647247A
- Authority
- CN
- China
- Prior art keywords
- micro
- user
- group
- information
- text message
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Abstract
The present invention provides key node recognition methods is propagated based on the micro-blog information for improving PageRank algorithms, step includes：1) information collection；2) the highest text message of hop count in the micro-group is determined；3) text message obtained obtains the text message for using "@" user identifier, which is B；4) text message obtained calculates high frequency words；5) it is compared with the high frequency words after the microblogging text message of acquisition user B, determine whether the hop count highest text message of the member in A micro-groups is forwarded by user B, it such as determines user's B forwardings, determines whether B has passed to user C, to form the propagation path of information；6) it determines the micro-group where the user on propagation path, network is paid close attention to build micro-group；7) micro-group with strongest influence power is determined by WeiboRank algorithms.The excavation of middle key node between present invention realization micro-group and in group conveniently carries out group's discovery without setting up complicated data model, has higher stability.
Description
Technical field
It is the present invention relates to data analysis field, more particularly to a kind of to be propagated based on the micro-blog information for improving PageRank algorithms
Key node recognition methods.
Background technology
Microblogging since the advent of the world just becomes the significant product of epoch network, this has the mutual of strong " from media " attribute
Working application just becomes important social media with instantaneity in fact and convenience.But the weak tendency that public credibility is " from media ",
Some key users are engaged in topic operation in information communication process, thus are easy out in vulgar content, rumour, deceptive information etc.
Existing, micro-blog information confidence level is just at a big problem, in addition the spread speed of internet information is exceedingly fast, microblogging public feelings information
Content can form social public security and threaten.
The influence power node during the analysis of micro blog network information data is propagated with mined information is studied, is found with strongest influence power
It is all critically important to propagate node, is monitored to micro blog network marketing, public sentiment and controls unreal microblogging public opinion and have positive effect.
Micro-group is the abbreviation of microblogging group, can polymerize with it is identical hobby or same label user, by it is all therewith
Corresponding topic is all gathered inside micro-group.Microblog users in same micro-group have the higher degree of polymerization and active degree,
Correlative study points out that micro-blog information is more easy to propagate in micro-group, thus unreal micro-blog information is more easy to propagate in group.
Invention content
Discovery method it is an object of the invention to a kind of key node for proposing to propagate based on micro-blog information between micro-group.
The present invention provides a kind of micro-blog informations improving PageRank algorithms to propagate key node recognition methods, step packet
It includes：
1) it acquires the information of all members in a certain micro-group A, deliver microblogging list information, microblogging text message, Yong Huguan
It is information；
2) the highest text message of hop count in the micro-group is determined；
3) text message for using "@" user identifier is obtained by the text message that step 2 obtains, which is B；
4) high frequency words are calculated by the text message that step 3 obtains；
5) it is compared with the high frequency words after the microblogging text message of acquisition user B, determines the forwarding time of the member in A micro-groups
Whether number highest text message is forwarded by user B, such as determines user's B forwardings, using above-mentioned identical method, determines that user B is
It is no that the hop count highest text message of member in A micro-groups has been passed to user C, to form the propagation path of information；
6) it determines the micro-group where the user on propagation path, network is paid close attention to build micro-group；
7) micro-group with strongest influence power, the calculating step of the WeiboRank algorithms are determined by WeiboRank algorithms
For：
71) direct appeal F1 (v) is calculated, user v is located at center, according to the distance of information propagation path, we
It will get up for 1 all node sets with central point distance, form a concentric circles, form N1 layers, it is seen that the node in this layer
Number, as the bean vermicelli set of user v, are denoted as F1 (v), and the size of F (v) characterizes the direct appeal of user v；
72) coverage R is calculated, will get up for 2 all node sets with central point distance, form N2 layers, it is seen that should
Node in layer is the bean vermicelli set of each user in F1 (v), is denoted as F2 (v), and so on, until a maximum concentric circles NM
In layer, until all nodes are leaf node, coverage
Distance value of the point apart from node centered on i；
73) the WR values of user v are calculated, WR values are defined as the product of the direct appeal and average information load of user v
J indicates the connection node that information can be transmitted to by node v；
dvjIndicate the distance between node v and node j；
Characterize average information load；
The WR values of user v are defined as the product of the direct appeal and average information load of user v, the WR values of user v
Bigger, the influence power for characterizing the user is bigger, and the status in micro blog network is more crucial.
Preferred embodiment is：Acquire Sina weibo and microblogging Liang great mainstream medias of Tencent data.
Preferred embodiment is：The User ID that member in a certain micro-group is acquired using WEB crawler technologies is obtained by api interface
Take each relevant parameter of the micro-blog information of JSON file formats；By calling the microblogging of API to read interface function.
Preferred embodiment is：Information is broadcast to me to classify, different class clusters is obtained, according in all kinds of clusters using SVM algorithm
Including the most information of forwarding number class cluster is ranked up from high to low, obtain the highest class cluster of forwarding number.
Preferred embodiment is：High frequency words are obtained using TF-IDF algorithms.
Preferred embodiment is：High frequency words quantity is α, and it is β that most high frequency words are totally propagated in micro-group, takes same cardinality, obtains
Ratio γ=α/β is propagated to high frequency words.
Preferred embodiment is：When body WR values are not much different in micro-group, high frequency words propagate number frequency acquisition δ values,
δ=δ * γ.δ values can be used as another reference parameter, and δ values are bigger, illustrate that micro-group proportion in micro-group relationship is bigger,
Finally obtain the crucial micro-group node during micro-group relationship is propagated.
Beneficial effects of the present invention are as follows：
The key node between micro-group in micro-blog information communication process can be accurately analyzed by this method.For micro-
The Effective selection of rich network marketing, unreal microblogging public opinion and illegal information activities information, system is to user information and key
Word information carries out data clusters analysis, the critical behavioural characteristic of information is extracted, when system delivers micro-blog information from personage
Between, frequency, high frequency words and the microblogging micro-group information of text end@key persons and related person, realize between micro-group and group
The excavation of interior middle key node conveniently carries out group's discovery without setting up complicated data model, has higher
Stability.
Description of the drawings
Fig. 1 is the flow diagram of the present invention；
Fig. 2 is micro-group relational network figure；
Fig. 3 is that micro-group micro-blog information propagates figure in embodiment.
Specific implementation mode
Present invention will be described in further detail below with reference to the accompanying drawings, to enable those skilled in the art with reference to specification text
Word can be implemented according to this.
It should be appreciated that such as " having ", "comprising" and " comprising " term used herein do not allot one or more
The presence or addition of a other elements or combinations thereof.
As shown in Figure 1, the present invention provides a kind of micro-blog informations improving PageRank algorithms to propagate key node identification
Method, step include：
1) it acquires the information of all members in a certain microblogging group A, deliver microblogging list information, microblogging text message, user
Relation information；
2) the highest text message of hop count in the micro-group is determined；
3) text message for using "@" user identifier is obtained by the text message that step 2 obtains, which is B；
4) high frequency words are calculated by the text message that step 3 obtains；
5) it is compared with the high frequency words after the microblogging text message of acquisition user B, determines the forwarding time of the member in A micro-groups
Whether number highest text message is forwarded by user B, such as determines user's B forwardings, using above-mentioned identical method, determines that user B is
It is no that the hop count highest text message of member in A micro-groups has been passed to user C, to form the propagation path of information；
6) it determines the micro-group where the user on propagation path, network is paid close attention to build micro-group；
7) micro-group with strongest influence power, the calculating step of the WeiboRank algorithms are determined by WeiboRank algorithms
For：
71) direct appeal F1 (v) is calculated, user v is located at center, according to the distance of information propagation path, we
It will get up for 1 all node sets with central point distance, form a concentric circles, form N1 layers, it is seen that the node in this layer
Number, as the bean vermicelli set of user v, are denoted as F1 (v), and the size of F (v) characterizes the direct appeal of user v；
72) coverage R is calculated, will get up for 2 all node sets with central point distance, form N2 layers, it is seen that should
Node in layer is the bean vermicelli set of each user in F1 (v), is denoted as F2 (v), and so on, until a maximum concentric circles NM
In layer, until all nodes are leaf node, coverage
Distance value of the point apart from node centered on i；
73) the WR values of user v are calculated, WR values are defined as the product of the direct appeal and average information load of user v
J indicates the connection node that information can be transmitted to by node v；
dvjIndicate the distance between node v and node j；
Characterize average information load；
The WR values of user v are defined as the product of the direct appeal and average information load of user v, the WR values of user v
Bigger, the influence power for characterizing the user is bigger, and the status in micro blog network is more crucial.
Preferred embodiment is：Acquire Sina weibo and microblogging Liang great mainstream medias of Tencent data.
Preferred embodiment is：The User ID that member in a certain micro-group is acquired using WEB crawler technologies is obtained by api interface
Take each relevant parameter of the micro-blog information of JSON file formats；By calling the microblogging of API to read interface function.
Preferred embodiment is：Information is broadcast to me to classify, different class clusters is obtained, according in all kinds of clusters using SVM algorithm
Including the most information of forwarding number class cluster is ranked up from high to low, obtain the highest class cluster of forwarding number.
Preferred embodiment is：High frequency words are obtained using TF-IDF algorithms.
Preferred embodiment is：High frequency words quantity is α, and it is β that most high frequency words are totally propagated in micro-group, takes same cardinality, obtains
Ratio γ=α/β is propagated to high frequency words.
Preferred embodiment is：When body WR values are not much different in micro-group, high frequency words propagate number frequency acquisition δ values,
δ=δ * γ.δ values can be used as another reference parameter, and δ values are bigger, illustrate that micro-group proportion in micro-group relationship is bigger,
Finally obtain the crucial micro-group node during micro-group relationship is propagated.
Embodiment
The present invention provides a kind of micro-blog informations improving PageRank algorithms to propagate key node recognition methods, including：
1) nearly three months hot issues are acquired first, heat searches related propositions and the phase such as seniority among brothers and sisters and responsive type topic public opinion
Community information is closed, the User ID of member in a certain micro-group is acquired using WEB crawler technologies, by api interface, obtains JSON
Each relevant parameter of micro-blog information of file format；By calling the microblogging of API to read interface function " statuses/user_
Timeline ", you can obtain the microblogging list that user delivers.Obtaining the field needed includes：
Created_at string microblogging creation times
Id int64 microbloggings ID
The microblogging ID of idstr string character string types
Text string micro-blog information contents
Source string microbloggings source
In_reply_to_status_id string reply ID
The user information field of user object microblogging authors
The former micro-blog information field that retweeted_status object are forwarded is returned when the microblogging is forwarding microblogging
It returns
Reposts_countint forwarding numbers
Interface function " friendships/show " is read, to obtain the details of two customer relationships.Interface function
" friendships/followers " obtains the bean vermicelli list of user respectively, then judge to whether there is between user bean vermicelli/
Concern relation.User other users ID of interest is acquired, the user collected in the micro-group is only limited in；It is micro- by obtaining simultaneously
The information of rich user's real identification, can make analysis more comprehensive, and research is more deep；
2) according to the text text micro-blog information contents acquired in step 1, in conjunction with classical taxonomy algorithm SVM algorithm to microblogging
Information is classified, and different class clusters is obtained, according to the most information of forwarding number included in all kinds of clusters from high to low to class
Cluster is ranked up, and obtains the highest class cluster of forwarding number.Using highest class cluster as sample, it is related to obtain news messages in such cluster
All users are obtained, the cluster Type of Collective between user is completed here according to customer relationship, obtaining relevant all users has this
A little users text most with forwarding information is propagated can build microblogging propagation relational network, such as style class hot ticket comes
Build communication network；
3) highest class cluster propagates information as sample in obtaining step 2, and high frequency words are obtained using TF-IDF algorithms, calculates
Content of text similarity, detailed process：Several keywords (such as 10) are taken out from every article, are merged into a collection
It closes, calculates word frequency of the every article for the word in this set；
4) the parsing data based on step 2 and step 3 obtain the micro-blog information of party B-subscriber after party A-subscriber@party B-subscriber, calculate
The similitude of n document of two user version information of AB between any two, so that it may to obtain the microblogging text message of party A-subscriber@party B-subscriber
Whether there is propagation in party B-subscriber's microblogging, if any propagation, whether analysis analog information has passed to C user, according to ABC user institute
Micro-group group so that build micro-group pay close attention to network, as shown in Figure 2；
5) according to the propagation relationship between the user of step 4, we propagate scale to information using cascade method and carry out
Tracking can have by judging every micro-blog information at two in the case where customer relationship network determines between the user of side
Relationship and the information such as whether propagate between micro-group, and then the association on realizing micro-group and propagating, builds micro-group relational network.And according to
All micro-groups where user carry out text cluster to micro-group, and obtain and totally propagate most high frequency words in micro-group as β；
6) WeiboRank algorithms calculate the appeal and coverage for considering user itself, we count in micro-group
The total amount WR values of analog information are propagated, WR is higher, illustrates that micro-group proportion in micro-group relationship is bigger.Identical information is in micro-group
Propagation number is more, can equally increase the proportion of micro-group propagation, counts the number that analog information is propagated in same micro-group, setting
For parameter δ.Above we also refer to, micro-group be can polymerize with it is identical hobby or same label user, if
If having the people of same label more to meet in the topic information and micro-group of propagation, which can widely pass in the micro-group
It broadcasts, has obtained the high frequency words of analog information above, setting high frequency words quantity is α, and totally propagating most high frequency words in micro-group is
β takes same cardinality, obtains high frequency words and propagates ratio γ=α/β.When body WR values are not much different in micro-group, high frequency words disseminator
Number frequency acquisition δ values, δ=δ * γ.δ values can be used as another reference parameter, and δ values are bigger, illustrate that micro-group compares in micro-group relationship
It is again bigger, finally obtain the crucial micro-group node during micro-group relationship is propagated；Page Rank algorithms are based on passing through " concern/bean vermicelli "
Relationship builds microblog users relational network, and the PR values of user are the use for being delivered evenly to chain and going out in Page Rank algorithms
What family was got on, result of which can ignore the significance level of user itself, due to not accounting for the weak tie between corporations
Hinged node can not solve the problems, such as maximizing influence.And WeiboRank algorithm synthesis considers the direct appeal of user, shadow
Ring range, for some features organization's microblogging account in calculate node advantageously, use WeiboRank algorithms
The key node between micro-group and user is calculated, the key node during micro-blog information is propagated is obtained.
7) specific micro-group WR value example analysis
As shown in figure 3, information is forwarded or is issued by someone during information is propagated, and@three different micro-group A, B, C,
The WR values of three micro-groups are cleared out respectively.Include three people, A in micro-group A1、A2And A3.It, can be with according to the definition of WeiboRank
Calculate A1Go out number of degrees F1 (v) be 1, coverage R be 1, average information loadFor (1 × 1)/1=
1, then node A1WR values be 1 × 1=1, A2Go out number of degrees F1 (v) be 2, coverage R be 4, average information loadFor (1 × 1+2 × 2)/4=1.5, then node A2WR values be 1.5 × 2=3, A3Go out number of degrees F1 (v) be 3, shadow
It is 11 to ring range R, average information loadFor (1 × 3+2 × 6+3 × 2)/11=1.82, then node A3WR values
For 1.82 × 3=5.46, in general the overall WR values of micro-group A are 1+3+5.46=9.46.Same method calculates separately out B1
WR values be 3, B2WR values be 1.75 and B3WR values be 4.88, the overall WR values of micro-group B are 9.63, C1WR values be
7.43、C2WR values be 17.33, the overall WR values of micro-group C are 24.76.
The overall WR values of micro-group A and micro-group B are not much different it can be seen from statistical result, although in micro-group C groups only
Two interpersonal communication information, but the number totally propagated is more, thus summation WR values are more than micro-group AB, thus micro-group C is propagating letter
Proportion is greater than two groups of AB during breath.
Compare micro-group A and micro-group B again, the number that two groups propagate all is 3 people, and the WR of group A and group B differences are not
Greatly, the high frequency words of the text message of crowd A and group B and the overall high frequency words of 3 people propagation information are calculated, according in step 6
Method calculates δ values, is worth size by δ to differentiate the importance of crowd A and group B propagated in micro-group.Thus it can obtain
The micro-group that information most influences in propagating.
Although the embodiments of the present invention have been disclosed as above, but its is not only in the description and the implementation listed
With it can be fully applied to various fields suitable for the present invention, for those skilled in the art, can be easily
Realize other modification, therefore without departing from the general concept defined in the claims and the equivalent scope, the present invention is simultaneously unlimited
In specific details and legend shown and described herein.
Claims (7)
1. propagating key node recognition methods based on the micro-blog information for improving PageRank algorithms, which is characterized in that step includes：
1) it acquires the information of all members in a certain micro-group A, deliver microblogging list information, microblogging text message, customer relationship letter
Breath；
2) the highest text message of hop count in the micro-group is determined；
3) text message for using "@" user identifier is obtained by the text message that step 2 obtains, which is B；
4) high frequency words are calculated by the text message that step 3 obtains；
5) it is compared with the high frequency words after the microblogging text message of acquisition user B, determines the hop count of the member in A micro-groups most
Whether whether high text message is forwarded by user B, such as determine user's B forwardings, using above-mentioned identical method, determine user B A
The hop count highest text message of member in micro-group has passed to user C, to form the propagation path of information；
6) it determines the micro-group where the user on propagation path, network is paid close attention to build micro-group；
7) determine that micro-group with strongest influence power, the calculating step of the WeiboRank algorithms are by WeiboRank algorithms：
71) direct appeal F1 (v) is calculated, user v is located at center, and according to the distance of information propagation path, we will be with
Central point distance is got up for 1 all node sets, forms a concentric circles, forms N1 layers, it is seen that the interstitial content in this layer,
The as bean vermicelli set of user v, is denoted as F1 (v), and the size of F (v) characterizes the direct appeal of user v；
72) coverage R is calculated, will get up for 2 all node sets with central point distance, form N2 layers, it is seen that in the layer
Node be each user in F1 (v) bean vermicelli set, be denoted as F2 (v), and so on, until maximum NM layers of a concentric circles
In, until all nodes are leaf node, coverage
Distance value of the point apart from node centered on i；
73) the WR values of user v are calculated, WR values are defined as the product of the direct appeal and average information load of user v
J indicates the connection node that information can be transmitted to by node v；
dvjIndicate the distance between node v and node j；
Levy average information load；
The WR values of user v are defined as the product of the direct appeal and average information load of user v, and the WR values of user v are bigger,
The influence power for characterizing the user is bigger, and the status in micro blog network is more crucial.
2. according to claim 1 propagate key node recognition methods based on the micro-blog information for improving PageRank algorithms,
It is characterized in that, acquisition Sina weibo and microblogging Liang great mainstream medias of Tencent data.
3. according to claim 1 propagate key node recognition methods based on the micro-blog information for improving PageRank algorithms,
It is characterized in that, acquiring the User ID of member in a certain micro-group using WEB crawler technologies, by api interface, JSON texts are obtained
Each relevant parameter of micro-blog information of part format；By calling the microblogging of API to read interface function.
4. according to claim 1 propagate key node recognition methods based on the micro-blog information for improving PageRank algorithms,
Classify it is characterized in that, broadcasting information to me using SVM algorithm, different class clusters is obtained, included in all kinds of clusters
The most information of forwarding number is from high to low ranked up class cluster, obtains the highest class cluster of forwarding number.
5. according to claim 1 propagate key node recognition methods based on the micro-blog information for improving PageRank algorithms,
It is characterized in that, obtaining high frequency words using TF-IDF algorithms.
6. according to claim 1 propagate key node recognition methods based on the micro-blog information for improving PageRank algorithms,
It is characterized in that, high frequency words quantity is α, it is β that most high frequency words are totally propagated in micro-group, takes same cardinality, obtains high frequency words
Propagation ratio γ=α/β.
7. according to claim 6 propagate key node recognition methods based on the micro-blog information for improving PageRank algorithms,
It is characterized in that, when body WR values are not much different in micro-group, high frequency words propagate number frequency acquisition δ values, δ=δ * γ.δ values can be made
For another reference parameter, δ values are bigger, illustrate that micro-group proportion in micro-group relationship is bigger, finally obtain in the propagation of micro-group relationship
Crucial micro-group node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810338554.0A CN108647247A (en) | 2018-04-16 | 2018-04-16 | Key node recognition methods is propagated based on the micro-blog information for improving PageRank algorithms |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810338554.0A CN108647247A (en) | 2018-04-16 | 2018-04-16 | Key node recognition methods is propagated based on the micro-blog information for improving PageRank algorithms |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108647247A true CN108647247A (en) | 2018-10-12 |
Family
ID=63746524
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810338554.0A Pending CN108647247A (en) | 2018-04-16 | 2018-04-16 | Key node recognition methods is propagated based on the micro-blog information for improving PageRank algorithms |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108647247A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110110974A (en) * | 2019-04-17 | 2019-08-09 | 福建天泉教育科技有限公司 | The recognition methods of crucial leader of opinion and computer readable storage medium |
| CN110674290A (en) * | 2019-08-09 | 2020-01-10 | 国家计算机网络与信息安全管理中心 | Relationship prediction method, device and storage medium for overlapping community discovery |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105260474A (en) * | 2015-10-29 | 2016-01-20 | 俞定国 | Microblog user influence computing method based on information interaction network |
| CN106603387A (en) * | 2016-12-20 | 2017-04-26 | 西南石油大学 | Microblog forwarding path prediction method and system based on microblog forwarding relationship |
| CN106599243A (en) * | 2016-12-20 | 2017-04-26 | 西南石油大学 | Method and system for predicting micro-blog forwarding path based on micro-blog themes |
| CN106991160A (en) * | 2017-03-30 | 2017-07-28 | 武汉大学 | A kind of microblogging propagation prediction method based on user force and content |
| CN107633260A (en) * | 2017-08-23 | 2018-01-26 | 上海师范大学 | A kind of social network opinion leader method for digging based on cluster |
-
2018
- 2018-04-16 CN CN201810338554.0A patent/CN108647247A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105260474A (en) * | 2015-10-29 | 2016-01-20 | 俞定国 | Microblog user influence computing method based on information interaction network |
| CN106603387A (en) * | 2016-12-20 | 2017-04-26 | 西南石油大学 | Microblog forwarding path prediction method and system based on microblog forwarding relationship |
| CN106599243A (en) * | 2016-12-20 | 2017-04-26 | 西南石油大学 | Method and system for predicting micro-blog forwarding path based on micro-blog themes |
| CN106991160A (en) * | 2017-03-30 | 2017-07-28 | 武汉大学 | A kind of microblogging propagation prediction method based on user force and content |
| CN107633260A (en) * | 2017-08-23 | 2018-01-26 | 上海师范大学 | A kind of social network opinion leader method for digging based on cluster |
Non-Patent Citations (1)
| Title |
|---|
| 陆静: "基于微博网络信息挖掘与传播的关键技术研究", 《中国博士学位论文全文数据库信息科技辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110110974A (en) * | 2019-04-17 | 2019-08-09 | 福建天泉教育科技有限公司 | The recognition methods of crucial leader of opinion and computer readable storage medium |
| CN110110974B (en) * | 2019-04-17 | 2022-03-29 | 福建天泉教育科技有限公司 | Key opinion leader identification method and computer readable storage medium |
| CN110674290A (en) * | 2019-08-09 | 2020-01-10 | 国家计算机网络与信息安全管理中心 | Relationship prediction method, device and storage medium for overlapping community discovery |
| CN110674290B (en) * | 2019-08-09 | 2023-03-10 | 国家计算机网络与信息安全管理中心 | Relationship prediction method, device and storage medium for overlapping community discovery |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Nobari et al. | Characteristics of viral messages on Telegram; The world’s largest hybrid public and private messenger | |
| Wang et al. | Understanding the power of opinion leaders’ influence on the diffusion process of popular mobile games: Travel Frog on Sina Weibo | |
| Cole-Lewis et al. | Social listening: a content analysis of e-cigarette discussions on Twitter | |
| CN104021233B (en) | Social network friend recommendation method based on community discovery | |
| Farseev et al. | Harvesting multiple sources for user profile learning: a big data study | |
| Nettleton | Data mining of social networks represented as graphs | |
| Pal et al. | Identifying topical authorities in microblogs | |
| Yuan et al. | We know how you live: exploring the spectrum of urban lifestyles | |
| Saez-Trumper et al. | Finding trendsetters in information networks | |
| US20080070209A1 (en) | Identifying influential persons in a social network | |
| Zhang et al. | Discover millions of fake followers in Weibo | |
| Song et al. | Who are the spoilers in social media marketing? Incremental learning of latent semantics for social spam detection | |
| CN104834695A (en) | Activity recommendation method based on user interest degree and geographic position | |
| CN103179198A (en) | Topic influence individual digging method based on relational network | |
| Cao et al. | Collusion-aware detection of review spammers in location based social networks | |
| Wang et al. | Classification of private tweets using tweet content | |
| Li et al. | What are Chinese talking about in hot weibos? | |
| Li et al. | Fake reviews tell no tales? dissecting click farming in content-generated social networks | |
| Yang et al. | Comparison and modelling of country-level microblog user and activity in cyber-physical-social systems using Weibo and Twitter data | |
| CN108647247A (en) | Key node recognition methods is propagated based on the micro-blog information for improving PageRank algorithms | |
| Feng et al. | New emoji requests from Twitter users: when, where, why, and what we can do about them | |
| Mittal et al. | Pinned it! A large scale study of the Pinterest network | |
| Yang et al. | Content Mining and Network Analysis of Microblog Spam. | |
| Mudgal et al. | Analysis of tweets to find the basis of popularity based on events semantic similarity | |
| Amini et al. | Marketing in a random network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181012 |
|
| RJ01 | Rejection of invention patent application after publication |