CN107273396A - A kind of social network information propagates the system of selection of detection node - Google Patents
A kind of social network information propagates the system of selection of detection node Download PDFInfo
- Publication number
- CN107273396A CN107273396A CN201710144748.2A CN201710144748A CN107273396A CN 107273396 A CN107273396 A CN 107273396A CN 201710144748 A CN201710144748 A CN 201710144748A CN 107273396 A CN107273396 A CN 107273396A
- Authority
- CN
- China
- Prior art keywords
- user
- topic
- node
- similitude
- keyword
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 27
- 238000004891 communication Methods 0.000 claims abstract description 30
- 238000005295 random walk Methods 0.000 claims abstract description 16
- 238000004364 calculation method Methods 0.000 claims abstract description 15
- 238000000034 method Methods 0.000 claims abstract description 15
- 230000007480 spreading Effects 0.000 claims description 13
- 230000005540 biological transmission Effects 0.000 claims description 9
- 238000000205 computational method Methods 0.000 claims description 5
- 239000011159 matrix material Substances 0.000 claims description 4
- 230000004927 fusion Effects 0.000 claims description 3
- 238000013508 migration Methods 0.000 claims description 3
- 230000005012 migration Effects 0.000 claims description 3
- 230000007704 transition Effects 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 2
- 238000004422 calculation algorithm Methods 0.000 abstract description 13
- 230000002452 interceptive effect Effects 0.000 abstract description 7
- 230000000694 effects Effects 0.000 abstract description 4
- 230000000644 propagated effect Effects 0.000 abstract description 3
- 230000006855 networking Effects 0.000 abstract description 2
- 230000007547 defect Effects 0.000 abstract 1
- 238000009792 diffusion process Methods 0.000 description 5
- 239000002245 particle Substances 0.000 description 4
- 238000011160 research Methods 0.000 description 3
- 238000011524 similarity measure Methods 0.000 description 3
- 102100034799 CCAAT/enhancer-binding protein delta Human genes 0.000 description 2
- 101000945965 Homo sapiens CCAAT/enhancer-binding protein delta Proteins 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000001902 propagating effect Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 241000209202 Bromus secalinus Species 0.000 description 1
- 244000025254 Cannabis sativa Species 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000003012 network analysis Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000005303 weighing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Abstract
Detection node system of selection is propagated the present invention relates to a kind of social network information.The present invention defines topic similitude, keyword similitude, and sets up user's topic interest preference model of " user's topic keyword " three levels；The similarity calculation method of user node in the community network of aggregators relation and node topic preference；Community network Information Communication detection node sort method based on random walk strategy.Instant invention overcomes the respective defect of past various algorithms.The present invention considers user structure similitude and user's topic preference similitude, balance parameters can dynamically be adjusted according to different community networks, with more preferable detection efficiency, and based on the relational structure and interactive structure in community network are collectively promoted in terms of network evolution, the relations problems of social networking relationships structure and interactive structure are considered from deeper level, satisfied effect is achieved.
Description
Technical field
The present invention relates to Information Communication detection method, more particularly to a kind of social network information propagates the selection of detection node
Method.
Background technology
Community network (Social Network) refers to the complex network body formed between Social Individual by social relationships
System, its relation between the individual and individual in society is constituted.In recent years, with Twitter, Facebook, microblogging, micro-
Letter etc. is developed rapidly for the online community network of representative, the Information Communication (Information based on community network
Diffusion) also more and more deeply and extensive, communication target includes media event, social hotspots, fashion, or new hair
Bright, new creation, new thought, it is also possible to be network rumour etc..The letter in Information Communication and traditional media in community network
Breath propagation is compared, the features such as showing extensive property, multimode state property, real-time, rapidity, and it is public to economic society and country
The influence of safety is more and more deep.In " the Egyptian revolution " of outburst in 2011, criminal utilizes Twitter and Facebook
Started a rumour without restraint Deng community network or media, instigate the will of the people, fallacious message and criminal organizations activity are propagated, in the push wave of social media
Help under billows, riot is by extremely amplification and rapid development.After being broken out in April, 2013 Sichuan " Yaan earthquake ", microblogging, which turns into, most to be had
The information spreading medium of power, all kinds of government affairs microbloggings, leader of opinion, grass roots account etc. make full use of the diffusion of information ability of microblogging,
Earthquake rescue prompting is issued, is that earthquake relief work plays a positive role but on the other hand, also has criminal using microblogging propagation
Rumour, cheats the public, causes the social uneasy and common people panic, brings extremely bad consequence for the Information Communication in community network,
How quick obtaining Information Communication situation therein, in time find current popular focus incident or bad climate, turn into
Urgent problem to be solved, this monitors and safeguarded that national public safety is significant for public sentiment.For large scale community net
Network and mass data information, ensure that Detection results would generally choose limited in communication network to reduce while testing cost
Node detects whole net as observation node by tracking the state change of these observation nodes or analyzing the information of its issue
Information Communication situation in network.
Before the present invention makes, in recent years, some researchers are studied Information Communication detection method,
Such as：" the Locating the source of of publication in document Physical Review Letters, 2012,109 (6)
Diffusion in Iarge scale network " to how to determine that Information Communication source is studied in network, by
Sparsely placement sensor in network, obtains the different time that observation node infects information, article provides an efficient algorithm, right
Any tree-shaped communication network can be in O (N) in the time, can be within O (N3) times with certain precision for any figure of propagating
Determine Information Communication source.Proceedings of the 10th ACMConference on Electronic Commerce are printed
Step on " A.Social influence and the diffusion of user created content " are to Information Communication
Early adopter (Early Adopter) research show, these people generally without many follower (node in-degree compared with
It is small), their social networks line duration is also below average line duration .The 18th ACMSIGKDD International
" " of Conference on Knowledge Discovery and Data Mining publications have studied trend promoter
(Trendsetter) the characteristics of, trend promoter is early adopter and the disseminator of the focus trend occurred in network, article
Binding time factor of evolution, based on PageRank thoughts, gives an algorithm for excavating different topic field trend promoters
.the 13th ACM SIGKDD International Conference on Knowledge Discovery and Data
Published on Mining " Cost effective out break detection in networks " be directed to blog network in
Information breaks out detection (Outbreak Detection) and proposes the greedy algorithm based on time module feature (Sub Modularity)
CELF.Article by infomation detection problem it is abstract be the maximized object function R (A) of one group of needs, A represents to need to dispose sensor
Observation node set.R (A) can be by k nodal test to Information Communication cascade quantity, or thus bring
The decrement articles of Information Communication number of the infected, which demonstrate R (A), has time module feature, based on this one heuristic greed of proposition
" the A Novel Algorithm for published on algorithm CELF.CHINESE JOURNAL OF COMPUTERS
Information Diffusion Detection in Social Network " propose a kind of transmission capacity sort algorithm
DiffRank, transmission capacity most strong k node is chosen according to arithmetic result as node is observed and detects overall network information
Propagate situation,
In existing correlative study, most of algorithm have ignored using the relational structure in community network as starting point
Interactive structure, it is impossible to reach gratifying effect.
The content of the invention
The purpose of the present invention be that overcome drawbacks described above there is provided a kind of social network information propagate detection node selection
Method.
The technical scheme is that：
A kind of social network information propagates the system of selection of detection node, and it is mainly technically characterized by：
4) topic similitude, keyword similitude are defined, and sets up the user of " user-topic-keyword " three levels
Topic interest preference model；
5) in the community network of aggregators relation and node topic preference user node similarity calculation method；
6) the community network Information Communication detection node sort method based on random walk strategy.
The user's topic interest preference model for setting up " user-topic-keyword " three levels, its step：
1) each user's issue and the topic information received are obtained from the journal file of community network；
If 2) communicated between two users, one is produced from user is sent to by the oriented of transmission user
Side；
3) when including a keyword in topic, then a nonoriented edge is produced from theme to the keyword；
4) user issues a theme, then an oriented even side is set up from user to topic, if user receives a words
Topic, then set up an oriented even side from topic to user；
5) similitude of topic is defined, using the two themes as end points if the similitude of two topics exceedes threshold value,
Set up a nonoriented edge；
6) define the similitude of keyword, if two keywords similitude exceed threshold value if using the two keywords as
End points, sets up a nonoriented edge.
The similarity calculation method step of user node in the community network of the aggregators relation and node topic preference
Suddenly it is：
1) computational methods of user structure similitude are provided；
2) computational methods of user's topic preference similitude are provided；
3) provide fusion user structure similitude and user's topic preference similitude community network user node it is similar
Property computational methods.
The community network Information Communication detection node sort method based on random walk strategy, specifically includes step：
1) definition and the calculation formula of Information Communication probability are provided；
2) probability of spreading generation probability of spreading figure and transition probability matrix between user node similarity and node are combined；
3) carry out having inclined random walk on probability of spreading figure, obtain the transmission capacity measurement of each node.
For case above, the present invention proposes a kind of aggregators structure from the angle of community network Information Communication
The user node importance ranking method of relation and node topic preference.
The advantage of the invention is that：It is similar that the selection of community network Information Communication detection node has considered user structure
Property and user's topic preference similitude, balance parameters can be dynamically adjusted according to different community networks, with more preferable inspection
Efficiency is surveyed, and based on the relational structure and interactive structure in community network are collectively promoted in terms of network evolution, from deeper
The secondary relations problems for considering social networking relationships structure and interactive structure, achieve satisfied effect.
Brief description of the drawings
Fig. 1 --- schematic flow sheet of the present invention.
Fig. 2 --- the present invention sets up user's topic preference pattern schematic diagram.
Embodiment
The present invention technical thought be：
1998 propose famous PageRank algorithms, and the algorithm is used to arrange the importance of Webpage node
Sequence, and the algorithm is applied successfully has founded Google search engine.The algorithm mainly make use of markov random walk model,
It is in order to which webpage is corresponding with the random walk model, webpage is corresponding with the particle in model, by the oriented link structure of webpage
Corresponding with particle advance, the link of such webpage redirects probability and is just successfully changed the probability transfer advanced for particle.Cause
There are different centrads and influence power for different web page joints, therefore web page joint is under markov random walk model
The click probability of acquisition may also be different, and the master that ranking is PageRank is carried out to web page joint according to different click probability
Want thought.Research shows that the node with nodes higher in-degree might not have a very big impact power.It is logical by these methods
What is often found is the core node in network, or certain field " leader of opinion ".Information Communication in community network is by node
The influence of influence power, but influence power maximum might not mean that transmission capacity is most strong, because influence power parser does not have
Consider node to factors such as the degree of participation of various information flows and issue article or the scales for propagating information.Existing algorithm is to society
Relational structure and interactive structure in meeting network collectively promote deficiency from the aspect of network evolution, do not consider society from deeper level
The relations problems of meeting cyberrelationship structure and interactive structure.
The technical characteristics of the present invention are embodied in：
1) user's topic interest preference model of " user-topic-keyword " three levels is set up
Particular technique route is：1. the interrelated relation of user, topic and keyword set up side between them； 2.
The all of user can intactly be preserved by user's topic interest preference model of " user-topic-keyword " three levels
Information, is that subsequent analysis provides the foundation.
2) similarity calculation method of user node in the community network of aggregators relation and node topic preference is designed.
Particular technique route：1. define the calculation formula of user structure similitude；2. define user's topic preference similitude
Calculation formula；3. setting balance parameter merges the social network of user structure similitude and user's topic preference similitude to define
The similarity calculation method of network user node.
The present invention is specifically described below, its flow is such as shown in " Fig. 1 --- schematic flow sheet of the present invention ".
1) topic similarity, keyword similitude are defined, and sets up the user of " user-topic-keyword " three levels
Topic interest preference model
User profile is propagated analysis and set up in user's topic interest preference model, then utilizes network analysis method
It is analyzed, the specific method step for setting up " user-topic-keyword " three level user topic interest preference models
It is rapid as follows：
● the action message of each user is obtained from the journal file of community network website, including：The communication information, transmission
With receive topic information etc.；
● the node set up in the network model of three level, network includes：User, topic and keyword, signal
Figure is as shown in " Fig. 2 --- the present invention sets up user's topic preference pattern schematic diagram ".
● the connection between node includes following several situations：If a) communicated between two users, produce
One from user is sent to the directed edge for being sent user；B) when including a keyword in topic, then from topic to the pass
Keyword produces a nonoriented edge；C) user issues a topic, then an oriented even side is set up from user to theme, if user
A topic is received, then an oriented even side is set up from theme to user.D) similitude of topic, i-th of theme and jth are defined
The Similarity measures formula of individual theme is：In formulaRepresent the key that the i themes are included
Set of words.When the similitude of two themes exceedes threshold value then using the two themes as end points, a nonoriented edge is set up；E) definition is closed
Similitude Sim (the kd of the similitude of keyword, i-th of keyword and j-th of keywordi, kdj) calculation formula is： In formulaRepresent that the set of topic occurs in i-th of keyword.When the similitude of two keywords
More than threshold value then using the two keywords as end points, a nonoriented edge is set up；
2) similarity calculation method of user node in the community network of aggregators relation and node topic preference is designed.
The phenomenon of " things of a kind come together, people of a mind fall into the same group " is widely present in community network, and research shows：Similarity and node between node
Between influence power there is positive correlation, be also to influence one of principal element of Information Communication.Node viAnd vjBetween it is similar
Degree Sim (vi, vj) represent, this patent chooses the structural similarity and two dimensions of user's topic preference similitude of node to weigh
Measure the similitude of node.The Similarity measures side of user node in the community network of aggregators relation and node topic preference
Method：
● user structure similitude uses formulaCalculating is obtained；
● user's topic preference similitude is obtained by calculating the inner product of two user's topic preference vectors,
● the Similarity measures side of user node in the community network of aggregators relation and node topic preference. Sims
(vi, vj)=b × Sims(vi, vj)+(1-b)×Simt(vi, vj), wherein parameter b ∈ [0,1] are similar for weighing user structure
With the ratio shared by user's topic preference.
3) the community network Information Communication detection node sort method based on random walk strategy.
It is implemented as follows：
● probability of spreading p (v are provided firsti, vj) calculation formula：
Define 1：In figure GRIn=(V, E), c is cascaded for information, if node vi∈ c, andThen information is from viPass
It is multicast to node vjProbability p (vi, vj) represent.
All information cascade c in C are decomposed, each cascade is decomposed into the individual single step (v of L (c)i→vj, ti), finally
Obtain GRThe corresponding total the number of transmissions n of=(V, E) Zhong Gebianij.It would generally be shown between the high node of probability of spreading more
Propagation times, therefore p (vi, vj) and viTo vjBetween propagation times be directly proportional, choose exponential relationship model.Therefore have
● it can be generated based on G with reference to probability of spreading between user node similarity and nodeR=(V, E) probability of spreading
Figure, its adjacency matrix AN×NRepresent,To AN×NCarry out row normalization
Obtain the transition probability matrix Q of random walkN×N, wherein
● carry out having inclined random walk on probability of spreading figure, each step of random walk is according to formula R=d × QR+
(1-d) × e iteration is carried out, and wherein R is a N-dimensional vector, and each components R (i) represents random walk and terminates rear corresponding node vi
Accessed probability, many particles simultaneously the migration on probability of spreading figure, each node is one " attractor ", its attraction
The transmission capacity of node is represented, walk process is divided into two, and part 1 presses probability d migration to the neighbor node of node；2nd
Divide the arbitrary node according to probability (1-d) random skip into network, the probability that each node is accessed randomly is determined by vectorial e,
There is inclined random walk to assign different values by each component to vectorial e, to represent that ion jumps to each node at random
Difference preference, that is, the attraction of each " attractor " it is different.The node in c is cascaded for Information Communication, it is received
Message is more early, and position represents that its transmissibility is stronger, the value of corresponding e (i) is also bigger about close to c front end.According to e's (i)
Value is ranked up the final row that obtained result is the community network Information Communication detection node based on random walk strategy
Sequence.
Claims (4)
1. a kind of social network information propagates detection node system of selection, it is characterized in that：
1) topic similitude, keyword similitude are defined, and sets up user's topic of " user-topic-keyword " three levels
Interest preference model；
2) in the community network of aggregators relation and node topic preference user node similarity calculation method；
3) the community network Information Communication detection node sort method based on random walk strategy.
2. a kind of social network information according to claim 1 propagates detection node system of selection, it is characterized in that：Set up
The step of user's topic interest preference model of " user-topic-keyword " three levels：
1) each user's issue and the topic information received are obtained from the journal file of community network；
If 2) communicated between two users, one is produced from user is sent to the directed edge for being sent user；
3) when including a keyword in topic, then a nonoriented edge is produced from theme to the keyword；
4) user issues a theme, then an oriented even side is set up from user to topic, if user receives a topic,
An oriented even side is set up from topic to user；
5) similitude of topic is defined, using the two themes as end points if the similitude of two topics exceedes threshold value, is set up
One nonoriented edge；
6) similitude of keyword is defined, using the two keywords as end if the similitude of two keywords exceedes threshold value
Point, sets up a nonoriented edge.
3. a kind of social network information according to claim 1 propagates detection node system of selection, it is characterized in that：Fusion section
The similarity calculation method step of user node is in the community network of point relation and node topic preference：
1) computational methods of user structure similitude are provided；
2) computational methods of user's topic preference similitude are provided；
3) the similitude meter of the community network user node of fusion user structure similitude and user's topic preference similitude is provided
Calculation method.
4. a kind of social network information according to claim 1 propagates detection node system of selection, it is characterized in that：Based on
The community network Information Communication detection node sort method of machine migration strategy, specifically includes step：
1) definition and the calculation formula of Information Communication probability are provided；
2) probability of spreading generation probability of spreading figure and transition probability matrix between user node similarity and node are combined；
3) carry out having inclined random walk on probability of spreading figure, obtain the transmission capacity measurement of each node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710144748.2A CN107273396A (en) | 2017-03-06 | 2017-03-06 | A kind of social network information propagates the system of selection of detection node |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710144748.2A CN107273396A (en) | 2017-03-06 | 2017-03-06 | A kind of social network information propagates the system of selection of detection node |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107273396A true CN107273396A (en) | 2017-10-20 |
Family
ID=60073569
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710144748.2A Pending CN107273396A (en) | 2017-03-06 | 2017-03-06 | A kind of social network information propagates the system of selection of detection node |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107273396A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108804689A (en) * | 2018-06-14 | 2018-11-13 | 合肥工业大学 | The label recommendation method of the fusion hidden connection relation of user towards answer platform |
| CN109213953A (en) * | 2018-08-13 | 2019-01-15 | 华东师范大学 | A kind of modeling method of social networks multi information propagation model |
| CN109726297A (en) * | 2018-12-28 | 2019-05-07 | 沈阳航空航天大学 | A kind of two subnetwork node prediction algorithms based on mutual exclusion strategy |
| CN109815345A (en) * | 2019-02-25 | 2019-05-28 | 南京大学 | A kind of knowledge mapping embedding grammar based on path |
| CN111723578A (en) * | 2020-06-09 | 2020-09-29 | 平安科技（深圳）有限公司 | Hot spot prediction method and device based on random walk model and computer equipment |
| CN111861122A (en) * | 2020-06-18 | 2020-10-30 | 北京航空航天大学 | Social network information credibility evaluation method based on propagation attribute similarity |
| CN112686765A (en) * | 2020-12-09 | 2021-04-20 | 天津大学 | Information propagation rule mining method based on social network |
| CN112765329A (en) * | 2020-12-31 | 2021-05-07 | 清华大学 | Method and system for discovering key nodes of social network |
| CN115185715A (en) * | 2022-09-13 | 2022-10-14 | 深圳市华云中盛科技股份有限公司 | Case popularity diffusion processing method based on social network information |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102880728A (en) * | 2012-10-31 | 2013-01-16 | 中国科学院自动化研究所 | Individualized ordering method for video searching results of famous persons |
| CN103412872A (en) * | 2013-07-08 | 2013-11-27 | 西安交通大学 | Micro-blog social network information recommendation method based on limited node drive |
| CN103530421A (en) * | 2012-11-02 | 2014-01-22 | 中国人民解放军国防科学技术大学 | Micro-blog based event similarity measuring method and system |
| CN103838806A (en) * | 2013-10-10 | 2014-06-04 | 哈尔滨工程大学 | Analysis method for subject participation behaviors of user in social network |
| CN104123352A (en) * | 2014-07-10 | 2014-10-29 | 西安理工大学 | Method for measuring influence of users on topic hierarchy for MicroBlog |
| CN104268230A (en) * | 2014-09-28 | 2015-01-07 | 福州大学 | Method for detecting objective points of Chinese micro-blogs based on heterogeneous graph random walk |
| CN104778213A (en) * | 2015-03-19 | 2015-07-15 | 同济大学 | Social network recommendation method based on random walk |
| CN106055627A (en) * | 2016-05-27 | 2016-10-26 | 西安电子科技大学 | Recognition method of key nodes of social network in topic field |
-
2017
- 2017-03-06 CN CN201710144748.2A patent/CN107273396A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102880728A (en) * | 2012-10-31 | 2013-01-16 | 中国科学院自动化研究所 | Individualized ordering method for video searching results of famous persons |
| CN103530421A (en) * | 2012-11-02 | 2014-01-22 | 中国人民解放军国防科学技术大学 | Micro-blog based event similarity measuring method and system |
| CN103412872A (en) * | 2013-07-08 | 2013-11-27 | 西安交通大学 | Micro-blog social network information recommendation method based on limited node drive |
| CN103838806A (en) * | 2013-10-10 | 2014-06-04 | 哈尔滨工程大学 | Analysis method for subject participation behaviors of user in social network |
| CN104123352A (en) * | 2014-07-10 | 2014-10-29 | 西安理工大学 | Method for measuring influence of users on topic hierarchy for MicroBlog |
| CN104268230A (en) * | 2014-09-28 | 2015-01-07 | 福州大学 | Method for detecting objective points of Chinese micro-blogs based on heterogeneous graph random walk |
| CN104778213A (en) * | 2015-03-19 | 2015-07-15 | 同济大学 | Social network recommendation method based on random walk |
| CN106055627A (en) * | 2016-05-27 | 2016-10-26 | 西安电子科技大学 | Recognition method of key nodes of social network in topic field |
Non-Patent Citations (1)
| Title |
|---|
| 张乐君等: "Research on Online Social Network Information Diffusion Detection Node Selection Algorithm Based on the Random Walk Model", 《JOURNAL OF COMPUTATIONAL AND THEORETICAL NANOSCIENCE》, 1 January 2016 (2016-01-01), pages 971 - 981 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108804689A (en) * | 2018-06-14 | 2018-11-13 | 合肥工业大学 | The label recommendation method of the fusion hidden connection relation of user towards answer platform |
| CN108804689B (en) * | 2018-06-14 | 2020-10-16 | 合肥工业大学 | Question-answering platform-oriented label recommendation method integrating user hidden connection relation |
| CN109213953B (en) * | 2018-08-13 | 2021-07-27 | 复旦大学 | Modeling method of social network multi-information propagation model |
| CN109213953A (en) * | 2018-08-13 | 2019-01-15 | 华东师范大学 | A kind of modeling method of social networks multi information propagation model |
| CN109726297A (en) * | 2018-12-28 | 2019-05-07 | 沈阳航空航天大学 | A kind of two subnetwork node prediction algorithms based on mutual exclusion strategy |
| CN109726297B (en) * | 2018-12-28 | 2022-12-23 | 沈阳航空航天大学 | Bipartite network node prediction algorithm based on mutual exclusion strategy |
| CN109815345A (en) * | 2019-02-25 | 2019-05-28 | 南京大学 | A kind of knowledge mapping embedding grammar based on path |
| CN111723578A (en) * | 2020-06-09 | 2020-09-29 | 平安科技（深圳）有限公司 | Hot spot prediction method and device based on random walk model and computer equipment |
| CN111723578B (en) * | 2020-06-09 | 2023-11-17 | 平安科技（深圳）有限公司 | Hot spot prediction method and device based on random walk model and computer equipment |
| CN111861122A (en) * | 2020-06-18 | 2020-10-30 | 北京航空航天大学 | Social network information credibility evaluation method based on propagation attribute similarity |
| CN111861122B (en) * | 2020-06-18 | 2022-10-18 | 北京航空航天大学 | Social network information credibility evaluation method based on propagation attribute similarity |
| CN112686765A (en) * | 2020-12-09 | 2021-04-20 | 天津大学 | Information propagation rule mining method based on social network |
| CN112765329A (en) * | 2020-12-31 | 2021-05-07 | 清华大学 | Method and system for discovering key nodes of social network |
| CN112765329B (en) * | 2020-12-31 | 2022-07-05 | 清华大学 | Method and system for discovering key nodes of social network |
| CN115185715A (en) * | 2022-09-13 | 2022-10-14 | 深圳市华云中盛科技股份有限公司 | Case popularity diffusion processing method based on social network information |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107273396A (en) | A kind of social network information propagates the system of selection of detection node | |
| Kochkina et al. | All-in-one: Multi-task learning for rumour verification | |
| CN104991956B (en) | Microblogging based on theme probabilistic model is propagated group and is divided and account liveness appraisal procedure | |
| Chen et al. | Battling the internet water army: Detection of hidden paid posters | |
| Del Vicario et al. | News consumption during the Italian referendum: A cross-platform analysis on facebook and twitter | |
| CN103927398A (en) | Microblog hype group discovering method based on maximum frequent item set mining | |
| CN106940732A (en) | A kind of doubtful waterborne troops towards microblogging finds method | |
| Zhao et al. | Chinese underground market jargon analysis based on unsupervised learning | |
| Tamine et al. | Social media-based collaborative information access: Analysis of online crisis-related twitter conversations | |
| CN103838806A (en) | Analysis method for subject participation behaviors of user in social network | |
| Samory et al. | Quotes reveal community structure and interaction dynamics | |
| Singh et al. | Rumour veracity estimation with deep learning for Twitter | |
| Li et al. | Multi-layer network for influence propagation over microblog | |
| Chang et al. | Exploration of a concept screening method in a crowdsourcing environment | |
| Jendoubi et al. | Classification of message spreading in a heterogeneous social network | |
| CN110851684A (en) | Social topic influence identification method and device based on ternary association graph | |
| Zygmunt | Role identification of social networkers | |
| Yong et al. | Rumors detection in sina weibo based on text and user characteristics | |
| Agarwal et al. | Analysis of contextual features’ granularity for fake news detection | |
| Huang et al. | Early detection of fake news based on multiple information features | |
| Li et al. | DeepPick: a deep learning approach to unveil outstanding users with public attainable features | |
| CN110458182A (en) | Based on the matched online vest detection method of similar subgraph | |
| Zhang et al. | Product Information Diffusion in a Social Network and Marketing Implications: A Case Study of Huawei Mobile Phone | |
| Wang et al. | Detecting inactive cyberwarriors from online forums | |
| Zhang et al. | Generating profiles for a lurking user by its followees' social context in microblogs |
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 |