CN108170842A - Hot microblog topic source tracing method based on tripartite graph model - Google Patents
Hot microblog topic source tracing method based on tripartite graph model Download PDFInfo
- Publication number
- CN108170842A CN108170842A CN201810040609.XA CN201810040609A CN108170842A CN 108170842 A CN108170842 A CN 108170842A CN 201810040609 A CN201810040609 A CN 201810040609A CN 108170842 A CN108170842 A CN 108170842A
- Authority
- CN
- China
- Prior art keywords
- user
- message
- path
- topic
- represent
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Fuzzy Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to social networks much-talked-about topic analysis field, more particularly to a kind of hot microblog topic source tracing method based on tripartite graph model, including:The application programming interface API provided using web crawlers or various social softwares obtains data;Propagation path model reduction is carried out to single Twitter message;Based on the propagation path reduction of single Twitter message, " message pathway user " topic tripartite graph is established;Message, path and the sequence of user's scoring sequence to being calculated according to topic tripartite graph, will be in the top N number of as much-talked-about topic source point sequence;The present invention is in influence of the limitation for node self attributes and topic independence and semantic technology to Source Tracing, HITS ranking thoughts are introduced in tripartite graph, in view of message and user influence each other relationship while, improve sort algorithm, improve the accuracy of algorithm.
Description
Technical field
The present invention relates to social networks much-talked-about topic analysis field, more particularly to a kind of microblogging heat based on tripartite graph model
Point topic source tracing method.
Background technology
Universal with social software application With the fast development of internet, online social software has been increasingly becoming people day
One of often exchange, most important means of communications and entertainment.One of the application product of microblogging as the internet new era, by it
The features such as sending the documents conveniently, commenting on free and unrelated own identification, is just grown rapidly in a short time.However, just because of
Its freedom sent the documents with comment, the public sentiment that much-talked-about topic and focus incident in micro blog network are caused are also more and more.
This just proposes new challenge to the management and control of public sentiment in online social networks.
In recent years, key message in topic communication network under micro blog network is increasingly becoming to the searching of key user related
The hot spot of expert's research.At this stage, the method that microblog topic is traced to the source is roughly divided into following several:Point based on content of text
It analyses, based on the analysis of user node influence power, the analysis based on communication network structure propagation tree in complex network.Wherein, it is based on
The analysis of content of text is mainly found out wherein similar to other message according to Message-text content similarities each under same topic
Highest several are spent, determines topic source.As when state China, Zhou Bin et al. exists《A kind of method that microblogging event source is found》
In, based on content of text, deliver the linking relationship between time and microblogging and find out time source;Based on user node influence power
Analysis mainly sorts to analyze, such as Cesar Henrique Comin et al. using the influence power of participating user in a network
《Identifying the starting point of a spreading process in complex networks》
In, by a kind of improved centrad measuring method, verified on ER networks and scales-free network;Based on communication network
Structure propagates the analysis of tree, as Sadikov et al. exists《Correcting for missing data in information
cascades》In for multi-data source, K tree-models are constructed, so as to show the researching value for tracing to the source and restoring propagation path.
In the correlative study that more than topic is traced to the source, laid particular emphasis on from different angles to the critical message in topic network
Or key user's node is excavated or sets about from Message-text content or participated under node influence power from complex network
Hand, and have ignored the relationship that influences each other between the propagation of topic and participating user.It is crucial however in practical topic network
Propagation path plays a crucial role the propagation of topic with key user.
Invention content
Against the above deficiency, the present invention is directed in research of tracing to the source at present and excessively stresses to have ignored much-talked-about topic in a certain respect
Propagation and participating user topic spread progradation the problem of, a kind of microblogging based on tripartite graph model of present invention proposition
Much-talked-about topic source tracing method, such as Fig. 1, including:
S1, the application programming interface API provided using web crawlers or various social softwares obtain data
It takes;
S2, propagation path model reduction is carried out to single Twitter message;
S3, based on the reduction of the propagation path of single Twitter message, establish " message-path-user " topic tripartite graph;
S4, the message to being calculated according to topic tripartite graph, path and the sequence of user's scoring sequence, will be in the top
It is N number of to be used as much-talked-about topic source point sequence.
Preferably, using the application programming interface API that web crawlers or various social softwares provide to data into
Row obtains, and specifically includes:
S11, data acquisition by the higher multiple messages of participation under Sina weibo webpage capture much-talked-about topic and disappear
The participating user of breath, the level-one forwarding number under userspersonal information, number of reviews, user's bean vermicelli number and user pay close attention to number
Mesh;
S12, simple data cleansing, data cleansing include deleting duplicated data, arrange invalid node etc..
Preferably, single Twitter message propagation path model restores, and specifically includes:
S21, tree-model is propagated according to the transmission of news path construction of acquisition, it is contemplated that comment user is impacted use
Family is without having secondary transmission capacity, so regarding the round of root node to each forwarding leaf node as one disappear
Propagation path is ceased, then can obtain message pathway collection and be combined into
S22, user message propagation drive is obtained, the message of user is propagated drive and is defined as:
Wherein, vk,miRepresent the user node v in i-th of messagek, ptRepresent t paths, Ω represents the propagation of message M
Number of paths, drive (vk,mi) be user message propagate drive,Represent user node vkUnder turn
Send out number,Represent user node vkUnder comment number.
Preferably, " message-path-user " topic tripartite graph is established, is specifically included:
S31, structure " message-path-user " topic tripartite graph model, according to same user may to different messages into
Row forwarding and comment, propagating multiple messages has overlapping, so as to form topic whole network, builds topic three based on this
Figure, is embodied as:
G=(M ∪ P ∪ V, A ∪ B);
Wherein, G represents three step graph model of topic, M={ M1,M2,M3,...,MiFor massage set, P={ P1,P2,
P3,...,PiFor message propagation path set, V={ V1,V2,V3,...,ViFor participating user gather, A for message-path it
Between weight matrix, weight matrixs of the B between path-user.
S32, user's topic propagate the calculating of drive, and same user may simultaneously participate in different under this much-talked-about topic
Message, defining user's topic propagation drive is:
Wherein, drive (vk,mi) for user message propagate drive, vk,miRepresent the user node in i-th of message
vk。
S33, hypertext topic search (Hypertext-Induced Topic Search, HITS) ranking thought is introduced,
It is excavated, i.e., message, path and user is dug using key element in loop iteration marking mechanism dialogue topic tripartite graph
Pick, wherein loop iteration marking mechanism include positive scoring process and reversed scoring process.
The present invention is true according to the propagation out-degree of node first with dendrogram model construction message propagation path-propagation tree
The propagation drive of fixed each node;Then build message-path-user topic tripartite graph, further portray message, path and
Influence relationship between user solves the problems, such as to ignore message and the correlation of user in the studies above;Finally, for node
Self attributes and the influence of topic independence and the limitation of semantic technology to Source Tracing, introduce HITS rankings in tripartite graph
Thought according to weight matrix of each set of node initial value vector sum between them, is calculated each using iterative cycles marking mechanism
The scores vector of set of node;Each set of node score value sequence is finally generated according to score.The present invention is in view of message and user's phase
While mutually influencing relationship, sort algorithm is improved, improves the accuracy of algorithm.
Description of the drawings
Fig. 1 is the overview flow chart of the hot microblog topic source tracing method the present invention is based on tripartite graph model;
Fig. 2 is the message propagation path model also artwork of the present invention;
Fig. 3 is the topic tripartite graph model of the present invention;
Fig. 4 is the cycle marking iterative manner illustraton of model of the present invention.
Specific embodiment
This patent in view of message and user influence each other relationship while, improve sort algorithm, propose that one kind is based on
The hot microblog topic source tracing method of tripartite graph model, such as Fig. 1, including:
S1, the application programming interface API provided using web crawlers or various social softwares obtain data
It takes;
S2, propagation path model reduction is carried out to single Twitter message;
S3, based on the reduction of the propagation path of single Twitter message, establish " message-path-user " topic tripartite graph;
S4, the message to being calculated according to topic tripartite graph, path and the sequence of user's scoring sequence, will be in the top
It is N number of to be used as much-talked-about topic source point sequence.
Preferably, using the application programming interface API that web crawlers or various social softwares provide to data into
Row obtains, and specifically includes:
S11, data acquisition by the higher multiple messages of participation under Sina weibo webpage capture much-talked-about topic and disappear
The participating user of breath, the level-one forwarding number under userspersonal information, number of reviews, user's bean vermicelli number and user pay close attention to number
Mesh;
S12, simple data cleansing, data cleansing includes deleting duplicated data, arranges invalid node, such as repeatedly forwarding
Node with commenting on same user is considered as invalid node, only sees Once dissemination as.
Preferably, single Twitter message propagation path model restores, such as Fig. 2, including:Topic whole network is divided, from list
Message is started with, and the topological structure for comment being forwarded to be formed using participating user isolates the propagation path of forwarding user, introduces tree
Each participating user is regarded as a node by shape model, and forwarding and comment relationship regards side as each time, builds participating user
Tree-model is propagated, and determines that the message of user propagates drive, specially:
S21, tree-model is propagated according to the transmission of news path construction of acquisition, it is contemplated that comment user is impacted use
Family is without having secondary transmission capacity, so regarding the round of root node to each forwarding leaf node as one disappear
Propagation path is ceased, then can obtain message pathway collection and be combined into
S22, user message propagation drive is obtained, the message of user is propagated drive and is defined as:
Wherein, vkRepresent user node, vk,miRepresent the user node v in i-th of messagek, ptRepresent t paths, Ω
Represent the propagation path quantity of message M, drive (vk,mi) be user message propagate drive,It represents
User node vkUnder forwarding number,Represent user node vkUnder comment number.
Preferably, " message-path-user " topic tripartite graph is established, first, on the basis of S2, is propagated and set with single
Based on form, such as Fig. 3, for example, for user V2, can be with passage path P1Spread news M1, can be with passage path P3Propagate it
His message, all message, path and user are merged together, and more message forwarding relational network is reconstructed, is formed more
Message, multipath propagation network mode build topic tripartite graph model;Then the topic for calculating participating user propagates drive.
Then, in order to accurately calculate the criticality of outbound message and participating user's node, HITS thoughts is introduced, are made in topic tripartite graph
With loop iteration marking mechanism, critical message, critical path, key user are excavated, specifically included:
S31, structure " message-path-user " topic tripartite graph model, according to same user may to different messages into
Row forwarding and comment, propagating multiple messages has overlapping, so as to form topic whole network, builds topic three based on this
Figure, is embodied as:
G=(M ∪ P ∪ V, A ∪ B);
Wherein, G represents three step graph model of topic, M={ M1,M2,M3,...,Mi... } and for massage set, P={ P1,P2,
P3,...,Pi... } and for message propagation path set, V={ V1,V2,V3,...,Vi... } gather for participating user, A is disappears
Weight matrix between breath-path, weight matrixs of the B between path-user.
S32, user's topic propagate the calculating of drive, and same user may simultaneously participate in different under this much-talked-about topic
Message, defining user's topic propagation drive is:
Wherein, drive (vk,mi) for user message propagate drive, vk,miRepresent the user node in i-th of message.
S33, HITS ranking thoughts are introduced, is dug using key element in loop iteration marking mechanism dialogue topic tripartite graph
Pick, i.e., excavate critical message, critical path and key user, and wherein loop iteration marking mechanism was given a mark including forward direction
Journey and reversed scoring process.
S331, message initial score vector X is calculated(0), path initial score vector Y(0), user's initial score vector W(0);
Wherein, αiRepresent be message propagate range, i.e. message MiCorresponding active path quantity;βjRepresent be
Path pjCorresponding node vkNumber, γkThat represent is user node vkTopic propagate drive, subscript T represent transposition
Matrix.
The weight matrix between weight matrix and path-user between S332, calculating message-path:
Weight matrix between message-path is expressed as:
Message-routine weight value matrix:
Path-message weight matrix:APM=AMP T;
Wherein, if having propagation path P, a in message Mi,j=1, otherwise, ai,j=0;It meanwhile can be AMPIt regards as
Transfer matrix between information node M and path node P, subscript T represent transposed matrix.
Path-user weight matrix:
User-routine weight value matrix:
Wherein, if propagation path PjIn include user node vk, then bk,j=1, otherwise bk,j=0,It is that user initially divides
It is worth vector W(0)K-th of element,It is path initial score Y(0)J-th of element.
The score value vector that S333, the massage set in tripartite graph, set of paths, user gather is updated:
Using loop iteration marking mechanism, forward direction-recycled back marking is carried out in tripartite graph, until each set score value
Convergence, i.e. less than one threshold epsilon.
Such as Fig. 4, positive scoring process includes scoring process 1 and process 2, and the element in message scoring vector X is each message
Node { M1,M2,...,Mi... scoring, the element in path scoring vector Y is each path node { P1,P2,...,Pj...}
Scoring, the element that user scores in vector W is each user node { V1,V2,...,Vk... scoring, process 1 be message-road
The scoring process of diameter, process 1 is according to n-th circulation message scoring vector X(n), message-routine weight value matrix AMPIt is initial with path
Score vector Y(0)Obtain (n+1)th circulating path scoring vector Y(n+1), scoring process of the process 2 for path-user, process 2
According to (n+1)th circulating path scoring vector Y(n+1), path-user weight matrix BPVWith user's initial score vector W(0)Obtain
N+1 cycle user's scoring vector W(n+1), specially:
The scoring process in message-path:Y(n+1)=μ AMPX(n)+(1-μ)Y(0);
The scoring process of path-user:W(n+1)=μ BPVY(n+1)+(1-μ)W(0);
Such as Fig. 4, reversed scoring process includes scoring process 3 and process 4, and wherein process 3 is the marking in user-path
Journey, user's scoring vector W that process 3 is obtained according to process 2(n+1), user-routine weight value matrix BVPWith path initial score to
Measure Y(0)The path scoring vector Y obtained in renewal process 1(n+1), process 4 is the scoring process of path-message, and process 4 is according to mistake
The path scoring vector Y obtained in journey 3(n+1), path-message weight matrix APMWith message initial score vector X(0)Obtain n-th+
1 circulation message scoring vector X(n+1), specially:
The scoring process in user-path:Y(n+1)=μ BVPW(n+1)+(1-μ)Y(0);
The scoring process of path-message:X(n+1)=μ APMY(n+1)+(1-μ)X(0);
Wherein, μ is proportion adjustable parameter, in order to avoid the score value of each set infinitely increases, is needed after the update of each iteration
Score value vector is normalized, make ∑iXi=1, ∑jYj=1, ∑kWk=1;When primary positive scoring process and primary anti-
After scoring process, X can be obtained(n+1), Y(n+1)And W(n+1), judge the score value of current each sequence and preceding primary each sequence
Score value difference whether be less than convergence threshold, if then terminating to calculate, obtain the collection of critical message, key user and critical path
It closes, otherwise then enables n=n+1, continue forward direction-recycled back marking.
Preferably, score is calculated according to topic tripartite graph in S3 and sorted, specifically included:
On the basis of S3, critical message, critical path, key user are ranked up from high to low respectively, generation point
Value sequence collection ordering by merging, using N number of sequence in the top as much-talked-about topic source point sequence;
The model is that a kind of hot microblog topic is traced to the source model, on the one hand using tripartite graph can consider comprehensively each node it
Between correlation advantage, by message, propagation path and participating user combine consideration, on the other hand, to HITS rank algorithms
It is improved, calculates the importance score of each key element, key element in topic communication process is excavated;
One of ordinary skill in the art will appreciate that all or part of step in the various methods of above-described embodiment is can
It is completed with instructing relevant hardware by program, which can be stored in a computer readable storage medium, storage
Medium can include:ROM, RAM, disk or CD etc..
Embodiment provided above has carried out the object, technical solutions and advantages of the present invention further detailed description, institute
It should be understood that embodiment provided above is only the preferred embodiment of the present invention, be not intended to limit the invention, it is all
Any modification, equivalent substitution, improvement and etc. made for the present invention within the spirit and principles in the present invention, should be included in the present invention
Protection domain within.
Claims (10)
1. the hot microblog topic source tracing method based on tripartite graph model, which is characterized in that including:
S1, the application programming interface API provided using web crawlers or various social softwares obtain data;
S2, propagation path model reduction is carried out to single Twitter message;
S3, based on the reduction of the propagation path of single Twitter message, establish " message-path-user " topic tripartite graph;
S4, the message to being calculated according to topic tripartite graph, path and the sequence of user's scoring sequence, will be in the top N number of
As much-talked-about topic source point sequence.
2. the hot microblog topic source tracing method according to claim 1 based on tripartite graph model, which is characterized in that described
Data obtain using the application programming interface API that web crawlers or various social softwares provide and are included:
S11, data acquisition pass through the higher multiple messages of participation under Sina weibo webpage capture much-talked-about topic and message
Participating user, the level-one forwarding number under userspersonal information, number of reviews, user's bean vermicelli number and user pay close attention to number;
S12, simple data cleansing, data cleansing include deleting duplicated data, arrange invalid node etc..
3. the hot microblog topic source tracing method according to claim 1 based on tripartite graph model, which is characterized in that described
Include to obtaining the reduction of single Twitter message propagation path model:
S21, tree-model is propagated according to the transmission of news path construction of acquisition, it is contemplated that comment user be affected user and
Do not have secondary transmission capacity, so regarding the round of root node to each forwarding leaf node as a piece of news pass
Path is broadcast, then can obtain message pathway collection and be combined into
S22, user message propagation drive is obtained, the message of user is propagated drive and is defined as:
Wherein, vk,miRepresent the user node v in i-th of messagek, ptRepresent t paths, Ω represents the propagation path of message M
Quantity, drive (vk,mi) be user message propagate drive,Represent user vkUnder forwarding number,Represent user vkUnder comment number.
4. the hot microblog topic source tracing method according to claim 1 based on tripartite graph model, which is characterized in that described
" message-path-user " topic tripartite graph is established to include:
S31, structure " message-path-user " topic tripartite graph model, may turn different messages according to same user
Hair and comment, propagating multiple messages has overlapping, so as to form topic whole network, builds topic tripartite graph based on this,
It is specific to represent to include:
G=(M ∪ P ∪ V, A ∪ B);
S32, same user may simultaneously participate in different messages under this much-talked-about topic, define user's topic and propagate drive
For:
S33, hypertext topic search HITS ranking thoughts are introduced, using crucial in loop iteration marking mechanism dialogue topic tripartite graph
Element is excavated;
Wherein, G represents three step graph model of topic, and M is massage set, and P is message propagation path set, and V gathers for participating user,
Weight matrixs of the A between message-path, weight matrixs of the B between path-user, drive (vk,mi) be user message
Propagate drive, vk,miRepresent the user node v in i-th of messagek。
5. the hot microblog topic source tracing method according to claim 4 based on tripartite graph model, which is characterized in that step
Using key element in loop iteration marking mechanism dialogue topic tripartite graph excavate in S33 and include:
S331, message initial score vector X is calculated(0), path initial score vector Y(0)With user's initial score vector W(0);
The weight matrix between weight matrix and path-user between S332, calculating message-path;
S333, using loop iteration give a mark mechanism in tripartite graph massage set, set of paths, user gather score value vector
It is updated, wherein loop iteration marking mechanism includes positive scoring process and reversed scoring process.
6. the hot microblog topic source tracing method according to claim 5 based on tripartite graph model, which is characterized in that step
Message initial score vector X in S331(0), path initial score vector Y(0)With user's initial score vector W(0)Calculating include:
Wherein, αiRepresent be message propagate range, i.e. message MiCorresponding active path quantity;βjThat represent is path pj
Corresponding node vkNumber, γkThat represent is user node vkTopic propagate drive, subscript T represent transposed matrix.
7. the hot microblog topic source tracing method according to claim 4 based on tripartite graph model, which is characterized in that described
Weight matrix expression between message-path includes:
Message-routine weight value matrix is:
Path-message weight matrix is:
Wherein, if having propagation path P, a in message Mi,j=1, otherwise, ai,j=0;It meanwhile can be AMPRegard message as
Transfer matrix between node M and path node P, βjThat represent is path pjCorresponding node vkNumber, subscript T represent
Transposed matrix.
8. the hot microblog topic source tracing method according to claim 4 based on tripartite graph model, which is characterized in that described
Weight matrix expression between path-user includes:
Path-user weight matrix is:
User-routine weight value matrix is:
Wherein, if propagation path PjIn include user node vk, then bk,j=1, otherwise bk,j=0,User's initial value to
Measure W(0)K-th of element,It is path initial score Y(0)J-th of element.
9. the hot microblog topic source tracing method according to claim 4 based on tripartite graph model, which is characterized in that described
Positive scoring process includes:
Y(n+1)=μ AMPX(n)+(1-μ)Y(0);
W(n+1)=μ BPVY(n+1)+(1-μ)W(0);
Wherein, μ be proportion adjustable parameter, Y(n+1)Represent the path scoring vector of (n+1)th cycle, W(n+1)It represents to follow for (n+1)th time
User's scoring vector of ring, AMPRepresent message-routine weight value matrix, BPVRepresent path-user weight matrix.
10. the hot microblog topic source tracing method according to claim 4 based on tripartite graph model, which is characterized in that institute
Reversed scoring process is stated to include:
Y(n+1)=μ BVPW(n+1)+(1-μ)Y(0);
X(n+1)=μ APMY(n+1)+(1-μ)X(0);
Wherein, μ be proportion adjustable parameter, Y(n+1)Represent the path scoring vector of (n+1)th cycle, X(n+1)It represents to follow for (n+1)th time
The message scoring vector of ring, BVPRepresent user-routine weight value matrix, APMRepresent path-message weight matrix.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810040609.XA CN108170842B (en) | 2018-01-16 | 2018-01-16 | Microblog hot topic tracing method based on three-part graph model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810040609.XA CN108170842B (en) | 2018-01-16 | 2018-01-16 | Microblog hot topic tracing method based on three-part graph model |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108170842A true CN108170842A (en) | 2018-06-15 |
CN108170842B CN108170842B (en) | 2021-12-14 |
Family
ID=62514924
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810040609.XA Active CN108170842B (en) | 2018-01-16 | 2018-01-16 | Microblog hot topic tracing method based on three-part graph model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108170842B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110070410A (en) * | 2019-03-07 | 2019-07-30 | 特斯联(北京)科技有限公司 | A kind of population social activity analysis method and system based on big data |
CN110837608A (en) * | 2019-11-07 | 2020-02-25 | 中科天玑数据科技股份有限公司 | Public opinion topic propagation path analysis system and method based on multi-source data |
CN110851684A (en) * | 2019-11-12 | 2020-02-28 | 重庆邮电大学 | Social topic influence identification method and device based on ternary association graph |
CN111523888A (en) * | 2020-04-16 | 2020-08-11 | 武汉有牛科技有限公司 | On-chain data and information traceability system based on block chain technology |
CN112256756A (en) * | 2020-10-22 | 2021-01-22 | 重庆邮电大学 | Influence discovery method based on ternary association diagram and knowledge representation |
CN112508726A (en) * | 2020-12-25 | 2021-03-16 | 东北电力大学 | False public opinion identification system based on information spreading characteristics and processing method thereof |
CN116992517A (en) * | 2023-09-28 | 2023-11-03 | 山东华云三维科技有限公司 | Collaborative modeling method, server and terminal for three-dimensional CAD model |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103177382A (en) * | 2013-03-19 | 2013-06-26 | 武汉大学 | Detection method for key propagation path and central node on microblog platform |
CN104915438A (en) * | 2015-06-25 | 2015-09-16 | 西安交通大学 | Method for acquiring PCU association data in specific topic microblogs |
CN106503230A (en) * | 2016-11-07 | 2017-03-15 | 贾志娟 | A kind of microblogging based on complex network propagates source tracing method |
CN107222381A (en) * | 2016-03-21 | 2017-09-29 | 北大方正集团有限公司 | The propagation path of microblog data determines method and apparatus |
-
2018
- 2018-01-16 CN CN201810040609.XA patent/CN108170842B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103177382A (en) * | 2013-03-19 | 2013-06-26 | 武汉大学 | Detection method for key propagation path and central node on microblog platform |
CN104915438A (en) * | 2015-06-25 | 2015-09-16 | 西安交通大学 | Method for acquiring PCU association data in specific topic microblogs |
CN107222381A (en) * | 2016-03-21 | 2017-09-29 | 北大方正集团有限公司 | The propagation path of microblog data determines method and apparatus |
CN106503230A (en) * | 2016-11-07 | 2017-03-15 | 贾志娟 | A kind of microblogging based on complex network propagates source tracing method |
Non-Patent Citations (1)
Title |
---|
张跃伟: ""基于微博客话题的热点预测及传播溯源"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110070410A (en) * | 2019-03-07 | 2019-07-30 | 特斯联(北京)科技有限公司 | A kind of population social activity analysis method and system based on big data |
CN110837608A (en) * | 2019-11-07 | 2020-02-25 | 中科天玑数据科技股份有限公司 | Public opinion topic propagation path analysis system and method based on multi-source data |
CN110837608B (en) * | 2019-11-07 | 2024-04-12 | 中科天玑数据科技股份有限公司 | Public opinion topic propagation path analysis system and method based on multi-source data |
CN110851684B (en) * | 2019-11-12 | 2022-10-04 | 重庆邮电大学 | Social topic influence recognition method and device based on ternary association graph |
CN110851684A (en) * | 2019-11-12 | 2020-02-28 | 重庆邮电大学 | Social topic influence identification method and device based on ternary association graph |
CN111523888B (en) * | 2020-04-16 | 2023-09-05 | 武汉有牛科技有限公司 | On-chain data and information tracing system based on block chain technology |
CN111523888A (en) * | 2020-04-16 | 2020-08-11 | 武汉有牛科技有限公司 | On-chain data and information traceability system based on block chain technology |
CN112256756B (en) * | 2020-10-22 | 2022-09-23 | 重庆邮电大学 | Influence discovery method based on ternary association diagram and knowledge representation |
CN112256756A (en) * | 2020-10-22 | 2021-01-22 | 重庆邮电大学 | Influence discovery method based on ternary association diagram and knowledge representation |
CN112508726B (en) * | 2020-12-25 | 2022-04-19 | 东北电力大学 | False public opinion identification system based on information spreading characteristics and processing method thereof |
CN112508726A (en) * | 2020-12-25 | 2021-03-16 | 东北电力大学 | False public opinion identification system based on information spreading characteristics and processing method thereof |
CN116992517A (en) * | 2023-09-28 | 2023-11-03 | 山东华云三维科技有限公司 | Collaborative modeling method, server and terminal for three-dimensional CAD model |
CN116992517B (en) * | 2023-09-28 | 2023-12-26 | 山东华云三维科技有限公司 | Collaborative modeling method, server and terminal for three-dimensional CAD model |
Also Published As
Publication number | Publication date |
---|---|
CN108170842B (en) | 2021-12-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108170842A (en) | Hot microblog topic source tracing method based on tripartite graph model | |
CN103064917B (en) | The high-impact customer group of a kind of specific tendency towards microblogging finds method | |
Hoang et al. | Virality and susceptibility in information diffusions | |
CN111222029A (en) | Method for selecting key nodes in network public opinion information dissemination | |
CN103279887B (en) | A kind of microblogging based on information theory propagates visual analysis method | |
CN104239385A (en) | Method for estimating relationships between topics, and system | |
CN106991617B (en) | Microblog social relationship extraction algorithm based on information propagation | |
CN103530428B (en) | A kind of put into practice the colleague of technical ability similarity based on developer and recommend method | |
CN108564479B (en) | System and method for analyzing hot topic propagation trend based on hidden link | |
CN106503230A (en) | A kind of microblogging based on complex network propagates source tracing method | |
Zhang et al. | Mechanism design for finding experts using locally constructed social referral web | |
Zhang et al. | Research on the information dissemination mechanisms of weibo in scale-free networks | |
Tong et al. | Dynamic analysis of the IFCD rumor propagation model under stochastic disturbance on heterogeneous networks | |
CN109816544B (en) | Information propagation model realization method and device based on contact probability | |
CN108520337B (en) | Riadry risk assessment method based on network risk entropy difference | |
CN110851684B (en) | Social topic influence recognition method and device based on ternary association graph | |
CN112269945B (en) | Information propagation prediction method based on rumor splitting rumor promotion and three-way cognitive game | |
Liu et al. | Nonlinear dynamic information propagation mathematic modeling and analysis based on microblog social network | |
Fang et al. | Social network public opinion research based on S-SEIR epidemic model | |
Li et al. | Multi-layer network for influence propagation over microblog | |
Wang et al. | Influence maximization by leveraging the crowdsensing data in information diffusion network | |
De Choudhury et al. | Dynamic prediction of communication flow using social context | |
Fu et al. | Preferential information dynamics model for online social networks | |
CN112256756B (en) | Influence discovery method based on ternary association diagram and knowledge representation | |
CN112712210B (en) | Sudden topic transmission scale prediction method, system, processing terminal and medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |