Proceedings of the 2013 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining 2013
DOI: 10.1145/2492517.2500231
|View full text |Cite
|
Sign up to set email alerts
|

Different approaches to community evolution prediction in blogosphere

Abstract: Predicting the future direction of community evolution is a problem with high theoretical and practical significance. It allows to determine which characteristics describing communities have importance from the point of view of their future behaviour. Knowledge about the probable future career of the community aids in the decision concerning investing in contact with members of a given community and carrying out actions to achieve a key position in it. It also allows to determine effective ways of forming opin… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

1
26
0

Year Published

2013
2013
2021
2021

Publication Types

Select...
4
3
1

Relationship

1
7

Authors

Journals

citations
Cited by 35 publications
(29 citation statements)
references
References 23 publications
1
26
0
Order By: Relevance
“…The reason for selected order is that some events, such as addition or deletion mean small change for groups. Moreover, some events cannot coexist with other ones (described in [17]) and position in order of such events is meaningless (such as the decay event). ) of all members of considered groups (see Section 5.3).…”
Section: Predicting Group Evolution Using Sgci Results and The Notionmentioning
confidence: 99%
“…The reason for selected order is that some events, such as addition or deletion mean small change for groups. Moreover, some events cannot coexist with other ones (described in [17]) and position in order of such events is meaningless (such as the decay event). ) of all members of considered groups (see Section 5.3).…”
Section: Predicting Group Evolution Using Sgci Results and The Notionmentioning
confidence: 99%
“…. , 20 and set a 60% overlap between frames; equivalent overlap has also been used in similar work [8]. Generally, the higher the percentage of overlap the smoother the evolution of communities is between consecutive frames.…”
Section: Methodsmentioning
confidence: 99%
“…Then community features, mostly related to the link structure, such as the size, cohesion, density, are computed. For instance in [8], [9], and [4] between one and five features have been computed. Those features are then used in the training of classifiers, including Support Vector Machines, Decision Trees, Naive Bayes, as well as ensemble methods such as Adaboost and Random Forests (see [4], [8], [17]).…”
Section: Related Workmentioning
confidence: 99%
“…For example, Gliwa et al ( 2013 ) predict community evolution (growing shrinking, splitting, merging and dissolving) from network properties (centralization, density, size and cohesion). Their focus is on community or group evolution rather than the evolutionary changes in the entire network and the relations among the changes of individual nodes.…”
Section: Network Evolutionmentioning
confidence: 99%