Detection and Interpretation of Communities in Complex Networks: Practical Methods and Application
read more
Citations
Surprise maximization reveals the community structure of complex networks
Unsupervised Learning and Multipartite Network Models: A Promising Approach for Understanding Traditional Medicine
An O(n2) algorithm for detecting communities of unbalanced sizes in large scale social networks
Countering the social ignorance of 'social' network analysis and data mining with ethnography : a case study of the Singapore blogosphere
References
Data Mining: Practical Machine Learning Tools and Techniques
Community structure in social and biological networks
Fast unfolding of communities in large networks
Finding and evaluating community structure in networks.
Fast unfolding of communities in large networks
Related Papers (5)
Frequently Asked Questions (9)
Q2. What have the authors stated for future works in "Detection and interpretation of communities in complex networks: methods and practical application" ?
The same remark holds for the measures designed to study the significance [ 72,73 ] and topological properties [ 15 ] of the community structure.
Q3. What is the first algorithm to implement a simulated annealing approach?
The NetCarto algorithm [23] (C code available on demand to its authors) implements a simulated annealing approach, which allows it to get very close to the actual optimum, but makes it in turn very slow.
Q4. What does the fact that nodes are very close to indicate?
The fact they are all very close to , including the network value, indicates nodes are very dominantly connected to other nodes from the same communities.
Q5. What is the common application of network modeling?
Network modeling has been used for years in many application fields: biological, social, technological, communication, information (see [1] for a very comprehensive review of applied studies).
Q6. How was the measure extended to weighted links?
The measure was extended to weighted links similarly to what was done for the edge-betweenness, i.e. using a normalization based on the weight of the considered link [48].
Q7. How did Girvan and Newman define their edgebetweenness measure?
Newman proposed extensions of both measures for weighted links [27], by normalizing edgebetweenness with the considered link weight, and by using weights to process the random walker transition probabilities.
Q8. What is the advantage of the pattern-based approach?
Although this is a drawback in the context of complex networks analysis, due to their size, the pattern-based approach still has an interesting advantage: it allows specifying more precisely the internal structure of the communities.
Q9. What are the two properties of link centrality?
There are several definitions for this notion, but link centrality is basically related to two properties: the number of pairs of nodes the link is connecting (directly or not) and how likely these connections are to be used.