Graph Mining: Community Detection: Learn by finding answers to the following questions. Can you answer the following questions on Community Detection?

Graph Mining: Community Detection: Learn by finding answers to the following questions. Can you answer the following questions on Community Detection?

What is a community anyway? Describe from your real-world/social understanding of it?

Can you relate it to the Graphs concept in Computer Science?

In your daily life do you interact with a community in a real world? or in your communication with others using technologies?

What is Graph as defined in Computer Science Books/Practice? Can you formally represent a Graph? What are the components of a graph?

Is a whole Graph a Community? or Graphs can be subsets (Subgraph) as well?

What is a Sub-graph anyway?

Can you identify communities in real-world networks? i.e. real world networks that can be represented as graphs?

Give examples of real-world Graphs as well.

Can you detect a community in the tissues or the organs in the human body? Can you represent these as Graphs?

Can you detect a community in Friends on social media with similar interests? i.e. is it a community? Can you represent these/this as Graphs?

Can you detect/define a community for the Neighborhoods where a hurricane will likely pass? i.e. is it a community?

Is it a community: People that are likely to be affected by a contagious disease?

Will the properties of a community (subgraph) be the same like the whole graph (i.e. average properties of the whole graph)? or will it differ? When? How? Why? Examples?

How does the formation of communities affect? i.e. How does the existence or not of communities affect? You can provide generic answers or just explain with examples as well.

Do you think that you are a part/node of communities in online social networks such as Facebook, twitter? If so, how and when you are affected? If so, how and when you can affect others?

Can communities affect rumor spreading or disease spreading or epidemic spreading? Can you also identify how a rumor/disease will spread i.e. from the source i.e. which path will it take, how will it flow, who will get affected?

Is it that the spreading will only affect the nodes/parts of the communities? Do you see any limitations of this thought?

Who else can get affected? Can you predict that?

What is Link prediction? What are Link Prediction Algorithms?

Is it a possibility that current communities can be fine tuned? i.e. remove links/nodes and add new links/nodes? Why, why not? How? Is it the right thing to do? why, why not?

What is the difference between clustering and community detection?

What is thought to be of more real world entitities/geared? Community Detection or Clustering?

What is thought to be more geared towards structural Properties such as Min-Cut?

What is Min-Cut anyway?

For modeling, understanding and analysis purpose - can we consider them the same/similar? Community Detection or Clustering?
Ans: Yes

What is Cluster Analysis i.e. Community Detection process? i.e. what will be your process to identify clusters/communities?

Provide the names of some Community Detection Algorithms. Can you explain them as well? Can you formally define/represent and/or show pictorially what they do and how they work (steps as well)? Check the answer section, after you give it a try. You must have studied them in Computer Networks or related courses/concepts (Data Communication, CCNA, CCNP, Vehicle Routing)

How can you evaluate that the detected communities/clusters are good/great communities/clusters?

Answers:
What is a community?
a subset of nodes within the graph such that connections between the nodes (in the subset) are denser than connections with the rest of the network (i.e. from the subset to the outer nodes)

What is Cluster Analysis i.e. Community Detection process? i.e. what will be your process to identify clusters/communities?
"The process of dividing nodes of a graph into possibly overlapping, subsets, where nodes in each subset are considered related by some similarity measure"

Provide the names of some Community Detection Algorithms
Maximal-Clique Enumeration, K- Spanning Tree, Shared Nearest Neighbor, Highly Connected Components, MinCut, Betweenness Based Algorithms, Louvain Modularity, CNM Algorithm

Resources:
Defining and identifying communities in networks
https://www.pnas.org/content/101/9/2658

Community Structure:
https://en.wikipedia.org/wiki/Community_structure

Graph Clustering:
https://www.csc2.ncsu.edu/faculty/nfsamato/practical-graph-mining-with-R/slides/pdf/Graph_Cluster_Analysis.pdf

Sayed Ahmed

Linkedin: https://ca.linkedin.com/in/sayedjustetc

Blog: http://Bangla.SaLearningSchool.com, http://SitesTree.com
Online and Offline Training: http://Training.SitesTree.com

Affiliate Links:
Hottest Deals on Amazon USA: http://tiny.cc/38lddz

Hottest Deals on Amazon CA: http://tiny.cc/bgnddz

Hottest Deals on Amazon Europe: http://tiny.cc/w4nddz