Download as pdf or txt
Download as pdf or txt
You are on page 1of 3

Ramaiah Institute of Technology, Bangalore – 560054

Department of Master of Computer Applications


MCAE24 Social Network Analysis
Question Bank
Unit III
1. Discuss the applications of link analysis. (110)
2. Describe signed networks. (110)
3. Explain the balance theory of undirected signed networks with example. (111-
112)
4. Explain strength of a tie with an example. Discuss its significance with real-world
application. (114)
5. Based on the network shown below, calculate the neighbourhood overlap of (A,
M) and (A, E). Write your observation on the results. (115)

6. Explain the formation of triadic closure. (115-116)


7. Consider the network given in the Figure. Determine how likely the following nodes
are to form a link. (117)
i) Nodes 2 and 6 to form a link, leading to a triadic closure among 1, 2 and 6.
ii) Nodes 3 and 4 to form a link, leading to a triadic closure among 2, 3 and 4.

8. Discuss the importance of local bridges/weak ties. (118)


9. Define PageRank. Explain the procedure to calculate PageRank in a random walk.
(OR)
How do you measure probability of a random walk from node A to node B
following a meta path P? (119-121)
10. Discuss the advantages and disadvantages of PageRank. (125)
11. Define link prediction, discuss the applications of link prediction and temporal
changes in a network.
12. Describe the confusion matrix based evaluation metrics. Calculate the various
confusion matrix based scores considering the following samples available for the
actual links (AL) and predicted links (PL) sets. Write the inference from the values
obtained (12)

Alp = Actual Positive Links ={(a,b),(b,c),(c,d)}


ALn = Actual Negative Links ={(a,d),(a,c),(b,d)}
PLp = Predicted Positive Links ={(a,b),(b,c),(b,d)}
PLn = Predicted Negative Links ={(a,d),(a,c),(c,d)}

UNIT IV
1. Discuss the various applications of Community detection algorithms. (146)
2. Describe different types of communities with neat diagrams. (148)
3. Explain node-centric community detection. (151-153)
4. Describe the following: (151-153)
a. K-cliques
b. K-clan
c. K-club
d. K-plex
e. K-core
(OR)

5. Compare and contrast the following:


a. K-cliques, K-clan, and K-club
b. K-plex and K-core
6. Discuss the drawback in node-centric community detection. Explain how it can be
resolved in modularity and community detection. (154-155)
7. Consider the following network. (156-157)

Calculate the difference between the actual number of edges and expected number
of edges. Also find the modularity values of the community structures.

8. Explain Fast Greedy algorithm to detect the community using modularity


maximization.
9. Explain Louvain algorithm to detect the community using modularity
maximization.
10. Describe Clique Percolation approach to detect overlapping communities. Solve
for the given network. 165-166
11. Describe Link Partition approach to detect overlapping communities. Solve for the
given network. 167-168

You might also like