NPTEL Social Network Analysis Week 11 Assignment Answers 2024
1. Which network neighborhoods are captured by the second and first-order proximities in the LINE representation method?
- Local, Global
- Global, Local
- Local, Local
- Global, Global
Answer :- For Answers Click Here
2. Which of the following statements about random walk-based methods is true?
- They are slow because they run random walks from a single node simultaneously.
- They are fast because they can run random walks from multiple nodes simultaneously.
- They are only applicable to undirected networks.
- They do not capture local structures in the network.
Answer :- For Answers Click Here
3. Which types of information networks is the “LINE” network embedding method suitable for?
- Undirected only
- Directed only
- Weighted only
- All of the above
Answer :-
4. In the given network based on the LINE proximity system:
- B-E has high second-order proximity, and F-A has high second-order proximity.
- D-F has high second-order proximity, and E-A has high first-order proximity.
- D-F has high first-order proximity, and E-A has high second-order proximity.
- F-E has high first-order proximity, and F-A has high second-order proximity.
Answer :-
5. In a graph convolutional network with a depth of L, when applied to a graph G(V,E), how many learnable weight matrices are required?
- |V|
- L
- |E|*L
- L*L
Answer :-
6. For the link prediction task, which proximity measure should be considered?
- First order
- Second order
Answer :- For Answers Click Here
7. What are the characteristics of the random walks generated in the DeepWalk algorithm?
- Conditioned on some probability
- Uniform
- Variable length
- All of the above
Answer :-
8. For the given graph, identify the depth-first search walk.
- {1,2,3,4,5,6}
- {1,3,6,2,4.5}
- {1,2,5,3,6,4}
- {1,2,4,5,3,6}
Answer :-
9. In the context of the node2vec algorithm, when the return parameter, denoted as ‘p,’ is set to a relatively high value, what effect does this have on the behavior of the random walker in terms of neighborhood exploration?
- it primarily directs the random walker to focus on the local neighborhood, emphasizing nearby nodes.
- It predominantly encourages the random walker to venture into the global neighborhood, broadening the scope of exploration.
- It facilitates a balanced exploration between global and local neighborhoods, treating them equally important.
- It effectively inhibits the random walker from engaging in any form of neighborhood exploration whatsoever.
Answer :-
10. In the context of the R-GCN variant of graph convolutional networks, for which specific aspects are distinct weight matrices learned?
- Communities
- Rankings
- Cliques
- Relations
Answer :- For Answers Click Here
