In this question, you will be working with two sample social graphs. Each line represents an edge in the network between two nodes separateedd by space.

Link Prediction

For each of the networks do the following, dont't use libraries, implement eveything from scratch:

Remove $20\%$ of the edges randomly

Use the following methods to add the same number of edges you removed

Common Neighbors, Jaccard Similarity, Adamic$/$Adar$,$ Preferential Attachment, simRank$(0\mathrm{.}5),$ simRank$(0\mathrm{.}2)$

a$.$ Evaluate and Demonstrate the precision and recall for each of the methods you tried in each of the networks, which method had the highest performance in each network? Limit simRank to $3$ hops

b$.$ Do you see a link between the degree distribution of the network and the performance of the methods you tried? Illustrate your observations along with the rationale of the difference if any.