All visualizations - louvain and random walk - are stored in the ~/results folder. It is recommended to download the results .pdf files, to better zoom on the individual nodes.

The project is running on Anaconda distribution of Python 3.7. Run the Runner.py File to start the louvain detection, network visualization and random walk.

The aim of this project is to build a tool to identify communities and influencers in a social network. The tool should be able to:

• Load a social graph
• Run community detection and some analytical algorithms
• Visualize the network

• Load the following Facebook dataset (The nodes represent the users and links represent the friendships between users) provided from: http://snap.stanford.edu/data/egonets-Facebook.html

• Implement Louvain method of community detection
• Implement Degree centrality measure
• Implement Random Walk algorithm Perform the following tasks using the previous implementations:
• Identify users’ communities in the Facebook network using Louvain method.
• Identify top 2 users with the highest degree centrality in each community.
• Distribute three messages across the Facebook network using the random walk algorithm, start each walk from a different community using its highest degree centrality user. Terminate the walk when at least one node from each community has been reached.

• Visualize the output of Louvain method by coloring nodes according to their assigned communities.
• Visualize the output of applying Random Walk algorithm, by highlighting the sequence of nodes selected in a path.