Using com-Youtube as dataset, with 1.134.890 nodes and 2.987.624 edges. Our goal is find the k users with highest influence, adopting local search (random-restart hill climbing) and two heuristics: number of friends and number of groups.
Implemented in Python 3.
Parameters:
- k: number of influencers to find
- heuristic_id: FRIENDS or GROUPS to choose which heuristic function to use
Running:
cd [...]/influencer_search/src
python main.py