Link Prediction on Hike’s Social Network Social networks are highly dynamic; they grow and change quickly over time through the addition of new edges, signifying the appearance of new interactions in the underlying social structure. The fundamental computational problem underlying social-network evolution is the Link Prediction problem:
Hike is a social platform and predicting links in their network forms the basis for recommending new friends to our users with whom they can possibly start a chat. High-quality recommendations help in strengthening the network by aiding the creation of new social connections between existing users. It also helps in the retention of new users by helping them find friends as they join the platform.
Can you develop an algorithm to predict whether a Hike user will chat another Hike user who is part of his/her phone contact book?
LINK https://datahack.analyticsvidhya.com/contest/hikeathon/
12th place Solution
Research Papers Studied:
https://arxiv.org/pdf/1411.5118.pdf
http://www.leonidzhukov.net/hse/2016/networks/papers/SNDA11.pdf
http://www.leonidzhukov.net/hse/2016/networks/papers/NowellKleinberg07linkprediction.pdf