Critique of Graph Cluster Randomization: Network Exposure to Multiple Universes

Abstract

In this paper I will be reviewing and evaluating the work of Johan Ugander and Jon Kleinberg (both from Cornell University) and Brian Karrer and Lars Backstrom (both from Facebook), entitled Graph Cluster Randomization: Network Exposure to Multiple Universes, presented at the 2013 International Conference of Knowledge DIscovery and Data Mining. After a brief summary of the paper and its findings, I present the author's background and related previous work - to find that graph cluster randomization proves to be a truly useful method for social networks and has a wide array of future applicability. Subsequenlty I conduct a brief overview of related work and possible improvements, followed by my personal reflections and suggested future work pathways. I finish with a brief conclusion.

Keywords: social network, randomization, A/B testing, network exposure

#Summary

In this paper I will be reviewing and evaluating the work of Johan Ugander and Jon Kleinberg (both from Cornell University) and Brian Karrer and Lars Backstrom (both from Facebook), entitled Graph Cluster Randomization: Network Exposure to Multiple Universes (GCR), presented at the 2013 International Conference of Knowledge DIscovery and Data Mining and featured on Facebook's research page (Ugander 2013).

References

  1. Johan Ugander, Brian Karrer, Lars Backstrom, Jon Kleinberg. Graph cluster randomization. In Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD 13. Association for Computing Machinery (ACM), 2013. Link