Introduction

The Cuban contact-tracing detection system set up in 1986 allowed the reconstruction and analysis of the sexual network underlying the HIV/AIDS epidemic (5,389 vertices and 4,073 edges, giant component of 2,386 nodes and 3,168 edges), shedding light onto the spread of HIV and the role of contact-tracing. I had the unique opportunity of working on this data set with my colleagues Stéphan Clémençon, Hector De Arazoza and Viet Chi Tran.

Publications

We published our results in several papers:

Methodological papers

We published three papers on the graphical tools that helped us to understand the data set:

1. Visual Mining of Epidemic Networks, Stéphan Clémençon, Hector De Arazoza, Fabrice Rossi and Viet-Chi Tran, in Advances in Computational Intelligence (Proceedings of the International Work Conference on Artificial Neural Networks, IWANN 2011), edited by Joan Cabestany, Ignacio Rojas and Gonzalo Joya, volume 6692, pages 276-283, Torremolinos (Spain), June 2011;
2. Analyse exploratoire d'un graphe : le cas de la contamination par le VIH à Cuba, Fabrice Rossi, Stéphan Clémençon, Hector De Arazoza and Viet-Chi Tran, in Congrès SMAI 2011, 5ème Biennale Française des Mathématiques Appliquées, Guidel (France), May 2011;
3. Hierarchical clustering for graph visualization, Stéphan Clémençon, Hector De Arazoza, Fabrice Rossi and Viet-Chi Tran, in Proceedings of XVIIIth European Symposium on Artificial Neural Networks (ESANN 2011), pages 227-232, Bruges (Belgium), April 2011.

Analysis paper

We published a very detailed analysis of the data set in A statistical network analysis of the HIV/AIDS epidemics in Cuba, Stéphan Clémençon, Hector De Arazoza, Fabrice Rossi and Viet-Chi Tran, Social Network Analysis and Mining, volume 5, number 1, pages 58, December 2015. This paper is complemented by a technical report with additional results and by the video below.

Video of the epidemics

See the paper A statistical network analysis of the HIV/AIDS epidemics in Cuba for details.