Outbreak detection is the task of detecting outbreaks in networks, where given a network and a dynamic process spreading over this network, we have to select a set of nodes to detect this process as effectively as possible. Detecting contaminants in water networks, malicious software in computer networks, infected hosts in social networks, popular and controversial posts on online sites, are all instances of outbreak detection. We will look at competing algorithms of the “Battle of the Water Sensor Networks” (BWSN), which was a competition held at the “Annual Water Distribution Systems Analysis Symposium” at Cincinnati, Ohio in August 2006. This competition required the submission of algorithms that can find the points where a set number of sensors can be placed so as to detect outbreaks in the most efficient manner possible. Although we have looked briefly at the CELF algorithm in Chap. 9, we will look at it in detail in this chapter on how it works on the water sensor problem and detects posts of interest in Blogspace.
- 1.Goyal, Amit, Wei Lu, and Laks V.S. Lakshmanan. 2011. Celf++: Optimizing the greedy algorithm for influence maximization in social networks. In Proceedings of the 20th international conference companion on World wide web, 47–48. ACM.Google Scholar
- 2.Krause, Andreas, and Carlos Guestrin. 2005. A note on the budgeted maximization of submodular functions. 2005.Google Scholar
- 4.Leskovec, Jure, Andreas Krause, Carlos Guestrin, Christos Faloutsos, Jeanne VanBriesen, and Natalie Glance. 2007. Cost-effective outbreak detection in networks. In Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining, 420–429. ACM.Google Scholar
- 5.Ostfeld, Avi, James G. Uber, Elad Salomons, Jonathan W. Berry, William E. Hart, Cindy A. Phillips, Jean-Paul Watson, Gianluca Dorini, Philip Jonkergouw, Zoran Kapelan, et al. 2008. The battle of the water sensor networks (BWSN): A design challenge for engineers and algorithms. Journal of Water Resources Planning and Management 134 (6): 556–568.CrossRefGoogle Scholar