Graph Structure Learning from Unlabeled Data for Early Outbreak Detection
- 27 March 2017
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Intelligent Systems
- Vol. 32 (2), 80-84
- https://doi.org/10.1109/MIS.2017.25
Abstract
Processes such as disease propagation and information diffusion often spread over some latent network structure that must be learned from observation. Given a set of unlabeled training examples representing occurrences of an event type of interest (such as a disease outbreak), the authors aim to learn a graph structure that can be used to accurately detect future events of that type. They propose a novel framework for learning graph structure from unlabeled data by comparing the most anomalous subsets detected with and without the graph constraints. Their framework uses the mean normalized log-likelihood ratio score to measure the quality of a graph structure, and it efficiently searches for the highest-scoring graph structure. Using simulated disease outbreaks injected into real-world Emergency Department data from Allegheny County, the authors show that their method learns a structure similar to the true underlying graph, but enables faster and more accurate detection.Keywords
This publication has 6 references indexed in Scilit:
- Scalable Detection of Anomalous Patterns With Connectivity ConstraintsJournal of Computational and Graphical Statistics, 2015
- Dynamic Pattern Detection with Temporal Consistency and Connectivity ConstraintsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2013
- Fast Subset Scan for Spatial Pattern DetectionJournal of the Royal Statistical Society Series B: Statistical Methodology, 2012
- Inferring networks of diffusion and influencePublished by Association for Computing Machinery (ACM) ,2010
- Detection of emerging space-time clustersPublished by Association for Computing Machinery (ACM) ,2005
- Upper level set scan statistic for detecting arbitrarily shaped hotspotsEnvironmental and Ecological Statistics, 2004