Landmine probability maps

The following paper creates landmine probability maps almost precisely in the way that I have wanted to create lost person behavior probability maps. (I'll add some more notes on it later.)

Estimating the Probability of Landmine Contamination. LTC Stephen R. Riese, Donald E. Brown and Yacov Y. Haimes. Military Operations Research 11:3, 2006. MOR V.11 No.3.  (Paywall, sorry.)

This paper introduces a new approach to forecasting landmine contamination in war-torn areas. The Probability of Mine (PoM) forecast incorporates information not used in traditional predictive pattern analysis, and provides more accurate estimates to support decision-making in the allocation of scarce demining resources, relocation of refugees, or planning of peacekeeping operations. That new information is spatial feature data that helps drive human behavior, in this case the choice of where to place landmines. Data gathered by the U.S. Army in Bosnia in 1995 and 1996 provides real world data to test and evaluate the model. The PoM forecast is built upon an empirical Bayesian prediction model that uses examples of areas known to be mined, examples of areas known to be free of mines, feature data (e.g., location of roads, vegetation, fighting lines), and an initial estimate of the overall mine density within the region. Results demonstrate that the PoM model is able to make accurate probabilistic forecasts on the presence of landmines, as well as provide measures of forecast quality that address both the model’s ability to account for uncertainty and the model’s predictive power.

Author: ctwardy

Charles Twardy started the SARBayes project at Monash University in 2000. Work at Monash included SORAL, the Australian Lost Person Behavior Study, AGM-SAR, and Probability Mapper. At George Mason University, he added the MapScore project and related work. More generally, he works on evidence and inference with a special interest in causal models, Bayesian networks, and Bayesian search theory, especially the analysis and prediction of lost person behavior. From 2011-2015, Charles led the DAGGRE & SciCast combinatorial prediction market projects at George Mason University, and has recently joined NTVI Federal as a data scientist supporting the Defense Suicide Prevention Office. Charles received a Dual Ph.D. in History & Philosophy of Science and Cognitive Science from Indiana University, followed by a postdoc in machine learning at Monash.

Leave a Reply

Your email address will not be published. Required fields are marked *