Bayesian network modelling

The complex problem of environmental management is scoped by both time and geometric dynamics. Bayesian networks provide a probabilistic framework to deal with these dynamics. Because of its conditional independence assumption, it can handle high-dimensional spaces very well. It also provides a powerful ‘what-if’ analysis interface for managerial and scenario support. The use of Bayesian networks and similar graph theory in order to approximate Bayes-Nash equilibria is a novel research area. The application field is computation of optimal persuit strategies.

The poaching of rhino has exhibited a dramatic annual increase since 2008. Whereas there were only 83 rhinos poached in South Africa in 2008, the number have increased to over 1000 in 2013. The number is current approximately 1175 per year at the end of 2015. This is slightly down from the previous year’s number of 1215 owing to increased efforts by the rangers, even though the number of incursions by poachers has continued to increase. Poaching co-occurs with other crimes such as Illegal weapons, Illegal immigration, money laundering, VAT fraud, drugs, murder and attempted murder, corruption, house robberies for large calibre rifles, entrapment and transnational organised crime. The black market price of rhino horn has increased drastically from R8000 per kilogram to more than R1 000 000 per kilogram in 2015, making this form of environmental crime highly profitable. Researcher in the department developed and implemented a geo-spatial predictive model that could be used to understand the poaching problem, to reason about rhino poaching in the park, as well as make predictions. The model is integrated in a information collaboration platform developed by the CSIR and is operational within the Kruger National Park.

Dr Alta de Waal - Principal investigator Department of Statistics, University of Pretoria, Pretoria South Africa E-mail: alta.de [email protected] Telephone: +27 (0) 12 420 3441
Lisa Kirkland - PhD student Email: [email protected]