Evaluation of Predictive Models forWildlife Poaching Activity through Controlled Field Test in Uganda
Loading...
Date
2018
Journal Title
Journal ISSN
Volume Title
Publisher
AAAI Conference on Artificial Intelligence
Abstract
Worldwide, conservation agencies employ rangers to protect
conservation areas from poachers. However, agencies lack the
manpower to have rangers effectively patrol these vast areas
frequently. While past work modeled poachers behavior
so as to aid rangers in planning future patrols, those models
predictions were not validated by extensive field tests.We
conducted two rounds of field tests in Ugandas Queen Elizabeth
Protected Area to evaluate our proposed spatio-temporal
model that predicts poaching threat levels. In the first round,
a one-month field test was conducted to test the predictive
power of the model and in the second round an eight-month
test was conducted to evaluate the selectiveness power of the
model. To our knowledge, this is the first time that a predictive
model is evaluated through such an extensive field test
in this domain. These field tests will be extended to another
park in Uganda, Murchison Fall Protected Area. Once such
models are evaluated in the field, they can be used to generate
efficient and feasible patrol routes for the park rangers.
Description
Keywords
Citation
Gholami, S., Ford, B., Kar, D., Fang, F., Tambe, M., Plumptre, A., ... & Mabonga, J. (2018, June). Evaluation of predictive models for wildlife poaching activity through controlled field test in uganda. In Workshops at the Thirty-Second AAAI Conference on Artificial Intelligence.