Pattern-based prediction of population outbreaks

The complexity and practical importance of insect outbreaks have made the problem predicting a focus recent research. We propose Pattern-Based Prediction (PBP) method for population outbreaks. It uses information on previous time series values that precede an outbreak event as predictors future outbreaks, which can be helpful when monitoring pest species. illustrate methodology using simulated datasets aphid obtained in wheat crops Southern Brazil. average test accuracy 84.6% simulation studies implemented with stochastic models 95.0% aphids Our results show PBP method's feasibility benchmarked our against established state-of-the-art machine learning methods: Support Vector Machines, Deep Neural Networks, Long Short Term Memory Random Forests. yielded competitive performance associated higher true-positive rates most comparisons while providing interpretability rather than being black-box method. is improvement over current tools, especially by non-specialists, such ecologists aiming to use quantitative approach monitoring. provide Python through pypbp package.