New approaches based on machine learning to predict insect outbreaks 

New approaches based on machine learning to predict insect outbreaks 

Gabriel Palma, Trinity College Dublin.

20th of November 2025

Abstract: 

The complexity and practical importance of insect outbreaks have made predicting them a focus of recent research. Also, the economic damage caused by sudden increases in insect pest populations in agriculture highlights the importance of developing algorithms to predict this biological phenomenon. These algorithms can be designed based on insect abundance and climate time series, thanks to modern insect-monitoring systems that collect region-specific data. In this presentation, I will introduce the Pattern-based Prediction (PBP) algorithm and a new approach that combines machine learning and time-series embedding to predict insect outbreaks.