C. Capinha

Large language models overcome the challenges of unstructured text data in ecology

This study evaluates three large language models for automating ecological data extraction from text, finding GPT 4 performs best across tasks with high accuracy.

Drivers of temporal bias in biodiversity recording by citizen scientists

This study identifies factors driving biodiversity recording activity by citizen scientists, finding that weekends, mild weather, and spring months correlate with higher recording frequency.

Supporting early detection of biological invasions through short-term spatial forecasts of detectability

This paper predicts phenological stages of invasive species using environmental data and opportunistic observations to improve early detection.

Biodiversity monitoring in Europe: user and policy needs

The paper identifies gaps in biodiversity monitoring in Europe and recommends establishing a coordinated network to improve data harmonization and policy response to biodiversity loss.

Recent land use and land cover pressures on Iberian peatlands

This study analyzes land use and land cover changes in Iberian peatlands using historical data and environmental models to identify factors driving changes and conservation needs.

Predicting the timing of ecological phenomena across regions using citizen science data

This paper introduces a framework to predict ecological event timing using citizen science data, showing accurate predictions for Japanese beetles and winter chanterelles across regions.

Forecasting the numbers of disease vectors with deep learning

This paper uses deep learning to forecast disease vector numbers, achieving high accuracy in predicting changes in Aedes aegypti egg counts.

Incorporating environmental time series into species distribution models

This paper develops a deep learning approach for species distribution models using environmental time series data, achieving accurate predictions without simplifying temporal and spatial variables.

Invasive alien mammals of European Union concern

This study reviews the ecology of invasive alien mammals in the European Union, finding that most species were introduced as pets and are expanding their ranges, with significant impacts on biodiversity and human health.

Deep learning classification of temporal data in ecology

This paper applies deep learning to classify ecological temporal data without manual feature extraction, showing improved accuracy across three ecological case studies.

A machine learning approach for the spatiotemporal forecasting of ecological phenomena using dates of species occurrence records

This paper develops a machine learning approach to forecast ecological phenomena using species occurrence records, showing improved accuracy compared to date-based null models.