Habitat Suitability Assessment of Three Passerine Birds Using Ensemble Learning with Diverse Models

The application of machine learning (ML) for predicting species habitat suitability has become increasingly popular. However, using a single ML algorithm may not provide optimal predictions for a given dataset, making it challenging to achieve high accuracy. Therefore, this study proposes a novel approach to assess habitat suitability of three redstarts species (P. Moussieri, P. Ochruros, and P. Phoenicurus) based on ensemble learning techniques. Initially, eight ML models namely MLP, SVM, KNN, Decision Trees (DT), Gradient Boosting Classifier (GB), Random Forest (RF), AdaBoost (AB), and Quadratic Discriminant Analysis (QDA) were trained individually. Then, based on the diversity of these base-learners, seven heterogeneous ensembles of two up to eight models were constructed for each species dataset. This study presents a thorough modeling framework for implementing heterogeneous ensembles and enhancing the overall comprehension of their performance in comparison to single models. The performance of this experiment was evaluated using: (1) six performance measures (AUC, sensitivity, specificity, accuracy, Kappa, and TSS), (2) Borda Count ranking method, and (3) Scott Knott statistical test. Results showed the potential of heterogeneous ensembles for predicting habitat suitability of the three redstarts birds. The proposed ensembles consistently outperformed single models across all datasets.