Predicting diarrhoea outbreaks with climate change

space with the smallest structural risk using kernel functions. We used the Scikit-Learn package (https://scikit-learn.org/) to develop all our SVM method with a Radial Basis Function Kernel (RBF) for predictions as shown in Fig 4. For all the methods used in this study, we prepared our input data in a lag format (described in the experiment section), no manual feature extraction step was conducted. Both deep learn- ing methods (CNN and LSTM) were implemented with the Keras and TensorFlow (https:// keras.io/) deep learning library. The methods were configured to make reproducible results Fig 3. Basic structure of the Long-term Short Term (LSTM) method with two LSTM layers. https://doi.org/10.1371/journal.pone.0262008.g003 Fig 4. Structure of the Support Vector Machine (SVM) regression method. The mappings of the input vectors and the final output is discerned with the RBF kernel function. https://doi.org/10.1371/journal.pone.0262008.g004 PLOS ONE Predicting diarrhoea outbreaks with climate change PLOS ONE | https://doi.org/10.1371/journal.pone.0262008 April 19, 2022 7 / 18