Model Evaluation with Weighted Threshold Optimization (and the “mewto” R package)

mewto is an R package that allows users to experiment with different thresholds for classification of prediction results in the case of binary classification problems and to interactively visualize model evaluation metrics, confusion matrices, the ROC and PR curves. It can also calculate the optimal threshold based on a weighted evaluation criterion and displays related performance metrics. The package enhances model performance evaluation in two main ways: by providing an interactive interface through which the user can perform exploratory analysis of model result data and visualize key metrics, as well as by developing a tool that can be used to find the optimum threshold for binary classification problems using a weighted optimization formula