ModelWise: Interactive Model Comparison for Model Diagnosis, Improvement and Selection

Model comparison is an important process to facilitate model diagnosis, improvement, and selection when multiple models are developed for a classification task. It involves careful concerning performance interpretation. Current visual analytics solutions often ignore the feature process. They either do not support detailed analysis of multi-class classifiers or rely on alone interpret results. Understanding how different make decisions, especially disagreements same instances, requires deeper understanding. We present ModelWise, method compare in terms performance, space, explanation. ModelWise adapts visualizations with rich interactions workflows achieve selection. considers subspaces generated use improves understanding by demonstrate usability two case studies, one small exemplar dataset another machine learning expert real-world perioperative data.