Active-Learning Class Activities and Shiny Applications for Teaching Support Vector Classifiers

Support vector classifiers are one of the most popular linear classification techniques for binary classification. Different from some commonly seen model fitting criteria in statistics, such as ordinary least squares criterion and maximum likelihood method, its algorithm depends on an optimization problem under constraints, which is unconventional to many students a second or third course statistics data science. As result, this topic often not intuitive more traditional statistical modeling tools. In order facilitate students’ mastery promote active learning, we developed in-class activities their accompanying Shiny applications teaching support classifiers. The designed materials aim at engaging through group work solidifying understanding via hands-on explorations. offer interactive demonstration changes components classifier when altering determining parameters. With goal benefiting broader science education community, have made publicly available. addition, detailed activity worksheet real example also provided online supplementary materials.