A machine learning software tool for multiclass classification

This paper describes code for a published article that can assist researchers with multiclass classification problems and analyse the performances of various machine learning models. Further, feature importance, feature correlation, variable clustering, confusion matrix and kernel density estimation were also implemented. The original study was published in Expert Systems with Applications, and this paper explains the code and workflow. Administrative healthcare data has been used as an example to run the code. The results and insights can assist healthcare stakeholders and policymakers reduce the negative impact of illness comorbidity and multimorbidity.