The multi-class prediction had gained popularity over the recent years. Thus, a natural question that is raised is how to measure the goodness of the fit. These metrics must take into account the imbalance in the data.
To be able to do this we will review regular metrics for two dimension
We will consider the geometric properties of the metrics
This would lead us to the metrics in the multi-class case
In particular, we will discuss the Mathew's Correlation Coefficient.
Uri Itai phd in applied math 2014 Under the supervision of prof. Nira Dyn
working in data science since 2012
Senior Data scientist in TRST AI