Member-only story
Matthews Correlation Coefficient is The Best Classification Metric You’ve Never Heard Of
Congratulations! You’ve built a binary classifier —a fancy-schmancy neural network using 128 GPUs with their dedicated power station, or perhaps a robust logistic regression model that runs on your good old ThinkPad. You’ve designed the model and fed the data; now the time has finally come to measure the classifier’s performance.
Don’t get me wrong: ROC curves are the best choice for comparing models. However, scalar metrics still remain popular among the machine-learning community with the four most common being accuracy, recall, precision, and F1-score. Scalar metrics are ubiquitous in textbooks, web articles, online courses, and they are the metrics that most data scientists are familiar with. But a couple of weeks ago, I stumbled upon another scalar metric for binary classification: the Matthews Correlation Coefficient (MCC). Following my “discovery”, I asked around and was surprised to find that many people in the field are not familiar with this classification metric. As a born-again believer, I’m here to spread the gospel!
Let’s start with a quick overview of the “Famous Four” metrics, including a discussion on why they are sometimes not very useful, or even downright misleading. Following that, I’ll introduce the other metric.