Hello World! I was told that my first blogpost should start with a joke, so here goes: Why don’t jokes work in octal? Because 7 10 11

When comparing different classification methods, the most frequent used metrics are accuracy, error rate and for visual representation, ROC curve. However when working on an imbalanced dataset, this metrics can be very deceiving. In this post we will look at what a ROC curve are, the problems with using this...

There is a lot to consider when dating around, and people are selective. Finding a prospective match one intends to spend the rest of their life with motivates selective behaviour, which is understandable. This however, gives rise to an inherent imbalance in available data.