Introduction We’ve been writing on the distribution density shapes expected for probability models in ROC (receiver operator characteristic) plots, double density plots, and normal/logit-normal densities frameworks. I thought I would re-approach the issue with a specific family of examples.
For classification problems I argue one of the biggest steps you can take to improve the quality and utility of your models is to prefer models that return scores or return probabilities instead of classification rules. Doing this also opens a second large opportunity for improvement: working with your domain […]
Recently, we showed how to use utility estimates to pick good classifier thresholds. In that article, we used model performance on an evaluation set, combined with estimates of rewards and penalties for correct and incorrect classifications, to find a threshold that optimized model utility. In this article, we will show […]
In a previous article we discussed why it’s a good idea to prefer probability models to “hard” classification models, and why you should delay setting “hard” classification rules as long as possible. But decisions have to be made, and eventually you will have to set that threshold. How do you […]
It’s a folk theorem I sometimes hear from colleagues and clients: that you must balance the class prevalence before training a classifier. Certainly, I believe that classification tends to be easier when the classes are nearly balanced, especially when the class you are actually interested in is the rarer one. […]
As John mentioned in his last post, we have been quite interested in the recent study by Fernandez-Delgado, et.al., “Do we Need Hundreds of Classifiers to Solve Real World Classification Problems?” (the “DWN study” for short), which evaluated 179 popular implementations of common classification algorithms over 120 or so data […]
Many data science projects and presentations are needlessly derailed by not having set shared business relevant quantitative expectations early on (for some advice see Setting expectations in data science projects). One of the most common issues is the common layman expectation of “perfect prediction” from classification projects. It is important […]
A bit more on the ROC/AUC The issue The receiver operating characteristic curve (or ROC) is one of the standard methods to evaluate a scoring system. Nina Zumel has described its application, but I would like to call out some additional details. In my opinion while the ROC is a […]