Some generic data visualization using ggplot2 package and UK Bakeoff data.

Not all passengers who buy a plane ticket show up at boarding. The no shows make flights occur with idle capacity and incur an opportunity cost for the operator. To compensate, airlines use overbooking (sale of seats above the flight capacity). But how many additional seats should we offer without it becoming a chronic passenger relocation problem?

Can you imagine a real business case where you apply machine learning to build and Random Forest classifier and its accuracy of the model isn’t the (only) main metric to pay attention? In real case scenarios the cost and benefits can affect a model in different aspects, this post exercises a business case where the return of an investment is dependent of the behavior of the precision metric.

This post explore Power Analysis technique. Power is the probability of detecting an effect, given that the effect is really there. In other words, it is the probability of rejecting the null hypothesis when it is in fact false. For example, let’s say that we have a simple study with drug A and a placebo group, and that the drug truly is effective; the power is the probability of finding a difference between the two groups.¹

“The only way to learn mathematics is to do mathematics.” - Paul Halmos. Taking time out of the day-to-day rush to finally learn how to use {tidymodels} for machine learning. These are the notebooks that I did to enter in this universe.

Jon Penney (2016)¹ explored whether the widespread publicity about NSA/PRISM surveillance (i.e., the Snowden revelations) in June 2013 was associated with a sharp and sudden decrease in traffic to Wikipedia articles on topics that raise privacy concerns. This post tries to reproduce some of this findings.