Probability and Statistics for Computer Science pp 115-137 | Cite as

# Useful Probability Distributions

## Abstract

We will use probability as a tool to resolve practical questions about data. Here are important example questions. We could ask **what process produced the data?** For example, I observe a set of independent coin flips. I would now like to know the probability of observing a head when the coin is flipped. We could ask **what sort of data can we expect in the future?** For example, what will be the outcome of the next election? Answering this requires collecting information about voters, preferences, and the like, then using it to build a model that predicts the outcome. We could ask **what labels should we attach to unlabelled data?** For example, we might see a large number of credit card transactions, some known to be legitimate and others known to be fraudulent. We now see a new transaction: is it legitimate? We could ask **is an effect easily explained by chance variations, or is it real?** For example, a medicine appears to help patients with a disease. Is there a real effect, or is it possible that by chance the patients we tested the medicine on felt better?