## About

Something predictable is showing a pattern and is therefore not truly random.

A high-entropy source is completely chaotic, is unpredictable, and is called true randomness.

Many forms of data mining model are predictive. For example, a model might predict income based on education and other demographic factors.

An accurate prediction function does not imply that the function is an accurate model of the phenomenon being analysed, just that it makes an accurate prediction given the data.

Predictive functions:

- Classification: Predict a class label based on other attributes. The learned attribute is Categorical (“nominal”).
- Regression: Predict a numeric value. The learned attribute is a continuous numeric value
- Ranking: Predict the order of relevance

It's easier to invent the future than try to predict it.

100% predictability = 0% innovation

The best way to predict the future is to invent it.

I've always been more interested in the future than in the past.

## Prediction vs Forecasting

Forecasting is the process of making predictions of the future based on past and present data (time serie) whereas prediction is a more more general term.

## Probability

Predictions have an associated probability that gives the degree of uncertainty.

Prediction probabilities are also known as confidence (How confident can I be of this prediction?).

This Prediction probability is one way to indicate:

- and wiki/uncertainty

## Example: Actionable Information

Data mining can derive actionable information from large volumes of data.

For example:

- a town planner might use a model that predicts income based on demographics to develop a plan for low-income housing.
- a car leasing agency might a use model that identifies customer segments to design a promotion targeting high-value customers.

## Documentation / Reference

- Jeopardy - Watson Game. See also Waston Academy (MOOC)