# Titarl | Beginner guide

## Overview

Titarl is a Data Mining algorithm designed to extract and use temporal rules from symbolic time series and time sequences (SSTS). Titarl rules can be directly interpreted (through the help of the Rule Viewer), used for predictions/detection, or used as input of other analysis. The Titarl binary also include code to automate the cross-validation evaluation of its rules as well as the implementation of various other Machine Learning algorithms (for comparison).

Titarl can learn rules such as:

- If there is an event A at time t, therefore there will be an event B between t+10 and t+20 (90% confidence).
- If there is an event A at time t and the value of sensor S is < 5 at time t, therefore there will be an event B between t+10 and t+20 (90% confidence).
- If there is an event A at time ta, and if there is an event B at time tb between time ta+5 and ta+8, therefore there will be an event C between tb+10 and t+20 (90% confidence).
- If there is an event A at time ta, and if there is NOT an event B at time tb between time ta+5 and ta+8, therefore there will be an event C between ta+10 and t+20 (90% confidence).
- If there is an event A at time ta, and if there is an event B at time tb between time ta+5 and ta+8, and if the value of the sensor S is >5.2, therefore there will be an event C between ta+10 and t+20 (90% confidence).

These rules are *very* simple examples. Titarl can of course learn much more complex rules. Titarl rules can be use for forecasting (prediction in the future), detection (prediction in the past) or just raw correlation analysis.

## Getting Titarl

Titarl is available on the download page.

Next, we invite you to read the Titarl's tutorial.

## Need help?

In case of questions or problems, contact me at mathieug@andrew.cmu.edu.