1. Mining Time-Series Databases

2. Data Mining
I – Introduction
The extraction of nontrivial, implicit and useful knowledge from the data
Data
Knowledge
Data Mining
Artificial Intelligence
Computer Science
Statistics
Information Retrieval

3. Data Mining goals To organize the data
I – Introduction
To find “structure” in the large amount of information available from different sources
To organize the data
To identify patterns that translate into new understandings and viable predictions
To discover relationships between data and phenomena that ordinary operations and routine analysis would otherwise overlook
Or make sense

4. Time Series People measure things: I – Introduction Oil price
Sócrates popularity
Blood pressure, etc.
and things change over time, creating a time series

5. Introduction
A Time-Series Database is a database that contains data for each point in time.
Examples:
Weather Data
Stock Prices

6. What to Mine? Full Periodic Patterns Partial Periodic Patterns
Every point in time contributes to the cyclic behavior of the time-series for each period.
e.g., describing the weekly stock prices pattern considering all the days of the week.
Partial Periodic Patterns
Describing the behavior of the time-series at some but not all points in time.
e.g., discovering that the stock prices are high every Saturday and small every Tuesday.

7. Time Series definition
I – Introduction
A (numeric) time series is a sequence of observations of a numeric property over time
-1,25
-1,00
0,01
0,05 …
5,45
0,00

8. Motivation to Work in Time Series
I – Introduction
Time series are ubiquitous
Most of the information (data) produced in a variety of areas are time series
e.g. about 50% of all newspaper graphics are time series
Other types of data can be converted to time series
Image from E. J. Keogh. A decade of progress in indexing and mining large time series databases. In VLDB, page 1268, 2006.

9. Time Series Examples I – Introduction electroencephalogram
Images from a variety of papers by E. J. Keogh. Available at:
electroencephalogram
physiology (muscle activation)
sensors
historical archives
motion data
ECG

10. Time Series Examples (cont.)
I – Introduction
Image from E. J. Keogh. A decade of progress in indexing and mining large time series databases. In VLDB, page 1268, 2006.
stocks data
sales
goods consumption
animal ECG
images
motion capture
handwritten character recognition
DNA sequences

11. Time Series data characteristics
I – Introduction
Analysis is hard, as we are typically dealing with massive data-sets:
One hour EEG: 1 GB of data
Typical weblog: 5 GB / week
MACHO database: 5 TB (growing 3 GB a day)
Stanford Linear Accelerator database: 500 TB
Quadratic complexity algorithms are insufficient
The data also present some distortions (noise, scaling effects, etc.) that make the analysis more difficult

12. Time Series Data Mining Tasks
I – Introduction
Image from E. J. Keogh. A decade of progress in indexing and mining large time series databases. In VLDB, page 1268, 2006.