Skip Navigation


5 Steps to Taming Big DataMichael Gibbs

5 Steps to Taming Big Data

Where does big data come from and how do researchers make sense of it all?
Ciprian Crainiceanu walks us through one example, from collection to analysis.

1. Accelerometers—small, reliable recorders of movement—are affixed at hip level to seniors in a study.

2. Memory sticks in the devices record 30 observations per second—10 each in three directions.

3. Raw data amassed over hours, days or weeks are downloaded into a computer. (One hour yields 108,000 data points per person—2.6 million per day.)

4. To tame the oceans of information (think thousands of people monitored for a month), Crainiceanu's statistical design team uses software to break long strings of data into meaningful indicators of motion called movelets.

5. Finally, movelets are stitched together into narratives that reveal the time and duration that subjects sat, stood, walked or lay down—creating a true guide to seniors' daily activities.

design element
Online Extras

Alain Labrique

Alain Labrique

Alain Labrique shows off a trove of low-cost technological treasures that support research from Kenya to Bangladesh.

Watch Now

Talk to Us

Amazed? Enthralled? Disappointed? We want to hear from you. Share your thoughts on articles and your ideas for new stories:

Download the PDF

Get a copy of all Feature articles in PDF format. Read stories offline, optimized for printing.

Download Now (4.2MB)