Vertigo Over the Seven V's of Big Data

• Published date: 01 March 2016
• DOI: http://doi.org/10.1002/jcaf.22145
• Date: 01 March 2016
• Author: Tim Chartier

81

© 2016 Wiley Periodicals, Inc.

Published online in Wiley Online Library (wileyonlinelibrary.com).

DOI 10.1002/jcaf.22145

C

o

m

m

e

n

t

a

r

y

Vertigo Over the Seven V’s of Big Data

Tim Chartier

Data is an intricate part of

today’s world and busi-

ness. Big data continues

to emerge and evolve—so much

so that even the concept of

what constitutes big data var-

ies. Originally, there were three

V’s of big data: volume, veloc-

ity, and variety. From the title

of this article, it may look like

I miscounted. In the beginning

there were three V’s, and with

this hot field there were then

four and then five and before

long seven, which we will dis-

cuss here. Let’s start with the

original three.

The volume of big data is

easy enough to see. The con-

tent of the entire World Wide

Web is estimated at upwards of

1 zettabyte, which is 1 trillion

gigabytes. That’s 100 million

times larger than the Library of

Congress.

The velocity of data isn’t

surprising as we often encoun-

ter it. Here are estimates, which

may even change by the time

of this publication, as to the

amount of data happening in

one minute of every day. In one

minute:

• YouTube users upload

72hours of new video

content.

• Google receives over two

million search queries.

• 200 million e-mail messages

are sent.

Finally, there is variety. A

decade ago, data fit neatly into

columns and rows, but not now.

Data comes in multiple, many

unstructured forms such as

tweets, status updates, wearable

device information, videos, and

images. When combined with

the volume of such data, the

landscape of mining for insight

has changed dramatically.

This original list grew to

seven V’s. In fact, just like our

velocity estimates may change

by the time you read this,

so might the number of V’s.

Theseven, though, are volume,

velocity, variety, variability,

veracity, visualization, and

value. A quick Internet search

can allow you more reading on

this expanded list.

I’d like to offer another V:

vertigo. It is easy to be over-

whelmed by today’s data and,

more importantly, analyzing it.

Recently, I met with an engi-

neering firm to discuss their

robust data sets. A lead mem-

ber of the group commented,

“It’s mind boggling to think

what all we can mine from our

data—so much so that we don’t

know where to start.”

Rather than ask probing

questions about their data,

visions of the use of analytics,

existing work, or current proj-

ects, I asked the same question

I ask my undergraduate stu-

dent researchers, “What’s the

simplest thing we can do first?

If we can’t do that, we’d need

to step back before we can do

anything.”

Have a data set that might

offer valuable insight? Are you

stuck deciding where to begin?

Look for a topic or question

that, if answered, could offer

interesting insight but is simple

enough that if not answered

would indicate you don’t have

a good enough handle on your

data to answer most other

questions.

Data analytics, espe-

ciallyof large data sets, is as

much exploration at some

stages as analysis. Answer-

ing simple questions can give

hintsas to informed next steps

to take.

The following quote of

Martin Luther King Jr. gives

insight. “You don’t have to

see the whole staircase, just

take the first step.” Wanting to

seethe end product is natural,