<s>
Click	B-Protocol
tracking	I-Protocol
is	O
when	O
user	O
click	O
behavior	O
or	O
user	O
navigational	O
behavior	O
is	O
collected	O
in	O
order	O
to	O
derive	O
insights	O
and	O
fingerprint	O
users	O
.	O
</s>
<s>
Research	O
has	O
also	O
been	O
conducted	O
to	O
explore	O
the	O
user	B-Application
experience	I-Application
of	O
privacy	O
with	O
making	O
user	O
personal	O
identification	O
information	O
individually	O
anonymized	O
and	O
improving	O
how	O
data	O
collection	O
consent	O
forms	O
are	O
written	O
and	O
structured	O
.	O
</s>
<s>
Click	B-Protocol
tracking	I-Protocol
is	O
relevant	O
in	O
several	O
industries	O
including	O
Human-Computer	O
Interaction	O
(	O
HCI	O
)	O
,	O
software	B-General_Concept
engineering	I-General_Concept
,	O
and	O
advertising	O
.	O
</s>
<s>
Email	B-Operating_System
tracking	I-Operating_System
,	O
link	O
tracking	O
,	O
web	O
analytics	O
,	O
and	O
user	O
research	O
are	O
also	O
related	O
concepts	O
and	O
applications	O
of	O
click	B-Protocol
tracking	I-Protocol
.	O
</s>
<s>
A	O
common	O
utilization	O
of	O
click	O
data	O
from	O
click	B-Protocol
tracking	I-Protocol
is	O
to	O
improve	O
results	O
 '	O
positions	O
from	O
search	O
engines	O
to	O
make	O
their	O
order	O
more	O
relevant	O
to	O
users	O
 '	O
needs	O
.	O
</s>
<s>
Click	B-Protocol
tracking	I-Protocol
employs	O
many	O
modern	O
techniques	O
such	O
as	O
machine	O
learning	O
and	O
data	B-Application
mining	I-Application
.	O
</s>
<s>
Examples	O
of	O
TRTs	O
include	O
radio	B-Application
frequency	I-Application
identification	I-Application
(	O
RFID	B-Application
)	O
,	O
credit	O
cards	O
,	O
and	O
store	O
video	O
cameras	O
.	O
</s>
<s>
Another	O
means	O
of	O
obtaining	O
user	O
input	O
is	O
eye-tracking	B-General_Concept
or	O
gaze	B-General_Concept
tracking	I-General_Concept
.	O
</s>
<s>
In	O
this	O
process	O
,	O
the	O
gaze-tracking	O
system	O
is	O
separated	O
into	O
its	O
own	O
panel	O
in	O
the	O
system	O
interface	O
,	O
and	O
the	O
user	B-Application
experience	I-Application
of	O
this	O
system	O
is	O
compromised	O
as	O
individuals	O
have	O
to	O
switch	O
between	O
the	O
panel	O
and	O
the	O
other	O
interface	O
features	O
.	O
</s>
<s>
Hence	O
,	O
researchers	O
created	O
their	O
own	O
web	O
browser	O
called	O
GazeTheWeb	O
(	O
GTW	O
)	O
,	O
and	O
the	O
focus	O
of	O
their	O
research	O
was	O
on	O
the	O
user	B-Application
experience	I-Application
.	O
</s>
<s>
Browser	B-Protocol
fingerprinting	I-Protocol
is	O
another	O
means	O
of	O
identifying	O
users	O
and	O
tracking	O
them	O
.	O
</s>
<s>
In	O
this	O
process	O
,	O
information	O
about	O
a	O
user	O
is	O
collected	O
from	O
their	O
web	O
browser	O
to	O
create	O
a	O
browser	B-Protocol
fingerprint	I-Protocol
.	O
</s>
<s>
A	O
browser	B-Protocol
fingerprint	I-Protocol
contains	O
information	O
about	O
a	O
device	O
,	O
its	O
operating	O
system	O
,	O
its	O
browser	O
,	O
and	O
its	O
configuration	O
.	O
</s>
<s>
HTTP	B-Protocol
headers	O
,	O
JavaScript	B-Language
,	O
and	O
browser	B-Protocol
plugins	I-Protocol
can	O
be	O
used	O
to	O
build	O
a	O
fingerprint	O
.	O
</s>
<s>
Browser	B-Protocol
fingerprints	I-Protocol
can	O
change	O
over	O
time	O
from	O
automatic	O
software	O
updates	O
or	O
user	O
browser	O
preference	O
adjustments	O
.	O
</s>
<s>
However	O
,	O
more	O
modern	O
tracking	O
software	O
utilizes	O
JavaScript	B-Language
in	O
order	O
to	O
track	O
cursor	O
behavior	O
.	O
</s>
<s>
This	O
software	O
also	O
generates	O
a	O
log	B-Application
analysis	I-Application
which	O
summarizes	O
user	O
cursor	O
activity	O
.	O
</s>
<s>
In	O
a	O
search	B-Protocol
session	I-Protocol
,	O
users	O
can	O
be	O
identified	O
using	O
cookies	B-Application
,	O
identd	O
protocol	O
,	O
or	O
their	O
IP	B-Protocol
address	I-Protocol
.	O
</s>
<s>
DoubleClick	O
Inc	O
.	O
is	O
an	O
example	O
of	O
a	O
company	O
that	O
has	O
such	O
a	O
database	O
and	O
partners	O
with	O
other	O
companies	O
to	O
aid	O
with	O
their	O
web	B-Application
mining	I-Application
.	O
</s>
<s>
Cookies	B-Application
are	O
added	O
to	O
HTTP	B-Protocol
(	O
Hypertext	B-Protocol
Transfer	I-Protocol
Protocol	I-Protocol
)	O
,	O
and	O
when	O
a	O
user	O
clicks	O
on	O
a	O
link	O
,	O
they	O
are	O
connected	O
to	O
the	O
associated	O
web	O
server	O
.	O
</s>
<s>
This	O
action	O
of	O
a	O
user	O
clicking	O
on	O
a	O
link	O
is	O
seen	O
as	O
a	O
request	O
,	O
and	O
the	O
server	O
“	O
responds	O
”	O
by	O
sending	O
the	O
user	O
's	O
information	O
,	O
and	O
this	O
information	O
is	O
a	O
cookie	B-Application
.	O
</s>
<s>
Cookies	B-Application
provide	O
a	O
“	O
bookmark	O
”	O
for	O
users’	O
sessions	O
on	O
a	O
website	O
,	O
and	O
they	O
store	O
user	O
login	O
information	O
and	O
the	O
pages	O
users	O
visit	O
on	O
a	O
website	O
.	O
</s>
<s>
Data	O
collected	O
via	O
cookies	B-Application
can	O
be	O
used	O
to	O
improve	O
websites	O
for	O
all	O
users	O
and	O
this	O
also	O
aids	O
with	O
user	O
profiling	O
for	O
advertising	O
.	O
</s>
<s>
When	O
data	B-Application
mining	I-Application
techniques	O
and	O
statistical	O
procedures	O
are	O
applied	O
to	O
understand	O
web	O
log	O
data	O
,	O
the	O
process	O
is	O
noted	O
as	O
log	B-Application
analysis	I-Application
or	O
web	B-Application
usage	I-Application
mining	I-Application
.	O
</s>
<s>
Web	B-Application
usage	I-Application
mining	I-Application
has	O
three	O
phases	O
.	O
</s>
<s>
The	O
tool	O
of	O
association	B-Algorithm
rule	I-Algorithm
mining	I-Algorithm
helps	O
with	O
finding	O
“	O
patterns	O
,	O
associations	O
,	O
and	O
correlations	O
”	O
among	O
pages	O
users	O
visit	O
in	O
a	O
search	B-Protocol
session	I-Protocol
.	O
</s>
<s>
Sequential	O
pattern	O
discovery	O
is	O
association	B-Algorithm
rule	I-Algorithm
mining	I-Algorithm
,	O
but	O
it	O
also	O
accounts	O
for	O
time	O
like	O
the	O
page	O
views	O
in	O
an	O
allotted	O
time	O
period	O
.	O
</s>
<s>
Classification	B-General_Concept
is	O
a	O
tool	O
that	O
allows	O
for	O
pages	O
to	O
be	O
added	O
to	O
groups	O
representing	O
certain	O
similar	O
qualities	O
.	O
</s>
<s>
Some	O
examples	O
of	O
tools	O
individuals	O
can	O
use	O
when	O
conducting	O
click	B-Protocol
analytics	I-Protocol
are	O
the	B-Application
Google	I-Application
Analytics	O
tool	O
In-Page	O
Analytics	O
,	O
ClickHeat	O
,	O
and	O
Crazy	O
Egg	O
.	O
</s>
<s>
When	O
given	O
Google	B-Application
Chrome	I-Application
warnings	O
,	O
70%	O
of	O
the	O
time	O
people	O
will	O
click	O
through	O
.	O
</s>
<s>
Pages	O
that	O
are	O
viewed	O
by	O
users	O
during	O
a	O
particular	O
search	B-Protocol
session	I-Protocol
constitute	O
click	O
data	O
.	O
</s>
<s>
Certain	O
user	O
actions	O
on	O
a	O
webpage	O
that	O
can	O
be	O
used	O
as	O
a	O
part	O
of	O
the	O
interpretation	O
process	O
include	O
bookmarking	B-General_Concept
,	O
saving	O
,	O
or	O
printing	O
a	O
particular	O
web	O
page	O
.	O
</s>
<s>
In	O
a	O
search	B-Protocol
session	I-Protocol
,	O
a	O
user	O
indicates	O
which	O
documents	O
they	O
are	O
more	O
interested	O
in	O
with	O
their	O
clicks	O
,	O
and	O
this	O
indicates	O
what	O
is	O
relevant	O
to	O
the	O
search	O
.	O
</s>
<s>
The	O
most	O
relevant	O
click	O
data	O
to	O
determine	O
relevance	O
of	O
results	O
is	O
often	O
the	O
last	O
viewed	O
web	O
page	O
rather	O
than	O
all	O
of	O
the	O
pages	O
clicked	O
on	O
in	O
a	O
search	B-Protocol
session	I-Protocol
.	O
</s>
<s>
Click	O
data	O
outside	O
of	O
search	B-Protocol
sessions	I-Protocol
can	O
also	O
be	O
used	O
to	O
improve	O
the	O
accuracy	O
of	O
relevant	O
results	O
for	O
users	O
.	O
</s>
<s>
Then	O
,	O
live	O
user	O
click	O
feedback	O
in	O
the	O
form	O
of	O
tracked	O
click-through	O
rates	O
(	O
CTR	O
)	O
in	O
search	B-Protocol
sessions	I-Protocol
can	O
be	O
used	O
to	O
rerank	O
the	O
results	O
based	O
on	O
the	O
data	O
.	O
</s>
<s>
Eye-tracking	B-General_Concept
research	O
indicates	O
that	O
users	O
exhibit	O
an	O
abundance	O
of	O
non-sequential	O
viewing	O
activity	O
when	O
looking	O
at	O
search	O
results	O
.	O
</s>
<s>
Supply-demand	O
mismatch	O
costs	O
can	O
be	O
reduced	O
through	O
click	B-Protocol
tracking	I-Protocol
.	O
</s>
<s>
Click	B-Protocol
tracking	I-Protocol
can	O
be	O
used	O
in	O
the	O
realm	O
of	O
advertising	O
,	O
but	O
there	O
is	O
the	O
potential	O
for	O
this	O
tool	O
to	O
be	O
used	O
negatively	O
.	O
</s>
<s>
In	O
the	O
2012	O
Fraud	O
Detection	O
in	O
Mobile	O
Advertising	O
(	O
FDMA	O
)	O
conference	O
,	O
competition	O
teams	O
were	O
tasked	O
with	O
having	O
to	O
use	O
data	B-Application
mining	I-Application
and	O
machine	O
learning	O
techniques	O
to	O
determine	O
“	O
fraudulent	O
publishers	O
”	O
from	O
a	O
given	O
dataset	O
.	O
</s>
<s>
In	O
order	O
to	O
know	O
what	O
emails	O
users	O
have	O
opened	O
,	O
email	O
senders	O
engage	O
in	O
email	B-Operating_System
tracking	I-Operating_System
.	O
</s>
<s>
When	O
a	O
third-party	B-General_Concept
is	O
associated	O
with	O
a	O
first-party	O
website	O
or	O
mobile	O
application	O
,	O
anytime	O
a	O
user	O
visits	O
the	O
first	O
party	O
website	O
or	O
mobile	O
application	O
,	O
their	O
information	O
will	O
be	O
sent	O
to	O
the	O
third-party	B-General_Concept
.	O
</s>
<s>
Third-party	B-General_Concept
tracking	O
generates	O
more	O
privacy	O
concerns	O
than	O
first-party	O
tracking	O
because	O
it	O
allows	O
for	O
many	O
website	O
or	O
application	O
records	O
about	O
a	O
particular	O
user	O
to	O
be	O
combined	O
,	O
yielding	O
better	O
user	O
profiles	O
.	O
</s>
<s>
Researchers	O
evaluated	O
the	O
reach	O
of	O
third-party	B-General_Concept
trackers	O
based	O
on	O
their	O
contact	O
with	O
users	O
rather	O
than	O
websites	O
,	O
so	O
more	O
"	O
popular	O
"	O
trackers	O
were	O
those	O
who	O
received	O
information	O
about	O
the	O
highest	O
number	O
of	O
people	O
rather	O
than	O
code	O
embedded	O
in	O
the	O
most	O
first-parties	O
.	O
</s>
<s>
Google	B-Application
and	O
Facebook	B-Application
were	O
deemed	O
as	O
the	O
first	O
and	O
second	O
largest	O
web	O
trackers	O
,	O
and	O
Google	B-Application
and	O
Twitter	B-Application
were	O
deemed	O
as	O
the	O
first	O
and	O
second	O
largest	O
mobile	O
trackers	O
.	O
</s>
