<s>
Educational	B-Algorithm
data	I-Algorithm
mining	I-Algorithm
(	O
EDM	O
)	O
is	O
a	O
research	O
field	O
concerned	O
with	O
the	O
application	O
of	O
data	B-Application
mining	I-Application
,	O
machine	O
learning	O
and	O
statistics	O
to	O
information	O
generated	O
from	O
educational	O
settings	O
(	O
e.g.	O
,	O
universities	O
and	O
intelligent	O
tutoring	O
systems	O
)	O
.	O
</s>
<s>
Educational	B-Algorithm
data	I-Algorithm
mining	I-Algorithm
refers	O
to	O
techniques	O
,	O
tools	O
,	O
and	O
research	O
designed	O
for	O
automatically	O
extracting	O
meaning	O
from	O
large	O
repositories	O
of	O
data	O
generated	O
by	O
or	O
related	O
to	O
people	O
's	O
learning	O
activities	O
in	O
educational	O
settings	O
.	O
</s>
<s>
For	O
example	O
,	O
several	O
learning	B-Application
management	I-Application
systems	I-Application
(	O
LMSs	O
)	O
track	O
information	O
such	O
as	O
when	O
each	O
student	O
accessed	O
each	O
learning	O
object	O
,	O
how	O
many	O
times	O
they	O
accessed	O
it	O
,	O
and	O
how	O
many	O
minutes	O
the	O
learning	O
object	O
was	O
displayed	O
on	O
the	O
user	O
's	O
computer	O
screen	O
.	O
</s>
<s>
As	O
interest	O
in	O
EDM	O
continued	O
to	O
increase	O
,	O
EDM	O
researchers	O
established	O
an	O
academic	O
journal	O
in	O
2009	O
,	O
the	O
Journal	O
of	O
Educational	B-Algorithm
Data	I-Algorithm
Mining	I-Algorithm
,	O
for	O
sharing	O
and	O
disseminating	O
research	O
results	O
.	O
</s>
<s>
In	O
2011	O
,	O
EDM	O
researchers	O
established	O
the	O
International	O
Educational	B-Algorithm
Data	I-Algorithm
Mining	I-Algorithm
Society	O
to	O
connect	O
EDM	O
researchers	O
and	O
continue	O
to	O
grow	O
the	O
field	O
.	O
</s>
<s>
With	O
the	O
introduction	O
of	O
public	O
educational	O
data	O
repositories	O
in	O
2008	O
,	O
such	O
as	O
the	O
Pittsburgh	O
Science	O
of	O
Learning	O
Centre	O
's	O
(	O
PSLC	O
)	O
DataShop	O
and	O
the	O
National	O
Center	O
for	O
Education	O
Statistics	O
(	O
NCES	O
)	O
,	O
public	O
data	O
sets	O
have	O
made	O
educational	B-Algorithm
data	I-Algorithm
mining	I-Algorithm
more	O
accessible	O
and	O
feasible	O
,	O
contributing	O
to	O
its	O
growth	O
.	O
</s>
<s>
The	O
user	B-Application
experience	I-Application
of	O
the	O
learner	O
and	O
their	O
overall	O
satisfaction	O
with	O
learning	O
are	O
also	O
measured	O
.	O
</s>
<s>
There	O
are	O
four	O
main	O
users	O
and	O
stakeholders	O
involved	O
with	O
educational	B-Algorithm
data	I-Algorithm
mining	I-Algorithm
.	O
</s>
<s>
For	O
younger	O
learners	O
,	O
educational	B-Algorithm
data	I-Algorithm
mining	I-Algorithm
can	O
also	O
inform	O
parents	O
about	O
their	O
child	O
's	O
learning	O
progress	O
.	O
</s>
<s>
Researchers	O
–	O
Researchers	O
focus	O
on	O
the	O
development	O
and	O
the	O
evaluation	O
of	O
data	B-Application
mining	I-Application
techniques	O
for	O
effectiveness	O
.	O
</s>
<s>
The	O
wide	O
range	O
of	O
topics	O
in	O
EDM	O
ranges	O
from	O
using	O
data	B-Application
mining	I-Application
to	O
improve	O
institutional	O
effectiveness	O
to	O
student	O
performance	O
.	O
</s>
<s>
As	O
research	O
in	O
the	O
field	O
of	O
educational	B-Algorithm
data	I-Algorithm
mining	I-Algorithm
has	O
continued	O
to	O
grow	O
,	O
a	O
myriad	O
of	O
data	B-Application
mining	I-Application
techniques	O
have	O
been	O
applied	O
to	O
a	O
variety	O
of	O
educational	O
contexts	O
.	O
</s>
<s>
The	O
first	O
phase	O
of	O
the	O
EDM	O
process	O
(	O
not	O
counting	O
pre-processing	B-Application
)	O
is	O
discovering	O
relationships	O
in	O
data	O
.	O
</s>
<s>
Several	O
algorithms	O
for	O
identifying	O
such	O
relationships	O
have	O
been	O
utilized	O
,	O
including	O
classification	B-General_Concept
,	O
regression	O
,	O
clustering	B-Algorithm
,	O
factor	O
analysis	O
,	O
social	O
network	O
analysis	O
,	O
association	B-Algorithm
rule	I-Algorithm
mining	I-Algorithm
,	O
and	O
sequential	B-Algorithm
pattern	I-Algorithm
mining	I-Algorithm
.	O
</s>
<s>
Discovered	O
relationships	O
must	O
then	O
be	O
validated	O
in	O
order	O
to	O
avoid	O
overfitting	B-Error_Name
.	O
</s>
<s>
A	O
large	O
amount	O
of	O
research	O
has	O
been	O
conducted	O
in	O
best	O
practices	O
for	O
visualizing	B-Application
data	I-Application
.	O
</s>
<s>
Of	O
the	O
general	O
categories	O
of	O
methods	O
mentioned	O
,	O
prediction	O
,	O
clustering	B-Algorithm
and	O
relationship	O
mining	O
are	O
considered	O
universal	O
methods	O
across	O
all	O
types	O
of	O
data	B-Application
mining	I-Application
;	O
however	O
,	O
Discovery	O
with	O
Models	O
and	O
Distillation	O
of	O
Data	O
for	O
Human	O
Judgment	O
are	O
considered	O
more	O
prominent	O
approaches	O
within	O
educational	B-Algorithm
data	I-Algorithm
mining	I-Algorithm
.	O
</s>
<s>
In	O
the	O
Discovery	O
with	O
Model	O
method	O
,	O
a	O
model	O
is	O
developed	O
via	O
prediction	O
,	O
clustering	B-Algorithm
or	O
by	O
human	O
reasoning	O
knowledge	O
engineering	O
and	O
then	O
used	O
as	O
a	O
component	O
in	O
another	O
analysis	O
,	O
namely	O
in	O
prediction	O
and	O
relationship	O
mining	O
.	O
</s>
<s>
Humans	O
can	O
make	O
inferences	O
about	O
data	O
that	O
may	O
be	O
beyond	O
the	O
scope	O
in	O
which	O
an	O
automated	O
data	B-Application
mining	I-Application
method	O
provides	O
.	O
</s>
<s>
For	O
the	O
use	O
of	O
education	O
data	B-Application
mining	I-Application
,	O
data	O
is	O
distilled	O
for	O
human	O
judgment	O
for	O
two	O
key	O
purposes	O
,	O
identification	O
and	O
classification	B-General_Concept
.	O
</s>
<s>
For	O
example	O
,	O
the	O
learning	B-General_Concept
curve	I-General_Concept
,	O
classic	O
to	O
educational	O
studies	O
,	O
is	O
a	O
pattern	O
that	O
clearly	O
reflects	O
the	O
relationship	O
between	O
learning	O
and	O
experience	O
over	O
time	O
.	O
</s>
<s>
Data	O
is	O
also	O
distilled	O
for	O
the	O
purposes	O
of	O
classifying	B-General_Concept
features	O
of	O
data	O
,	O
which	O
for	O
educational	B-Algorithm
data	I-Algorithm
mining	I-Algorithm
,	O
is	O
used	O
to	O
support	O
the	O
development	O
of	O
the	O
prediction	O
model	O
.	O
</s>
<s>
Classification	B-General_Concept
helps	O
expedite	O
the	O
development	O
of	O
the	O
prediction	O
model	O
,	O
tremendously	O
.	O
</s>
<s>
The	O
goal	O
of	O
this	O
method	O
is	O
to	O
summarize	O
and	O
present	O
the	O
information	O
in	O
a	O
useful	O
,	O
interactive	B-General_Concept
and	O
visually	O
appealing	O
way	O
in	O
order	O
to	O
understand	O
the	O
large	O
amounts	O
of	O
education	O
data	O
and	O
to	O
support	O
decision	O
making	O
.	O
</s>
<s>
Constructing	O
courseware	O
–	O
EDM	O
can	O
be	O
applied	O
to	O
course	O
management	O
systems	O
such	O
as	O
open	O
source	O
Moodle	B-Application
.	O
</s>
<s>
Moodle	B-Application
contains	O
usage	O
data	O
that	O
includes	O
various	O
activities	O
by	O
users	O
such	O
as	O
test	O
results	O
,	O
amount	O
of	O
readings	O
completed	O
and	O
participation	O
in	O
discussion	B-Application
forums	I-Application
.	O
</s>
<s>
Data	B-Application
mining	I-Application
tools	O
can	O
be	O
used	O
to	O
customize	O
learning	O
activities	O
for	O
each	O
user	O
and	O
adapt	O
the	O
pace	O
in	O
which	O
the	O
student	O
completes	O
the	O
course	O
.	O
</s>
<s>
New	O
research	O
on	O
mobile	O
learning	O
environments	O
also	O
suggests	O
that	O
data	B-Application
mining	I-Application
can	O
be	O
useful	O
.	O
</s>
<s>
Data	B-Application
mining	I-Application
can	O
be	O
used	O
to	O
help	O
provide	O
personalized	O
content	O
to	O
mobile	O
users	O
,	O
despite	O
the	O
differences	O
in	O
managing	O
content	O
between	O
mobile	B-Application
devices	I-Application
and	O
standard	O
PCs	B-Device
and	O
web	B-Application
browsers	I-Application
.	O
</s>
<s>
New	O
EDM	O
applications	O
will	O
focus	O
on	O
allowing	O
non-technical	O
users	O
use	O
and	O
engage	O
in	O
data	B-Application
mining	I-Application
tools	O
and	O
activities	O
,	O
making	O
data	O
collection	O
and	O
processing	O
more	O
accessible	O
for	O
all	O
users	O
of	O
EDM	O
.	O
</s>
<s>
In	O
October	O
2013	O
,	O
Coursera	O
offered	O
a	O
free	O
online	O
course	O
on	O
"	O
Big	B-Application
Data	I-Application
in	O
Education	O
"	O
that	O
taught	O
how	O
and	O
when	O
to	O
use	O
key	O
methods	O
for	O
EDM	O
.	O
</s>
<s>
This	O
course	O
moved	O
to	O
edX	B-Application
in	O
the	O
summer	O
of	O
2015	O
,	O
and	O
has	O
continued	O
to	O
run	O
on	O
edX	B-Application
annually	O
since	O
then	O
.	O
</s>
<s>
Considerable	O
amounts	O
of	O
EDM	O
work	O
are	O
published	O
at	O
the	O
peer-reviewed	O
International	O
Conference	O
on	O
Educational	B-Algorithm
Data	I-Algorithm
Mining	I-Algorithm
,	O
organized	O
by	O
the	O
International	O
Educational	B-Algorithm
Data	I-Algorithm
Mining	I-Algorithm
Society	O
.	O
</s>
<s>
EDM	O
papers	O
are	O
also	O
published	O
in	O
the	O
Journal	O
of	O
Educational	B-Algorithm
Data	I-Algorithm
Mining	I-Algorithm
(	O
JEDM	O
)	O
.	O
</s>
<s>
In	O
2011	O
,	O
Chapman	O
&	O
Hall/CRC	O
Press	O
,	O
Taylor	O
and	O
Francis	O
Group	O
published	O
the	O
first	O
Handbook	O
of	O
Educational	B-Algorithm
Data	I-Algorithm
Mining	I-Algorithm
.	O
</s>
<s>
This	O
resource	O
was	O
created	O
for	O
those	O
that	O
are	O
interested	O
in	O
participating	O
in	O
the	O
educational	B-Algorithm
data	I-Algorithm
mining	I-Algorithm
community	O
.	O
</s>
<s>
The	O
data	O
set	O
was	O
provided	O
by	O
the	O
DataShop	O
,	O
and	O
it	O
consisted	O
of	O
over	O
1,000,000	O
data	O
points	O
from	O
students	O
using	O
a	O
cognitive	B-Application
tutor	I-Application
.	O
</s>
<s>
Six	O
hundred	O
teams	O
competed	O
for	O
over	O
US$	O
8,000	O
in	O
prize	O
money	O
(	O
which	O
was	O
donated	O
by	O
Facebook	B-Application
)	O
.	O
</s>
<s>
The	O
winners	O
submitted	O
an	O
algorithm	O
that	O
utilized	O
feature	O
generation	O
(	O
a	O
form	O
of	O
representation	O
learning	O
)	O
,	O
random	B-Algorithm
forests	I-Algorithm
,	O
and	O
Bayesian	O
networks	O
.	O
</s>
<s>
Research	O
also	O
indicates	O
that	O
the	O
field	O
of	O
educational	B-Algorithm
data	I-Algorithm
mining	I-Algorithm
is	O
concentrated	O
in	O
western	O
countries	O
and	O
cultures	O
and	O
subsequently	O
,	O
other	O
countries	O
and	O
cultures	O
may	O
not	O
be	O
represented	O
in	O
the	O
research	O
and	O
findings	O
.	O
</s>
<s>
Privacy	O
–	O
Individual	O
privacy	O
is	O
a	O
continued	O
concern	O
for	O
the	O
application	O
of	O
data	B-Application
mining	I-Application
tools	O
.	O
</s>
<s>
As	O
users	O
become	O
savvy	O
in	O
their	O
understanding	O
of	O
online	O
privacy	O
,	O
administrators	O
of	O
educational	B-Algorithm
data	I-Algorithm
mining	I-Algorithm
tools	O
need	O
to	O
be	O
proactive	O
in	O
protecting	O
the	O
privacy	O
of	O
their	O
users	O
and	O
be	O
transparent	O
about	O
how	O
and	O
with	O
whom	O
the	O
information	O
will	O
be	O
used	O
and	O
shared	O
.	O
</s>
<s>
However	O
,	O
due	O
to	O
the	O
complexities	O
associated	O
with	O
detecting	O
and	O
preventing	O
digital	O
plagiarism	O
in	O
particular	O
,	O
educational	B-Algorithm
data	I-Algorithm
mining	I-Algorithm
tools	O
are	O
not	O
currently	O
sophisticated	O
enough	O
to	O
accurately	O
address	O
this	O
issue	O
.	O
</s>
