<s>
Inductive	B-Algorithm
miner	I-Algorithm
belongs	O
to	O
a	O
class	O
of	O
algorithms	O
used	O
in	O
process	B-Algorithm
discovery	I-Algorithm
.	O
</s>
<s>
A	O
number	O
of	O
techniques	O
such	O
as	O
alpha	B-Algorithm
miner	I-Algorithm
,	O
genetic	O
miner	O
,	O
work	O
on	O
the	O
basis	O
of	O
converting	O
an	O
event	O
log	O
into	O
a	O
workflow	O
model	O
,	O
however	O
,	O
they	O
do	O
not	O
produce	O
models	O
that	O
are	O
sound	O
all	O
the	O
time	O
.	O
</s>
<s>
Inductive	B-Algorithm
miner	I-Algorithm
relies	O
on	O
building	O
a	O
directly	O
follows	O
graph	O
from	O
event	O
log	O
and	O
using	O
this	O
graph	O
to	O
detect	O
various	O
process	O
relations	O
.	O
</s>
<s>
The	O
inductive	B-Algorithm
miner	I-Algorithm
technique	O
relies	O
on	O
the	O
detection	O
of	O
various	O
cuts	O
on	O
the	O
directly	O
follows	O
graph	O
created	O
using	O
the	O
event	O
log	O
.	O
</s>
<s>
The	O
core	O
idea	O
behind	O
inductive	B-Algorithm
miner	I-Algorithm
lies	O
in	O
the	O
unique	O
methodology	O
of	O
discovering	O
various	O
divisions	O
of	O
the	O
arcs	O
in	O
the	O
directly-follows	O
graph	O
,	O
and	O
using	O
the	O
smaller	O
components	O
after	O
division	O
to	O
represent	O
the	O
execution	O
sequence	O
of	O
the	O
activities.The	O
inductive	B-Algorithm
miner	I-Algorithm
algorithm	O
uses	O
the	O
directly-follows	O
graph	O
to	O
detect	O
one	O
of	O
the	O
following	O
cuts	O
.	O
</s>
<s>
Inductive	B-Algorithm
miner	I-Algorithm
with	O
fall	O
through	O
:	O
The	O
complex	O
event	O
log	O
sometimes	O
would	O
make	O
it	O
impossible	O
to	O
detect	O
any	O
cuts	O
using	O
the	O
above	O
techniques	O
.	O
</s>
<s>
Inductive	B-Algorithm
miner	I-Algorithm
frequency-based	O
:	O
The	O
less	O
frequent	O
relations	O
in	O
the	O
event	O
log	O
sometimes	O
creates	O
problems	O
in	O
detecting	O
any	O
type	O
of	O
cuts	O
.	O
</s>
<s>
Inductive	B-Algorithm
miner	I-Algorithm
for	O
big	O
data	O
:	O
This	O
includes	O
an	O
improvement	O
on	O
the	O
existing	O
inductive	B-Algorithm
miner	I-Algorithm
to	O
handle	O
big	O
data	O
sets	O
.	O
</s>
