<s>
Collocation	B-General_Concept
extraction	I-General_Concept
is	O
the	O
task	O
of	O
using	O
a	O
computer	O
to	O
extract	O
collocations	B-General_Concept
automatically	O
from	O
a	O
corpus	O
.	O
</s>
<s>
The	O
traditional	O
method	O
of	O
performing	O
collocation	B-General_Concept
extraction	I-General_Concept
is	O
to	O
find	O
a	O
formula	O
based	O
on	O
the	O
statistical	O
quantities	O
of	O
those	O
words	O
to	O
calculate	O
a	O
score	O
associated	O
to	O
every	O
word	O
pairs	O
.	O
</s>
<s>
Proposed	O
formulas	O
are	O
mutual	O
information	O
,	O
t-test	B-General_Concept
,	O
z	B-General_Concept
test	I-General_Concept
,	O
chi-squared	B-General_Concept
test	I-General_Concept
and	O
likelihood	B-General_Concept
ratio	I-General_Concept
.	O
</s>
<s>
Within	O
the	O
area	O
of	O
corpus	O
linguistics	O
,	O
collocation	B-General_Concept
is	O
defined	O
as	O
a	O
sequence	O
of	O
words	O
or	O
terms	B-Language
which	O
co-occur	O
more	O
often	O
than	O
would	O
be	O
expected	O
by	O
chance	O
.	O
</s>
