<s>
SemEval	B-General_Concept
(	O
Semantic	O
Evaluation	O
)	O
is	O
an	O
ongoing	O
series	O
of	O
evaluations	O
of	O
computational	B-General_Concept
semantic	I-General_Concept
analysis	I-General_Concept
systems	O
;	O
it	O
evolved	O
from	O
the	O
Senseval	B-General_Concept
word	O
sense	O
evaluation	O
series	O
.	O
</s>
<s>
The	O
purpose	O
of	O
the	O
SemEval	B-General_Concept
and	O
Senseval	B-General_Concept
exercises	O
is	O
to	O
evaluate	O
semantic	B-General_Concept
analysis	I-General_Concept
systems	O
.	O
</s>
<s>
"	O
Semantic	B-General_Concept
Analysis	I-General_Concept
"	O
refers	O
to	O
a	O
formal	O
analysis	O
of	O
meaning	O
,	O
and	O
"	O
computational	O
"	O
refer	O
to	O
approaches	O
that	O
in	O
principle	O
support	O
effective	O
implementation	O
.	O
</s>
<s>
The	O
first	O
three	O
evaluations	O
,	O
Senseval-1	O
through	O
Senseval-3	O
,	O
were	O
focused	O
on	O
word	B-General_Concept
sense	I-General_Concept
disambiguation	I-General_Concept
(	O
WSD	O
)	O
,	O
each	O
time	O
growing	O
in	O
the	O
number	O
of	O
languages	O
offered	O
in	O
the	O
tasks	O
and	O
in	O
the	O
number	O
of	O
participating	O
teams	O
.	O
</s>
<s>
Beginning	O
with	O
the	O
fourth	O
workshop	O
,	O
SemEval-2007	O
(	O
SemEval-1	O
)	O
,	O
the	O
nature	O
of	O
the	O
tasks	O
evolved	O
to	O
include	O
semantic	B-General_Concept
analysis	I-General_Concept
tasks	O
outside	O
of	O
word	B-General_Concept
sense	I-General_Concept
disambiguation	I-General_Concept
.	O
</s>
<s>
Triggered	O
by	O
the	O
conception	O
of	O
the	O
*	O
SEM	O
conference	O
,	O
the	O
SemEval	B-General_Concept
community	O
had	O
decided	O
to	O
hold	O
the	O
evaluation	O
workshops	O
yearly	O
in	O
association	O
with	O
the	O
*	O
SEM	O
conference	O
.	O
</s>
<s>
none	O
of	O
the	O
WSD	O
tasks	O
were	O
included	O
in	O
the	O
SemEval-2012	O
workshop	O
.	O
</s>
<s>
From	O
the	O
earliest	O
days	O
,	O
assessing	O
the	O
quality	O
of	O
word	B-General_Concept
sense	I-General_Concept
disambiguation	I-General_Concept
algorithms	O
had	O
been	O
primarily	O
a	O
matter	O
of	O
intrinsic	O
evaluation	O
,	O
and	O
“	O
almost	O
no	O
attempts	O
had	O
been	O
made	O
to	O
evaluate	O
embedded	O
WSD	O
components	O
”	O
.	O
</s>
<s>
Until	O
1990	O
or	O
so	O
,	O
discussions	O
of	O
the	O
sense	O
disambiguation	B-General_Concept
task	O
focused	O
mainly	O
on	O
illustrative	O
examples	O
rather	O
than	O
comprehensive	O
evaluation	O
.	O
</s>
<s>
In	O
April	O
1997	O
,	O
Martha	O
Palmer	O
and	O
Marc	O
Light	O
organized	O
a	O
workshop	O
entitled	O
Tagging	O
with	O
Lexical	O
Semantics	B-Application
:	O
Why	O
,	O
What	O
,	O
and	O
How	O
?	O
</s>
<s>
in	O
conjunction	O
with	O
the	O
Conference	O
on	O
Applied	O
Natural	B-Language
Language	I-Language
Processing	I-Language
.	O
</s>
<s>
At	O
the	O
time	O
,	O
there	O
was	O
a	O
clear	O
recognition	O
that	O
manually	O
annotated	O
corpora	O
had	O
revolutionized	O
other	O
areas	O
of	O
NLP	B-Language
,	O
such	O
as	O
part-of-speech	O
tagging	O
and	O
parsing	B-Language
,	O
and	O
that	O
corpus-driven	O
approaches	O
had	O
the	O
potential	O
to	O
revolutionize	O
automatic	O
semantic	B-General_Concept
analysis	I-General_Concept
as	O
well	O
.	O
</s>
<s>
Kilgarriff	O
recalled	O
that	O
there	O
was	O
“	O
a	O
high	O
degree	O
of	O
consensus	O
that	O
the	O
field	O
needed	O
evaluation	O
,	O
”	O
and	O
several	O
practical	O
proposals	O
by	O
Resnik	O
and	O
Yarowsky	O
kicked	O
off	O
a	O
discussion	O
that	O
led	O
to	O
the	O
creation	O
of	O
the	O
Senseval	B-General_Concept
evaluation	O
exercises	O
.	O
</s>
<s>
After	O
SemEval-2010	O
,	O
many	O
participants	O
feel	O
that	O
the	O
3-year	O
cycle	O
is	O
a	O
long	O
wait	O
.	O
</s>
<s>
For	O
this	O
reason	O
,	O
the	O
SemEval	B-General_Concept
coordinators	O
gave	O
the	O
opportunity	O
for	O
task	O
organizers	O
to	O
choose	O
between	O
a	O
2-year	O
or	O
a	O
3-year	O
cycle	O
.	O
</s>
<s>
The	O
SemEval	B-General_Concept
community	O
favored	O
the	O
3-year	O
cycle	O
.	O
</s>
<s>
Although	O
the	O
votes	O
within	O
the	O
SemEval	B-General_Concept
community	O
favored	O
a	O
3-year	O
cycle	O
,	O
organizers	O
and	O
coordinators	O
had	O
settled	O
to	O
split	O
the	O
SemEval	B-General_Concept
task	O
into	O
2	O
evaluation	O
workshops	O
.	O
</s>
<s>
The	O
SemEval	B-General_Concept
organizers	O
thought	O
it	O
would	O
be	O
appropriate	O
to	O
associate	O
our	O
event	O
with	O
the	O
*	O
SEM	O
conference	O
and	O
collocate	O
the	O
SemEval	B-General_Concept
workshop	O
with	O
the	O
*	O
SEM	O
conference	O
.	O
</s>
<s>
Thus	O
was	O
born	O
SemEval-2012	O
and	O
SemEval-2013	O
.	O
</s>
<s>
The	O
current	O
plan	O
is	O
to	O
switch	O
to	O
a	O
yearly	O
SemEval	B-General_Concept
schedule	O
to	O
associate	O
it	O
with	O
the	O
*	O
SEM	O
conference	O
but	O
not	O
every	O
task	O
needs	O
to	O
run	O
every	O
year	O
.	O
</s>
<s>
Senseval-1	O
took	O
place	O
in	O
the	O
summer	O
of	O
1998	O
for	O
English	O
,	O
French	O
,	O
and	O
Italian	O
,	O
culminating	O
in	O
a	O
workshop	O
held	O
at	O
Herstmonceux	O
Castle	O
,	O
Sussex	O
,	O
England	O
on	O
September	O
2	O
–	O
4	O
.	O
</s>
<s>
Senseval-2	O
took	O
place	O
in	O
the	O
summer	O
of	O
2001	O
,	O
and	O
was	O
followed	O
by	O
a	O
workshop	O
held	O
in	O
July	O
2001	O
in	O
Toulouse	O
,	O
in	O
conjunction	O
with	O
ACL	O
2001	O
.	O
</s>
<s>
Senseval-2	O
included	O
tasks	O
for	O
Basque	O
,	O
Chinese	O
,	O
Czech	O
,	O
Danish	O
,	O
Dutch	O
,	O
English	O
,	O
Estonian	O
,	O
Italian	O
,	O
Japanese	O
,	O
Korean	O
,	O
Spanish	O
and	O
Swedish	O
.	O
</s>
<s>
Senseval-3	O
took	O
place	O
in	O
March	O
–	O
April	O
2004	O
,	O
followed	O
by	O
a	O
workshop	O
held	O
in	O
July	O
2004	O
in	O
Barcelona	O
,	O
in	O
conjunction	O
with	O
ACL	O
2004	O
.	O
</s>
<s>
Senseval-3	O
included	O
14	O
different	O
tasks	O
for	O
core	O
word	B-General_Concept
sense	I-General_Concept
disambiguation	I-General_Concept
,	O
as	O
well	O
as	O
identification	O
of	O
semantic	O
roles	O
,	O
multilingual	O
annotations	O
,	O
logic	B-General_Concept
forms	I-General_Concept
,	O
subcategorization	O
acquisition	O
.	O
</s>
<s>
SemEval-2007	O
(	O
Senseval-4	O
)	O
took	O
place	O
in	O
2007	O
,	O
followed	O
by	O
a	O
workshop	O
held	O
in	O
conjunction	O
with	O
ACL	O
in	O
Prague	O
.	O
</s>
<s>
SemEval-2007	O
included	O
18	O
different	O
tasks	O
targeting	O
the	O
evaluation	O
of	O
systems	O
for	O
the	O
semantic	B-General_Concept
analysis	I-General_Concept
of	O
text	O
.	O
</s>
<s>
SemEval-2010	O
took	O
place	O
in	O
2010	O
,	O
followed	O
by	O
a	O
workshop	O
held	O
in	O
conjunction	O
with	O
ACL	O
in	O
Uppsala	O
.	O
</s>
<s>
SemEval-2010	O
included	O
18	O
different	O
tasks	O
targeting	O
the	O
evaluation	O
of	O
semantic	B-General_Concept
analysis	I-General_Concept
systems	O
.	O
</s>
<s>
SemEval-2012	O
took	O
place	O
in	O
2012	O
;	O
it	O
was	O
associated	O
with	O
the	O
new	O
*	O
SEM	O
,	O
First	O
Joint	O
Conference	O
on	O
Lexical	O
and	O
Computational	B-General_Concept
Semantics	I-General_Concept
,	O
and	O
co-located	O
with	O
NAACL	O
,	O
Montreal	O
,	O
Canada	O
.	O
</s>
<s>
SemEval-2012	O
included	O
8	O
different	O
tasks	O
targeting	O
at	O
evaluating	O
computational	O
semantic	O
systems	O
.	O
</s>
<s>
However	O
,	O
there	O
was	O
no	O
WSD	O
task	O
involved	O
in	O
SemEval-2012	O
,	O
the	O
WSD	O
related	O
tasks	O
were	O
scheduled	O
in	O
the	O
upcoming	O
SemEval-2013	O
.	O
</s>
<s>
SemEval-2013	O
was	O
associated	O
with	O
NAACL	O
2013	O
,	O
North	O
American	O
Association	O
of	O
Computational	O
Linguistics	O
,	O
Georgia	O
,	O
USA	O
and	O
took	O
place	O
in	O
2013	O
.	O
</s>
<s>
SemEval-2014	O
took	O
place	O
in	O
2014	O
.	O
</s>
<s>
It	O
was	O
co-located	O
with	O
COLING	O
2014	O
,	O
25th	O
International	O
Conference	O
on	O
Computational	O
Linguistics	O
and	O
*	O
SEM	O
2014	O
,	O
Second	O
Joint	O
Conference	O
on	O
Lexical	O
and	O
Computational	B-General_Concept
Semantics	I-General_Concept
,	O
Dublin	O
,	O
Ireland	O
.	O
</s>
<s>
There	O
were	O
10	O
different	O
tasks	O
in	O
SemEval-2014	O
evaluating	O
various	O
computational	O
semantic	O
systems	O
.	O
</s>
<s>
SemEval-2015	O
took	O
place	O
in	O
2015	O
.	O
</s>
<s>
It	O
was	O
co-located	O
with	O
NAACL-HLT	O
2015	O
,	O
2015	O
Conference	O
of	O
the	O
North	O
American	O
Chapter	O
of	O
the	O
Association	O
for	O
Computational	O
Linguistics	O
–	O
Human	O
Language	O
Technologies	O
and	O
*	O
SEM	O
2015	O
,	O
Third	O
Joint	O
Conference	O
on	O
Lexical	O
and	O
Computational	B-General_Concept
Semantics	I-General_Concept
,	O
Denver	O
,	O
USA	O
.	O
</s>
<s>
There	O
were	O
17	O
different	O
tasks	O
in	O
SemEval-2015	O
evaluating	O
various	O
computational	O
semantic	O
systems	O
.	O
</s>
<s>
The	O
framework	O
of	O
the	O
SemEval/Senseval	O
evaluation	O
workshops	O
emulates	O
the	O
Message	B-General_Concept
Understanding	I-General_Concept
Conferences	I-General_Concept
(	O
MUCs	O
)	O
and	O
other	O
evaluation	O
workshops	O
ran	O
by	O
ARPA	O
(	O
Advanced	O
Research	O
Projects	O
Agency	O
,	O
renamed	O
the	O
Defense	O
Advanced	O
Research	O
Projects	O
Agency	O
(	O
DARPA	O
)	O
)	O
.	O
</s>
<s>
Senseval-1	O
&	O
Senseval-2	O
focused	O
on	O
evaluation	O
WSD	O
systems	O
on	O
major	O
languages	O
that	O
were	O
available	O
corpus	O
and	O
computerized	O
dictionary	O
.	O
</s>
<s>
Senseval-3	O
looked	O
beyond	O
the	O
lexemes	O
and	O
started	O
to	O
evaluate	O
systems	O
that	O
looked	O
into	O
wider	O
areas	O
of	O
semantics	B-Application
,	O
such	O
as	O
Semantic	O
Roles	O
(	O
technically	O
known	O
as	O
Theta	O
roles	O
in	O
formal	O
semantics	B-Application
)	O
,	O
Logic	B-General_Concept
Form	I-General_Concept
Transformation	O
(	O
commonly	O
semantics	B-Application
of	O
phrases	O
,	O
clauses	O
or	O
sentences	O
were	O
represented	O
in	O
first-order	O
logic	B-General_Concept
forms	I-General_Concept
)	O
and	O
Senseval-3	O
explored	O
performances	O
of	O
semantics	B-Application
analysis	O
on	O
Machine	B-Application
translation	I-Application
.	O
</s>
<s>
As	O
the	O
types	O
of	O
different	O
computational	O
semantic	O
systems	O
grew	O
beyond	O
the	O
coverage	O
of	O
WSD	O
,	O
Senseval	B-General_Concept
evolved	O
into	O
SemEval	B-General_Concept
,	O
where	O
more	O
aspects	O
of	O
computational	O
semantic	O
systems	O
were	O
evaluated	O
.	O
</s>
<s>
The	O
SemEval	B-General_Concept
exercises	O
provide	O
a	O
mechanism	O
for	O
examining	O
issues	O
in	O
semantic	B-General_Concept
analysis	I-General_Concept
of	O
texts	O
.	O
</s>
<s>
The	O
topics	O
of	O
interest	O
fall	O
short	O
of	O
the	O
logical	O
rigor	O
that	O
is	O
found	O
in	O
formal	O
computational	B-General_Concept
semantics	I-General_Concept
,	O
attempting	O
to	O
identify	O
and	O
characterize	O
the	O
kinds	O
of	O
issues	O
relevant	O
to	O
human	O
understanding	O
of	O
language	O
.	O
</s>
<s>
The	O
first	O
major	O
area	O
in	O
semantic	B-General_Concept
analysis	I-General_Concept
is	O
the	O
identification	O
of	O
the	O
intended	O
meaning	O
at	O
the	O
word	O
level	O
(	O
taken	O
to	O
include	O
idiomatic	O
expressions	O
)	O
.	O
</s>
<s>
This	O
is	O
word-sense	B-General_Concept
disambiguation	I-General_Concept
(	O
a	O
concept	O
that	O
is	O
evolving	O
away	O
from	O
the	O
notion	O
that	O
words	O
have	O
discrete	O
senses	O
,	O
but	O
rather	O
are	O
characterized	O
by	O
the	O
ways	O
in	O
which	O
they	O
are	O
used	O
,	O
i.e.	O
,	O
their	O
contexts	O
)	O
.	O
</s>
<s>
The	O
tasks	O
in	O
this	O
area	O
include	O
lexical	O
sample	O
and	O
all-word	O
disambiguation	B-General_Concept
,	O
multi	O
-	O
and	O
cross-lingual	O
disambiguation	B-General_Concept
,	O
and	O
lexical	B-General_Concept
substitution	I-General_Concept
.	O
</s>
<s>
Given	O
the	O
difficulties	O
of	O
identifying	O
word	O
senses	O
,	O
other	O
tasks	O
relevant	O
to	O
this	O
topic	O
include	O
word-sense	B-General_Concept
induction	I-General_Concept
,	O
subcategorization	O
acquisition	O
,	O
and	O
evaluation	O
of	O
lexical	O
resources	O
.	O
</s>
<s>
The	O
second	O
major	O
area	O
in	O
semantic	B-General_Concept
analysis	I-General_Concept
is	O
the	O
understanding	O
of	O
how	O
different	O
sentence	O
and	O
textual	O
elements	O
fit	O
together	O
.	O
</s>
<s>
Other	O
tasks	O
in	O
this	O
area	O
look	O
at	O
more	O
specialized	O
issues	O
of	O
semantic	B-General_Concept
analysis	I-General_Concept
,	O
such	O
as	O
temporal	O
information	O
processing	O
,	O
metonymy	O
resolution	O
,	O
and	O
sentiment	O
analysis	O
.	O
</s>
<s>
The	O
tasks	O
in	O
this	O
area	O
have	O
many	O
potential	O
applications	O
,	O
such	O
as	O
information	B-General_Concept
extraction	I-General_Concept
,	O
question	B-Algorithm
answering	I-Algorithm
,	O
document	O
summarization	O
,	O
machine	B-Application
translation	I-Application
,	O
construction	O
of	O
thesauri	O
and	O
semantic	O
networks	O
,	O
language	O
modeling	O
,	O
paraphrasing	B-General_Concept
,	O
</s>
<s>
In	O
each	O
of	O
these	O
potential	O
applications	O
,	O
the	O
contribution	O
of	O
the	O
types	O
of	O
semantic	B-General_Concept
analysis	I-General_Concept
constitutes	O
the	O
most	O
outstanding	O
research	O
issue	O
.	O
</s>
<s>
For	O
example	O
,	O
in	O
the	O
word	B-General_Concept
sense	I-General_Concept
induction	I-General_Concept
and	I-General_Concept
disambiguation	I-General_Concept
task	I-General_Concept
,	O
there	O
are	O
three	O
separate	O
phases	O
:	O
</s>
<s>
No	O
other	O
resources	O
were	O
allowed	O
other	O
than	O
morphological	O
and	O
syntactic	O
Natural	B-Language
Language	I-Language
Processing	I-Language
components	O
,	O
such	O
as	O
morphological	O
analyzers	O
,	O
Part-Of-Speech	O
taggers	O
and	O
syntactic	O
parsers	B-Language
.	O
</s>
<s>
In	O
the	O
evaluation	O
phase	O
,	O
answers	O
of	O
to	O
the	O
testing	O
phase	O
were	O
evaluated	O
in	O
a	O
supervised	B-General_Concept
an	O
unsupervised	B-General_Concept
framework	O
.	O
</s>
<s>
The	O
unsupervised	B-General_Concept
evaluation	O
for	O
WSI	B-General_Concept
considered	O
two	O
types	O
of	O
evaluation	O
V	O
Measure	O
(	O
Rosenberg	O
and	O
Hirschberg	O
,	O
2007	O
)	O
,	O
and	O
paired	O
F-Score	O
(	O
Artiles	O
et	O
al.	O
,	O
2009	O
)	O
.	O
</s>
<s>
The	O
tables	O
below	O
reflects	O
the	O
workshop	O
growth	O
from	O
Senseval	B-General_Concept
to	O
SemEval	B-General_Concept
and	O
gives	O
an	O
overview	O
of	O
which	O
area	O
of	O
computational	B-General_Concept
semantics	I-General_Concept
was	O
evaluated	O
throughout	O
the	O
Senseval/SemEval	O
workshops	O
.	O
</s>
<s>
of	O
Tasks	O
Areas	O
of	O
study	O
Languages	O
of	O
Data	O
Evaluated	O
Senseval-1	O
3	O
Word	B-General_Concept
Sense	I-General_Concept
Disambiguation	I-General_Concept
(	O
WSD	O
)	O
-	O
Lexical	O
Sample	O
WSD	O
tasks	O
English	O
,	O
French	O
,	O
Italian	O
Senseval-2	O
12	O
Word	B-General_Concept
Sense	I-General_Concept
Disambiguation	I-General_Concept
(	O
WSD	O
)	O
-	O
Lexical	O
Sample	O
,	O
All	O
Words	O
,	O
Translation	O
WSD	O
tasks	O
Czech	O
,	O
Dutch	O
,	O
English	O
,	O
Estonian	O
,	O
Basque	O
,	O
Chinese	O
,	O
Danish	O
,	O
English	O
,	O
Italian	O
,	O
Japanese	O
,	O
Korean	O
,	O
Spanish	O
,	O
Swedish	O
Senseval-3	O
16	O
(	O
incl	O
.	O
</s>
<s>
2	O
cancelled	O
)	O
Logic	B-General_Concept
Form	I-General_Concept
Transformation	O
,	O
Machine	B-Application
Translation	I-Application
(	O
MT	O
)	O
Evaluation	O
,	O
Semantic	O
Role	O
Labelling	O
,	O
WSD	O
Basque	O
,	O
Catalan	O
,	O
Chinese	O
,	O
English	O
,	O
Italian	O
,	O
Romanian	O
,	O
Spanish	O
SemEval2007	O
19	O
(	O
incl	O
.	O
</s>
<s>
1	O
cancelled	O
)	O
Cross-lingual	O
,	O
Frame	O
Extraction	O
,	O
Information	B-General_Concept
Extraction	I-General_Concept
,	O
Lexical	B-General_Concept
Substitution	I-General_Concept
,	O
Lexical	O
Sample	O
,	O
Metonymy	O
,	O
Semantic	O
Annotation	O
,	O
Semantic	O
Relations	O
,	O
Semantic	O
Role	O
Labelling	O
,	O
Sentiment	O
Analysis	O
,	O
Time	O
Expression	O
,	O
WSD	O
Arabic	O
,	O
Catalan	O
,	O
Chinese	O
,	O
English	O
,	O
Spanish	O
,	O
Turkish	O
SemEval2010	O
18	O
(	O
incl	O
.	O
</s>
<s>
1	O
cancelled	O
)	O
Coreference	O
,	O
Cross-lingual	O
,	O
Ellipsis	O
,	O
Information	B-General_Concept
Extraction	I-General_Concept
,	O
Lexical	B-General_Concept
Substitution	I-General_Concept
,	O
Metonymy	O
,	O
Noun	O
Compounds	O
,	O
Parsing	B-Language
,	O
Semantic	O
Relations	O
,	O
Semantic	O
Role	O
Labeling	O
,	O
Sentiment	O
Analysis	O
,	O
Textual	O
Entailment	O
,	O
Time	O
Expressions	O
,	O
WSD	O
Catalan	O
,	O
Chinese	O
,	O
Dutch	O
,	O
English	O
,	O
French	O
,	O
German	O
,	O
Italian	O
,	O
Japanese	O
,	O
Spanish	O
SemEval2012	O
8	O
Common	O
Sense	O
Reasoning	O
,	O
Lexical	B-General_Concept
Simplification	I-General_Concept
,	O
Relational	O
Similarity	O
,	O
Spatial	O
Role	O
Labelling	O
,	O
Semantic	O
Dependency	O
Parsing	B-Language
,	O
Semantic	O
and	O
Textual	O
Similarity	O
Chinese	O
,	O
English	O
SemEval2013	O
14	O
Temporal	B-General_Concept
Annotation	I-General_Concept
,	O
Sentiment	O
Analysis	O
,	O
Spatial	O
Role	O
Labeling	O
,	O
Noun	O
Compounds	O
,	O
Phrasal	O
Semantics	B-Application
,	O
Textual	O
Similarity	O
,	O
Response	O
Analysis	O
,	O
Cross-lingual	O
Textual	O
Entailment	O
,	O
BioMedical	O
Texts	O
,	O
Cross	O
and	O
Multilingual	O
WSD	O
,	O
Word	B-General_Concept
Sense	I-General_Concept
Induction	I-General_Concept
,	O
and	O
Lexical	O
Sample	O
Catalan	O
,	O
French	O
,	O
German	O
,	O
English	O
,	O
Italian	O
,	O
Spanish	O
SemEval2014	O
10	O
Compositional	O
Distributional	O
Semantic	O
,	O
Grammar	O
Induction	O
for	O
Spoken	O
Dialogue	O
Systems	O
,	O
Cross-Level	O
Semantic	O
Similarity	O
,	O
Sentiment	O
Analysis	O
,	O
L2	O
Writing	O
Assistant	O
,	O
Supervised	B-General_Concept
Semantic	B-Application
Parsing	I-Application
,	O
Clinical	O
Text	O
Analysis	O
,	O
Semantic	O
Dependency	O
Parsing	B-Language
,	O
Sentiment	O
Analysis	O
in	O
Twitter	O
,	O
Multilingual	O
Semantic	O
Textual	O
Similarity	O
English	O
,	O
Spanish	O
,	O
French	O
,	O
German	O
,	O
Dutch	O
,	O
SemEval2015	O
18	O
(	O
incl	O
.	O
</s>
<s>
The	O
Multilingual	O
WSD	O
task	O
was	O
introduced	O
for	O
the	O
SemEval-2013	O
workshop	O
.	O
</s>
<s>
The	O
task	O
is	O
aimed	O
at	O
evaluating	O
Word	B-General_Concept
Sense	I-General_Concept
Disambiguation	I-General_Concept
systems	O
in	O
a	O
multilingual	O
scenario	O
using	O
BabelNet	B-Application
as	O
its	O
sense	O
inventory	O
.	O
</s>
<s>
Unlike	O
similar	O
task	O
like	O
crosslingual	O
WSD	O
or	O
the	O
multilingual	B-General_Concept
lexical	I-General_Concept
substitution	I-General_Concept
task	O
,	O
where	O
no	O
fixed	O
sense	O
inventory	O
is	O
specified	O
,	O
Multilingual	O
WSD	O
uses	O
the	O
BabelNet	B-Application
as	O
its	O
sense	O
inventory	O
.	O
</s>
<s>
Prior	O
to	O
the	O
development	O
of	O
BabelNet	B-Application
,	O
a	O
bilingual	O
lexical	O
sample	O
WSD	O
evaluation	O
task	O
was	O
carried	O
out	O
in	O
SemEval-2007	O
on	O
Chinese-English	O
bitexts	O
.	O
</s>
<s>
To	O
facilitate	O
the	O
ease	O
of	O
integrating	O
WSD	O
systems	O
into	O
other	O
Natural	B-Language
Language	I-Language
Processing	I-Language
(	O
NLP	B-Language
)	O
applications	O
,	O
such	O
as	O
Machine	B-Application
Translation	I-Application
and	O
multilingual	O
Information	B-Library
Retrieval	I-Library
,	O
the	O
cross-lingual	O
WSD	O
evaluation	O
task	O
was	O
introduced	O
a	O
language-independent	O
and	O
knowledge-lean	O
approach	O
to	O
WSD	O
.	O
</s>
<s>
The	O
task	O
is	O
an	O
unsupervised	B-General_Concept
Word	B-General_Concept
Sense	I-General_Concept
Disambiguation	I-General_Concept
task	O
for	O
English	O
nouns	O
by	O
means	O
of	O
parallel	O
corpora	O
.	O
</s>
<s>
It	O
is	O
worth	O
noting	O
that	O
the	O
SemEval-2014	O
have	O
only	O
two	O
tasks	O
that	O
were	O
multilingual/crosslingual	O
,	O
i.e.	O
</s>
<s>
The	O
major	O
tasks	O
in	O
semantic	O
evaluation	O
include	O
the	O
following	O
areas	O
of	O
natural	B-Language
language	I-Language
processing	I-Language
.	O
</s>
<s>
The	O
following	O
table	O
shows	O
the	O
areas	O
of	O
studies	O
that	O
were	O
involved	O
in	O
Senseval-1	O
through	O
SemEval-2014	O
(	O
S	O
refers	O
to	O
Senseval	B-General_Concept
and	O
SE	O
refers	O
to	O
SemEval	B-General_Concept
,	O
e.g.	O
</s>
<s>
S1	O
refers	O
to	O
Senseval-1	O
and	O
SE07	O
refers	O
to	O
SemEval2007	O
)	O
:	O
</s>
<s>
SemEval	B-General_Concept
tasks	O
have	O
created	O
many	O
types	O
of	O
semantic	O
annotations	O
,	O
each	O
type	O
with	O
various	O
schema	O
.	O
</s>
<s>
In	O
SemEval-2015	O
,	O
the	O
organizers	O
have	O
decided	O
to	O
group	O
tasks	O
together	O
into	O
several	O
tracks	O
.	O
</s>
<s>
Here	O
lists	O
the	O
type	O
of	O
semantic	O
annotations	O
involved	O
in	O
the	O
SemEval	B-General_Concept
workshops	O
:	O
</s>
<s>
the	O
taxonomy	O
evaluation	O
task	O
in	O
SemEval-2015	O
was	O
under	O
the	O
Learning	O
Semantic	O
Relations	O
track	O
and	O
in	O
SemEval-2016	O
,	O
there	O
is	O
a	O
dedicated	O
track	O
for	O
Semantic	O
Taxonomy	O
with	O
a	O
new	O
Semantic	O
Taxonomy	O
Enrichment	O
task	O
.	O
</s>
