In order to build robust
<term>
automatic abstracting systems
</term>
, there is a need for better
<term>
training resources
</term>
than are currently available .
#30160In order to build robustautomatic abstracting systems, there is a need for better training resources than are currently available.
lr,15-1-E99-1015,ak
In order to build robust
<term>
automatic abstracting systems
</term>
, there is a need for better
<term>
training resources
</term>
than are currently available .
#30170In order to build robust automatic abstracting systems, there is a need for bettertraining resources than are currently available.
model,7-2-E99-1015,ak
In this paper , we introduce an
<term>
annotation scheme
</term>
for scientific articles which can be used to build such a
<term>
resource
</term>
in a consistent way .
#30184In this paper, we introduce anannotation scheme for scientific articles which can be used to build such a resource in a consistent way.
lr,20-2-E99-1015,ak
In this paper , we introduce an
<term>
annotation scheme
</term>
for scientific articles which can be used to build such a
<term>
resource
</term>
in a consistent way .
#30197In this paper, we introduce an annotation scheme for scientific articles which can be used to build such aresource in a consistent way.
model,5-3-E99-1015,ak
The seven categories of the
<term>
scheme
</term>
are based on
<term>
rhetorical moves of argumentation
</term>
.
#30208The seven categories of thescheme are based on rhetorical moves of argumentation.
other,9-3-E99-1015,ak
The seven categories of the
<term>
scheme
</term>
are based on
<term>
rhetorical moves of argumentation
</term>
.
#30212The seven categories of the scheme are based onrhetorical moves of argumentation.
model,6-4-E99-1015,ak
Our experimental results show that the
<term>
scheme
</term>
is stable , reproducible and intuitive to use .
#30223Our experimental results show that thescheme is stable, reproducible and intuitive to use.