101
Voted
ACL
2012
13 years 2 months ago
2012
Many machine translation (MT) evaluation metrics have been shown to correlate better with human judgment than BLEU. In principle, tuning on these metrics should yield better syste...
118
Voted
ICEIS
2010
IEEE
14 years 10 months ago
2010
IEEE
Regularities in the world are human defined. Patterns in the observed phenomena are there because we define and recognize them as such. Automatic pattern recognition tries to bridg...
121
Voted
COLING
2002
15 years 6 days ago
2002
tion. Our method was developed through the study of a corpus of abstracts written ssional abstractors. Relying on human judgment, we have evaluated indicativeness, informativeness,...
BIOSYSTEMS
2008
15 years 15 days ago
2008
: Knowing that human judgment can be fallible, we propose to distinguish the subjective ascription of a property, such as autonomy, from the genuine fact that an entity is characte...