Title
On the use of different loss functions in statistical pattern recognition applied to machine translation
Abstract
In pattern recognition, an elegant and powerful way to deal with classification problems is based on the minimisation of the classification risk. The risk function is defined in terms of loss functions that measure the penalty for wrong decisions. However, in practice a trivial loss function is usually adopted (the so-called 0-1 loss function) that do no make the most of this framework. This work is focused on the study of different loss functions, and specially on those loss functions that do not depend on the class proposed by the system. Loss functions of this kind have allowed us to theoretically explain heuristics that are successfully used with very complex pattern recognition problem, such as (statistical) machine translation. A comparative experimental work has also been carried out to compare different proposals of loss functions in the practical scenario of machine translation.
Year
DOI
Venue
2008
10.1016/j.patrec.2007.06.015
Pattern Recognition Letters
Keywords
Field
DocType
trivial loss function,complex pattern recognition problem,direct translation rule,classification rules,statistical pattern recognition,comparative experimental work,classification risk,different loss function,different proposal,statistical machine translation,loss function,classification problem,machine translation,decision theory,bayes’ risk,risk function,pattern recognition
Pattern recognition,Computer science,Machine translation,Heuristics,Minimisation (psychology),Decision theory,Artificial intelligence,Risk function,Automatic translation,Pattern recognition problem
Journal
Volume
Issue
ISSN
29
8
Pattern Recognition Letters
Citations 
PageRank 
References 
0
0.34
18
Authors
4
Name
Order
Citations
PageRank
Jesús Andrés-Ferrer1737.52
D. Ortiz-Martínez220.74
I. García-Varea3483.44
francisco casacuberta41439161.33