P a e i 1j the probability of an alignment given

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: system, developed at IBM •  Simple genera1ve model to produce F given E=e1, e2, …eI •  Choose J, the number of words in F: F=f1, f2, …fJ •  Choose a 1 ­to ­many alignment A=a1, a2, …aJ •  For each posi1on in F, generate a word fj from the aligned word in E: ea j Dan Jurafsky IBM Model 1: Genera$ve Process 0 1 2 3 4 5 6 NULL J=9 Mary didn’t slap the green witch Maria no dió una bofetada a la bruja verde A= 1 2 3 3 3 0 4 6 5 1.  Choose J, the number of words in F: F=f1, f2, …fJ 2.  Choose a 1 ­to ­many alignment A=a1, a2, …aJ 3.  For each posi1on in F, generate a word fj from the aligned word in E: ea j Dan Jurafsky Compu$ng P(F | E) in IBM Model 1: P(F|E,A) •  Let ea j : the English word assigned to Spanish word fj t(fx,ey): probability of transla1ng ey as fx •  If we knew E, the alignment A, and J, then: J P( F | E, A) = ! t ( f j , ea j ) j =1 •  The probability of the Spanish sentence if we knew the English source, the alignment, and J Dan Jurafsky Compu$ng P(F | E) in IBM Model 1: P(A|E) •  A normaliza1on factor, since there are (I + 1)J possible alignments: ! P( A | E ) = ( I + 1)J •  The probability of an alignment given the English sentence. Dan Jurafsky Compu$ng P(F | E) in IBM Model 1: P(F,A|E) and then P(F|E) ! P( A | E ) = ( I + 1)J J P( F | E, A) = ! t ( f j , ea j ) j =1 The probability of genera1ng F through a par1cular alignment: J ! P ( F, A | E ) = t ( f j , ea ) J! ( I + 1) j=1 To...
View Full Document

This document was uploaded on 02/14/2014.

Ask a homework question - tutors are online