Natural Language Processing

1
Natural Language Processing

• Lecture Notes 8
• (Though most of this topic covered in class on
  the board)

2
Example PCFG
3
Ambiguous Sentence
P(TL) 1.5 x 10-6 P(TR) 1.7 x 10-6 P(S)
3.2 x 10-6
4
Lack of Sensitivity to the properties of
individual Words

• Lexical information can play an important role in
  selecting between alternative interpretations.
  Consider sentence
• "Moscow sent soldiers into Afghanistan."
• NP ? NP PPVP ? VP PP
• These give rise to 2 parse trees

5
PP Attachment Ambiguity
67 of PPs attach to NPs
33 of PPs attach to VPs
S
VP VP NP
NP
PP
NP
NP
N V N P
N Moscow sent soldiers into Afghanistan
N V N P
N Moscow sent soldiers into Afghanistan
6
Lexicalized Tree
7
Example (wrong)