{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

lect18-unsup - S t a t is t ic a l M o d e ls o f S e m a n...

Info iconThis preview shows pages 1–38. Sign up to view the full content.

View Full Document Right Arrow Icon
Statistical Models of Semantics and Unsupervised Language Discovery Lecture #18 Introduction to Natural Language Processing CMPSCI 585, Fall 2007 Andrew McCallum Computer Science Department University of Massachusetts Amherst Including slides from Chris Manning, Dan Klein, Rion Snow & Patrick Pantel.
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Attachment Ambiguity Where to attach a phrase in the parse tree? “I saw the man with the telescope.” – What does “with a telescope” modify? – Is the problem AI complete? Yes, but… – Proposed simple structural factors Right association [Kimball 1973] ‘low’ or ‘near’ attachment = ‘early closure’ of NP Minimal attachment [Frazier 1978] (depends on grammar) = ‘high’ or ‘distant’ attachment = ‘late closure’ (of NP)
Background image of page 2
Attachment Ambiguity “The children ate the cake with a spoon .” “The children ate the cake with frosting .” “Joe included the package for Susan .” “Joe carried the package for Susan .” Ford, Bresnan and Kaplan (1982): “It is quite evident, then, that the closure effects in these sentences are induced in some way by the choice of the lexical items.”
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Lexical acquisition, semantic similarity Previous models give same estimate to all unseen events. Unrealistic - could hope to refine that based on semantic classes of words • Examples – “Susan ate the cake with a durian.” – “Susan had never eaten a fresh durian before.” – Although never seen “eating pineapple” should be more likely than “eating holograms” because pineapple is similar to apples, and we have seen “eating apples”.
Background image of page 4
An application: selectional preferences Most verbs prefer arguments of a particular type. Such regularities are called selectional preferences or selectional restrictions . “Bill drove a…” Mustang, car, truck, jeep Selectional preference strength: how strongly does a verb constrain direct objects “see” versus “unknotted”
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Measuring selectional preference strength Assume we are given a clustering of (direct object) nouns. Resnick (1993) uses WordNet. Selectional association between a verb and a class Proportion that its summand contributes to preference strength. For nouns in multiple classes, disambiguate as most likely sense:
Background image of page 6
Selection preference strength (made up data) Noun class c P(c) P(c|eat) P(c|see) P(c|find) people 0.25 0.01 0.25 0.33 furniture 0.25 0.01 0.25 0.33 food 0.25 0.97 0.25 0.33 action 0.25 0.01 0.25 0.01 SPS S(v) 1.76 0.00 0.35 A(eat, food) = 1.08 A(find, action) = -0.13
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Selectional Preference Strength example (Resnick, Brown corpus)
Background image of page 8
But how might we measure word similarity for word classes? Vector spaces
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
But how might we measure word similarity for word classes? Vector spaces word-by-word matrix B
Background image of page 10
Similarity measures for binary vectors
Background image of page 11

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Cosine measure
Background image of page 12
Example of cosine measure on word-by-word matrix on NYT
Background image of page 13

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Probabilistic measures
Background image of page 14
Neighbors of word “company” [Lee]
Background image of page 15

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Learning syntactic patterns for automatic hypernym discovery Rion Snow, Daniel Jurafsky, and Andrew Y. Ng.
Background image of page 16
Background image of page 17

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Background image of page 18
Background image of page 19

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Background image of page 20
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}