1 / 8

Lexical Dependency Parsing

Lexical Dependency Parsing. Chris Brew. OhioState University. Using Lexical Dependencies. Lexical information crucial to parser success Original version is Magerman’s SPATTER Each is simpler than the last Often also with improved performance. The Task.

Download Presentation

Lexical Dependency Parsing

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Lexical Dependency Parsing Chris Brew OhioState University

  2. Using Lexical Dependencies • Lexical information crucial to parser success • Original version is Magerman’s SPATTER • Each is simpler than the last • Often also with improved performance

  3. The Task • Generate trees as in Wall Street Journal part of Penn Treebank • Collins provides new statistical model for P(T|S) • PCFGs used rules, DOP used fragments, LR used parser states • This uses Bigram Lexical Dependencies plus a few extras

  4. The components of the model • A model of base NPs P(B|S) • Obtained using bigram statistics and POS tags • A model of dependencies P(D|S,B) • A bijective mapping which can interconvert between • Trees • Pairings of base NP structure and dependencies

  5. A parse tree • Base NPs [John Smith][the president][IBM] [his resignation] [yesterday] • Treebank is linguistically odd here

  6. S(announced) NP(Smith) VP(announced) NP(Smith) NP(president) VBD NP(resignation) NP NNP NNP NP PP(of) announced PRP$ NN NN John Smith DT NN IN NP his resignation yesterday the president of NNP IBM Propagating head words • Small set of rules propagate heads

  7. VBD VP NP VP VBD NP announced [his Resignation] [yesterday] Extracted structure • Base NPs plus Dependencies • Dependencies labelled with triplesof nonterminals NB. Not all dependencies shown here NP S VP NP NP NP [John Smith] [the president] of [IBM]

  8. Statistical model • Gives probabilities to dependencies • So the probability of a rule like VP -> VBD NP NP, which involves two dependencies, is made from the probabilities of the components. VBD VP NP VBD VP NP announced [his Resignation] [yesterday]

More Related