684.02 Spring 2007 05/22/071 Beyond PCFGs Chris Brew Ohio State University.

Slides:



Advertisements
Similar presentations
School of something FACULTY OF OTHER School of Computing FACULTY OF ENGINEERING Chunking: Shallow Parsing Eric Atwell, Language Research Group.
Advertisements

Albert Gatt Corpora and Statistical Methods Lecture 11.
Proceedings of the Conference on Intelligent Text Processing and Computational Linguistics (CICLing-2007) Learning for Semantic Parsing Advisor: Hsin-His.
Probabilistic Earley Parsing Charlie Kehoe, Spring 2004 Based on the 1995 paper by Andreas Stolcke: An Efficient Probabilistic Context-Free Parsing Algorithm.
10. Lexicalized and Probabilistic Parsing -Speech and Language Processing- 발표자 : 정영임 발표일 :
September PROBABILISTIC CFGs & PROBABILISTIC PARSING Universita’ di Venezia 3 Ottobre 2003.
1 Statistical NLP: Lecture 12 Probabilistic Context Free Grammars.
March 1, 2009 Dr. Muhammed Al-Mulhem 1 ICS 482 Natural Language Processing Probabilistic Context Free Grammars (Chapter 14) Muhammed Al-Mulhem March 1,
In Search of a More Probable Parse: Experiments with DOP* and the Penn Chinese Treebank Aaron Meyers Linguistics 490 Winter 2009.
Part II. Statistical NLP Advanced Artificial Intelligence Part of Speech Tagging Wolfram Burgard, Luc De Raedt, Bernhard Nebel, Lars Schmidt-Thieme Most.
Probabilistic Parsing: Enhancements Ling 571 Deep Processing Techniques for NLP January 26, 2011.
PCFG Parsing, Evaluation, & Improvements Ling 571 Deep Processing Techniques for NLP January 24, 2011.
Albert Gatt LIN3022 Natural Language Processing Lecture 8.
Amirkabir University of Technology Computer Engineering Faculty AILAB Efficient Parsing Ahmad Abdollahzadeh Barfouroush Aban 1381 Natural Language Processing.
Are Linguists Dinosaurs? 1.Statistical language processors seem to be doing away with the need for linguists. –Why do we need linguists when a machine.
1/13 Parsing III Probabilistic Parsing and Conclusions.
Features and Unification
Grammar induction by Bayesian model averaging Guy Lebanon LARG meeting May 2001 Based on Andreas Stolcke’s thesis UC Berkeley 1994.
1/17 Probabilistic Parsing … and some other approaches.
Language Model. Major role: Language Models help a speech recognizer figure out how likely a word sequence is, independent of the acoustics. A lot of.
Probabilistic Parsing Ling 571 Fei Xia Week 5: 10/25-10/27/05.
SI485i : NLP Set 9 Advanced PCFGs Some slides from Chris Manning.
Robert Hass CIS 630 April 14, 2010 NP NP↓ Super NP tagging JJ ↓
Part II. Statistical NLP Advanced Artificial Intelligence Applications of HMMs and PCFGs in NLP Wolfram Burgard, Luc De Raedt, Bernhard Nebel, Lars Schmidt-Thieme.
Empirical Methods in Information Extraction Claire Cardie Appeared in AI Magazine, 18:4, Summarized by Seong-Bae Park.
Tree Kernels for Parsing: (Collins & Duffy, 2001) Advanced Statistical Methods in NLP Ling 572 February 28, 2012.
1 Statistical NLP: Lecture 10 Lexical Acquisition.
Probabilistic Parsing Reading: Chap 14, Jurafsky & Martin This slide set was adapted from J. Martin, U. Colorado Instructor: Paul Tarau, based on Rada.
Chapter 10: Compilers and Language Translation Invitation to Computer Science, Java Version, Third Edition.
1 Statistical Parsing Chapter 14 October 2012 Lecture #9.
1 Natural Language Processing Lecture 11 Efficient Parsing Reading: James Allen NLU (Chapter 6)
10/12/2015CPSC503 Winter CPSC 503 Computational Linguistics Lecture 10 Giuseppe Carenini.
THE BIG PICTURE Basic Assumptions Linguistics is the empirical science that studies language (or linguistic behavior) Linguistics proposes theories (models)
SI485i : NLP Set 8 PCFGs and the CKY Algorithm. PCFGs We saw how CFGs can model English (sort of) Probabilistic CFGs put weights on the production rules.
1 Semi-Supervised Approaches for Learning to Parse Natural Languages Rebecca Hwa
Review 1.Lexical Analysis 2.Syntax Analysis 3.Semantic Analysis 4.Code Generation 5.Code Optimization.
11 Chapter 14 Part 1 Statistical Parsing Based on slides by Ray Mooney.
Page 1 Probabilistic Parsing and Treebanks L545 Spring 2000.
Coarse-to-Fine Efficient Viterbi Parsing Nathan Bodenstab OGI RPE Presentation May 8, 2006.
Statistical Decision-Tree Models for Parsing NLP lab, POSTECH 김 지 협.
CPSC 503 Computational Linguistics
For Friday Finish chapter 23 Homework –Chapter 23, exercise 15.
Supertagging CMSC Natural Language Processing January 31, 2006.
Natural Language Processing Lecture 15—10/15/2015 Jim Martin.
Parsing Chapter 15. The Job of a Parser Examine a string and decide whether or not it is a syntactically well-formed member of L(G), and If it is, assign.
CPSC 422, Lecture 27Slide 1 Intelligent Systems (AI-2) Computer Science cpsc422, Lecture 27 Nov, 16, 2015.
Building Sub-Corpora Suitable for Extraction of Lexico-Syntactic Information Ondřej Bojar, Institute of Formal and Applied Linguistics, ÚFAL.
Parsing and Code Generation Set 24. Parser Construction Most of the work involved in constructing a parser is carried out automatically by a program,
Wei Lu, Hwee Tou Ng, Wee Sun Lee National University of Singapore
GRAMMARS & PARSING. Parser Construction Most of the work involved in constructing a parser is carried out automatically by a program, referred to as a.
/02/20161 Probabilistic Context Free Grammars Chris Brew Ohio State University.
Dependency Parsing Niranjan Balasubramanian March 24 th 2016 Credits: Many slides from: Michael Collins, Mausam, Chris Manning, COLNG 2014 Dependency Parsing.
Natural Language Processing : Probabilistic Context Free Grammars Updated 8/07.
Roadmap Probabilistic CFGs –Handling ambiguity – more likely analyses –Adding probabilities Grammar Parsing: probabilistic CYK Learning probabilities:
Probabilistic and Lexicalized Parsing. Probabilistic CFGs Weighted CFGs –Attach weights to rules of CFG –Compute weights of derivations –Use weights to.
Natural Language Processing Vasile Rus
CSC 594 Topics in AI – Natural Language Processing
PRESENTED BY: PEAR A BHUIYAN
Basic Parsing with Context Free Grammars Chapter 13
4 (c) parsing.
Probabilistic and Lexicalized Parsing
CSCI 5832 Natural Language Processing
CSCI 5832 Natural Language Processing
Kanat Bolazar February 16, 2010
Parsing Unrestricted Text
CSCI 5832 Natural Language Processing
Chapter 10: Compilers and Language Translation
Probabilistic Parsing
David Kauchak CS159 – Spring 2019
Artificial Intelligence 2004 Speech & Natural Language Processing
Presentation transcript:

Spring /22/071 Beyond PCFGs Chris Brew Ohio State University

Spring /22/072 Beyond PCFGs  Shift-reduce parsers  probabilistic LR parsers  Data-oriented parsers

Spring /22/073 Motivation  Get round the limitations of the PCFG model  Exploit knowledge about individual words  Build better language models

Spring /22/074 Shift-reduce  Simple version either shift a word from the input list to the parse stack or reduce two elements from top of parse stack to a single tree  Hermjakob and Mooney cmp-lg –structures rather than just trees and words –more complex parse action language –not just binary rules

Spring /22/075 Machine learning for shift-reduce  supervisor shows the system correct sequences of parsing actions  system tries to learn to predict correct actions –needs a feature language  as it learns, the supervisor has less need to override the actions chosen by the system.

Spring /22/076 Examples of the feature language –broad syntactic class of the third element on the stack –the tense of the first element of the input list –Does top element of stack contain an object? –Could top frame be an adjectival degree adverb (e.g. very)? –Is frame1 a possible agent/patient of frame2? –Do frame1 and frame2 satisfy subject-verb agreement?

Spring /22/077 Hand-crafted knowledge used  205 features, all moderately local (no references to 1000th element of the stack or anything like that)  4356 node lexical knowledge base  subcategorisation table for 242 verbs  But we learn the association between features and actions

Spring /22/078 Various different hybrid decision structures  best was a hierarchical list of decision trees which encoded information about the task. Schematically. –decide whether to do anything »if not, we are done »if so, decide whether to do a reduction  if so, decide which reduction  if not, decide what sort of shift to do

Spring /22/079 Evaluation  Corpus of 272 annotated sentences.  17-fold cross validation (17 blocks of 16 sentences each)  Precision of 92.7%, recall of 92.8%: average length 17.1 words, with 43.5 parse actions per sentence. Parseval measures.  Correct structure and labelling 26.8% (i.e. 1 in 4 sentences are completely correct)

Spring /22/0710 Comments on Hermjakob and Mooney  A lot of grunt work needed - but not as much as full rationalist NLP system  The knowledge used is micro-modular very small pieces of highly independent knowledge  Test set is small, sentences short  Fairly robust  Good on small scale tests in an English/German MT task

Spring /22/0711 Probabilistic LR Parsers  Briscoe and Carroll CL 19(1) pp 25-59)  PCFGsgive these subtrees same probability N N N N N N N N N N

Spring /22/0712 LR Parsing  Builds a parsing table which gives parsing actions and Gotos for possible combinations of parser state and input symbols  There may be parsing action conflicts, in which more than one action is available.  In programming language grammars, you almost never want conflicts.  In NL grammars, you have no escape!

Spring /22/0713 Probabilistic LR  When there is a conflict, non- deterministically execute all possible actions.  But score them according to a probability distribution.  So where do the probabilities come from? And what do they mean? See analysis in Stolcke’s paper relating them to his forward and inner probabilities.

Spring /22/0714 LR parsing using Alvey Tools Grammar  Wide coverage unification grammar written by Claire Grover and Ted Briscoe  Build LR tables from CF backbone of this grammar  Interactively build disambiguated training corpus by supervising choice of parse actions

Spring /22/0715 Evaluation  Very good performance on LDOCE noun definitions 76% correct structure and labelling  State of the art results in later work on tag sequence grammars where the available lexical information is more restricted. (54% correct structure and labelling)  Work underway to bring this technique to Wall Street Journal data for comparison with other methods

Spring /22/0716 Data-oriented parsing  Rens Bod: Enriching Linguistics with Statistics: Performance Models of Natural Language, Amsterdam Ph.D  Treebank data again (this time ATIS sentences)  Radical rejection of context-free assumption  Count subtrees of arbitrary depth, not rule applications

Spring /22/0717 A corpus

Spring /22/0718 Tree fragments Some of the fragments S NP Matthew VP V likes NP Euan NP Euan V likes NP Matthew S NPVP S NPVP VNP S Matthew VP S NP Matthew VP VNP Euan

Spring /22/0719 The probability of a tree  The probability of all the ways of making it out of fragments  The probability of a fragment is given as a ratio between the frequency of the fragment and the total frequency of all fragments having that root

Spring /22/0720 Complexity  It’s hard to find efficient algorithms for sewing together DOP trees (cf. Si’maan for solutions)  Only very small corpora feasible  In practice, depth may have to be limited.  Many tree fragments are very rare, so there is an issue about smoothing

Spring /22/0721 Evaluation  several variations studied, DOP4 geta parse accuracies around 80% without a hand- coded dictionary, DOP-5 around 90% with.  results to be interpreted with caution due to small size of corpus  Evaluation on Dutch OVIS domain suggests that DOP is not competitive with Groningen’s more labour intensive system (but maybe that’s not the point)

Spring /22/0722 Where to find out more  Papers by Bod, Carroll, Hermjakob.  Manning and Schütze ch 12.  (subarea Computation and Language)