Clayton Greenberg Princeton University April 21, 2013.

Slides:



Advertisements
Similar presentations
Pasco-Hernando Community College Tutorial Series.
Advertisements

Week 3b. Merge, feature checking CAS LX 522 Syntax I.
Composition 9 Sentences.
1 Developing Statistic-based and Rule-based Grammar Checkers for Chinese ESL Learners Howard Chen Department of English National Taiwan Normal University.
1 LIN 1310B Introduction to Linguistics Prof: Nikolay Slavkov TA: Qinghua Tang CLASS 18, March 13, 2007.
Probabilistic Parsing: Enhancements Ling 571 Deep Processing Techniques for NLP January 26, 2011.
Intelligent Systems (AI-2) Computer Science cpsc422, Lecture 27
Faculty Of Applied Science Simon Fraser University Cmpt 825 presentation Corpus Based PP Attachment Ambiguity Resolution with a Semantic Dictionary Jiri.
Inducing Information Extraction Systems for New Languages via Cross-Language Projection Ellen Riloff University of Utah Charles Schafer, David Yarowksy.
Syntax LING October 11, 2006 Joshua Tauberer.
1/17 Acquiring Selectional Preferences from Untagged Text for Prepositional Phrase Attachment Disambiguation Hiram Calvo and Alexander Gelbukh Presented.
Designed by Elisa Paramore
Syntax.
SI485i : NLP Set 9 Advanced PCFGs Some slides from Chris Manning.
Misplaced and Dangling Modifiers. Misplaced Modifiers Modifier ◦ Describes, clarifies, or gives more detail about other words in a sentence ◦ Can be a.
Participles and Gerunds. Verbal Phrases Participles end in either –ing or –ed, or are irregularly formed. Act as adjectives HINT: Can be taken out of.
Albert Gatt LIN 3098 Corpus Linguistics. In this lecture Some more on corpora and grammar Construction Grammar as a theoretical framework Collostructional.
PARSING David Kauchak CS457 – Fall 2011 some slides adapted from Ray Mooney.
Robert Hass CIS 630 April 14, 2010 NP NP↓ Super NP tagging JJ ↓
Thumbs Up or Thumbs Down? Semantic Orientation Applied to Unsupervised Classification on Reviews Peter D. Turney Institute for Information Technology National.
BİL711 Natural Language Processing1 Statistical Parse Disambiguation Problem: –How do we disambiguate among a set of parses of a given sentence? –We want.
Probabilistic Parsing Reading: Chap 14, Jurafsky & Martin This slide set was adapted from J. Martin, U. Colorado Instructor: Paul Tarau, based on Rada.
S HIFTING W ORD O RDER IN THE S ENTENCE Lesson 28 By Joseph C. Blumenthal.
Subjects and Verbs. What is a subject? What is a verb?
Jennie Ning Zheng Linda Melchor Ferhat Omur. Contents Introduction WordNet Application – WordNet Data Structure - WordNet FrameNet Application – FrameNet.
Improving Subcategorization Acquisition using Word Sense Disambiguation Anna Korhonen and Judith Preiss University of Cambridge, Computer Laboratory 15.
Arabic Tokenization, Part-of-Speech Tagging and Morphological Disambiguation in One Fell Swoop Nizar Habash and Owen Rambow Center for Computational Learning.
12/06/1999 JHU CS /Jan Hajic 1 Introduction to Natural Language Processing ( ) Statistical Parsing Dr. Jan Hajič CS Dept., Johns Hopkins Univ.
A Cascaded Finite-State Parser for German Michael Schiehlen Institut für Maschinelle Sprachverarbeitung Universität Stuttgart
Unsupervised Word Sense Disambiguation REU, Summer, 2009.
Analyzing Transcripts: Creating Needs. 2 Confidential and Proprietary Applied Marketing Science, Inc. Identifying Needs In this part of the VOC process.
Correlative Conjunctions Created by Kathryn Reilly.
Misuse of Articles By: Liz M. LaboyWorkshop four Albanice FloresProf. C. Garcia Jennifer M. Serrano ENGL 245.
In Need of Modification Correcting and Avoiding Misplaced or Dangling Modifiers.
Grammars Grammars can get quite complex, but are essential. Syntax: the form of the text that is valid Semantics: the meaning of the form – Sometimes semantics.
Rules, Movement, Ambiguity
CSA2050 Introduction to Computational Linguistics Parsing I.
The FRAME Routine Key Topic is about… So What? (What’s important to understand about this?) Main idea.
Misplaced Modifiers 11/18/14. Misplaced Modifiers A word, phrase, or clause that seems to modify the wrong word or word group in a sentence is a misplaced.
Recognizing Stances in Online Debates Unsupervised opinion analysis method for debate-side classification. Mine the web to learn associations that are.
Pronouns. Cases of Personal Pronouns Case/exam.UseExampleCondition Nominative (subject case) I, we, you, he, she, it, they Subject Predicate pronoun We.
LING 001 Introduction to Linguistics Spring 2010 Syntactic parsing Part-Of-Speech tagging Apr. 5 Computational linguistics.
Attempto Controlled English Norbert Fuchs et al. Dept CS, Univ Zurich, Switzerland as (mis-)interpreted by Peter Clark.
Daily Routines Gary Ovando.
CPSC 422, Lecture 27Slide 1 Intelligent Systems (AI-2) Computer Science cpsc422, Lecture 27 Nov, 16, 2015.
SYNTAX.
APA NJ APA Teacher Training 2 What is the Purpose of the APA? To measure performance of students with the most significant cognitive disabilities.
Identifying the Subject and the Verb. In order for a sentence to be complete, it must contain at least one main clause. A main clause contains an independent.
Using Wikipedia for Hierarchical Finer Categorization of Named Entities Aasish Pappu Language Technologies Institute Carnegie Mellon University PACLIC.
From Words to Senses: A Case Study of Subjectivity Recognition Author: Fangzhong Su & Katja Markert (University of Leeds, UK) Source: COLING 2008 Reporter:
English II—February 4, 2015 Bell work: Have you ever visited a new place? How did it make you feel? Was it what you expected? Homework: – Study Island.
ACTIVE & PASSIVE VOICE Verbs can appear in two different voices: Active Voice Passive Voice.
Sentence Clarity and Combining English I Writing Tips/Grammar Review.
Parts of a Sentence Object of a Verb.  Complement; does NOT identify or modify the subject  Noun, pronoun, or word group that completes the meaning.
NLP. Parsing ( (S (NP-SBJ (NP (NNP Pierre) (NNP Vinken) ) (,,) (ADJP (NP (CD 61) (NNS years) ) (JJ old) ) (,,) ) (VP (MD will) (VP (VB join) (NP (DT.
Thursday, September 16, Announcements Movie night, Friday September 25, at 6:00 pm. Bring your family and friends. There will be vocabulary review.
SUBJECT/VERB AGREEMENT Practical English III. The Basics  The basic rule states that a singular subject takes a singular verb, while a plural subject.
Clayton Greenberg Princeton University May 16, 2013.
Subjects! You have added [brackets]! You have underlined things twice! Now…..
Relative clauses. You use a relative clause to give more information about a noun in the main clause of the sentence. The relative clause always follows.
Purdue University Writing Lab Misplaced Modifiers Consider the different meanings in the following sentences: The dog under the tree bit Carrie. vs. The.
 V = verb: action in the sentence  S = subject: noun or pronoun performing the action  DO = direct object: comes after an action verb and answers the.
The 4 “C’s” Be clear, concise, considerate, and correct!
DISC Behavior Profile Module 00-2 Modified: 9/20/2018.
Probabilistic and Lexicalized Parsing
Probabilistic and Lexicalized Parsing
Using the Right Word Part Two
Parts of Speech II.
Direct Objects.
SUCCESS CRITERIA: UNDERSTAND BASIC SENTENCES FOR GRADE E.
Presentation transcript:

Clayton Greenberg Princeton University April 21, 2013

One morning, I shot an elephant in my pajamas. How he got into my pajamas I’ll never know. -- Groucho Marx

0 join board as director V 1 is chairman of N.V. N 2 named director of conglomerate N 3 caused percentage of deaths N 5 using crocidolite in filters V 6 bring attention to problem V 9 is asbestos in products N 12 led team of researchers N 13 making paper for filters N 16 including three with cancer N 18 is finding among those N 22 is one of nations N

* Algorithm for devising an ordered list of human-readable forms. * In vanilla form, does not account for probabilistic calculations. * Well suited to classifications that obey Markov assumption. * Allows for “discovery” of linguistic insights.

* Algorithm for devising an ordered list of human-readable forms. * In vanilla form, does not account for probabilistic calculations. * Well suited to classifications that obey Markov assumption. * Allows for “discovery” of linguistic insights.

There is a “general tendency” that complements merge first, then adjuncts, then specifiers. of PPs tend to be complements: 1) He is a student of linguistics with a math minor. 2) *He is a student with a math minor of linguistics.

Either by implementing a TBL system or having the insight to divide the corpus into preposition groups, one would discover that the Ratnaparkhi corpus shows of PPs attaching to nouns… 99.1% of the time. The only exceptions are phrasal verbs like accuse of and misclassifications. And, of happens to be the most frequent preposition in English.

to PPs usually attach to verbs. When then attach to nouns, they are usually nominalizations of verbs. From word-specific rules induced by TBL: If the object of the preposition can be used as an instrument for the verb, the preposition likely is VP-attach.

* Ordered hierarchy of synsets, per POS. * Synsets contain words, glosses, hypernyms and hyponyms. * Equipped to help with data sparsity. * Provides semi-unsupervised dimensions to data. * Perhaps able to capture generalizations about PP-attachment behaviors.

* I ate my food with a spork * I ate my food with an eating utensil (hypernym) * I ate my food with a fork * I ate my food with a spoon

* Used WordNet hierarchies and TBL to induce rules for PP attachment. * Accuracy: 81.8% * That’s decently good, but you get most of the way there by defining a default for each preposition (~75%).

I saw the woman with the telescope. I saw the woman with the red glasses. I saw the woman with the handbag.

* Beyond hierarchies and glosses. * Teleological links: subjects were asked what do you use a shovel for? Scissors? Oven? * Evocation database: subject were given pairs of words and asked how much one word evokes the other. See -> telescope (weak). Telescope -> see (strong).

* Cleaned up Ratnaparkhi’s corpus. * Found the original sentence for each quadruple in the Penn Treebank. * Programmatically determined three disambiguating glosses for each quadruple. * Presented MTurkers with the original and glosses. They are asked to rate meaning quality from 1 to 5.

EXAMPLES: Original sentence: One morning, I shot an elephant in my pajamas GOOD: shot in my pajamas, an elephant EXCELLENT: shot while in my pajamas, an elephant FAIR: an elephant could be in my pajamas This sentence has two possible meanings. In one meaning, the shooter is wearing pajamas. In the other, the elephant is wearing pajamas. The first meaning is more reasonable, so it gets a higher rating. The "while" makes the second phrase clearer than the first: the shooter is not shooting into the pajamas. Therefore, the second phrase gets a higher rating than the first. Original sentence: He saw the woman with the telescope. EXCELLENT: saw with the telescope, the woman NEUTRAL: saw while with the telescope, the woman EXCELLENT: the woman could be with the telescope This sentence has two possible meanings. In one meaning, "he" used the telescope to see the woman. In the other, the woman has the telescope. Both meanings seem equally reasonable. However, perhaps you have a preference for one over the other. There is NO CORRECT answer. The second phrase is less clear about whether he used the telescope to see the woman, so it gets a lower rating. Original sentence: He saw the woman with the handbag. POOR: saw with the handbag, the woman POOR: saw while with the handbag, the woman EXCELLENT: the woman could be with the handbag This sentence has one possible meaning. The woman has the handbag. "He" did not use the handbag to see the woman. This is the only CORRECT answer. If the sentence has only one meaning, please rate the phrase with the correct meaning with EXCELLENT and the other(s) with POOR. If you do not get most of these correct, we may not accept your submission for payment.

* Combine: data from experiment, Ratnaparkhi classifications, WordNet data. * The weights assigned to the experiment data will show its usefulness. * We hope that these graded determinations will be reflected in the system. * Phase 1: train to answer the binary question. * Phase 2: predict opinions on three glosses.

* Over the summer, I adapted a version of TBL to translate prepositions from Persian into English. * The system exclusively used the word forms, a programmatically determined “governor” (essentially the attachment site), and the object.

Hassan bx Maryym azdwaj Krd. Literally, John with Mary got married ‘John married Mary.’ So, we need a rule that says, “If the governor is azdwaj Krdn, delete bx instead of translating it to with.

* What we wanted: * What we got:

* Finish writing the code. * Connect to frame semantics? * Investigate further why people have the intuitions that they do when the context is underspecified. * Explore other learning algorithms.