Natural Language Processing Semantics Compositional
|
|
- Lynette Cummings
- 7 years ago
- Views:
Transcription
1 Natural Language Processing Semantics Compositional Mats Dahllöf Institutionen för lingvistik och filologi December 2013
2 Natural language processing and semantics Two directions : Analysis: NL meaning representation NL free, in principle. Classification, retrieval. Synthesis (generation): meaning representation NL NL designed. Easier. Meaning representation logic.
3 Logic and semantics Aristotelian logic important ever since. Inference. (Patterns in AL: syllogisms.) E.g.: Premise: No reptiles have fur. Premise: All snakes are reptiles. Conclusion: No snakes have fur. Modern logic develops, late 19th Century more general and systematic. Formal semantics in linguistics and philosophy based on logic (20th Century). Applications in language technology and computer science.
4 Formal and computational semantics Computational semantics is the study of how to automate the process of constructing and reasoning with meaning representations of natural language expressions. (Wikipedia). Early systems rule-based, most famous example: Montague grammar (1970). Sophisticated mechanisms for translation of English into a very rich logic. (NLTK: a version of this.) Language technology: Recent interest in data-driven and machine learning-based methods. More robust less sophistcated/fine-grained.
5 Semantics in NLP NLP semantics is typically more limited in scope than NL semantics as analysed in linguistics and philosophy. NLP applications often handle semantic aspects without having explicitly semantic components, e.g. in machine translation. Other aspects of language morphology, syntax, etc. can be seen as support systems for semantics: The purpose of language lies in the use of expressions as carriers of semantic meaning. And that is what many NLP systems have to respect, e.g. MT, retrieval, classification, etc.
6 Semantics and truth Semantics, meanings and states of affairs: What a sentence means: a structure involving (lexical) concepts and relations among them. Can be articulated as a semantic representation. E.g. I ate a turkey sandwich. in predicte logic: x(ate(i, x) Sandwich(x) ContainsTurkey(x)) A sentence and the semantic representation of a sentence is also the representation of a possible state of affairs.
7 Semantics and truth Correspondence theory of truth: If the content of a sentence corresponds to an actual state of affairs if it is true; otherwise, it is false. Ignoring philosophical complications, in many cases we can extract knowledge from texts. E.g. Warmer climate entails increased release of carbon dioxide by inland lakes. (From uu.se press release.) Related issue: Which texts should we trust? Many sentences are difficult to formalize in logic. (Modality, conditionality, vague quantification, tense, etc.)
8 Semantics vs pragmatics/discourse What does a word, a phrase, a text segment mean as an NL expression? ( Linguistic meaning semantics.) Conventional, static, systemic aspect of meaning. What does the author intend to convey by means of a word, a phrase, a text segment? ( Speaker meaning pragmatics/discourse.) Contextual, dynamic aspect of meaning. The two aspects depend on each other, of course.
9 Semantics vs pragmatics/discourse After a stellar two-year run, the bond market is stumbling and a number of investors are betting that stocks will post better returns in the coming months. (Wall Street Journal.) semantics pragmatics/discourse communicative intentions e.g. a wish to advise, caution, etc. relation to context (coherence) open reference specific reference intended (reference resolution)
10 Semantics vs pragmatics/discourse After a stellar two-year run, the bond market is stumbling and a number of investors are betting that stocks will post better returns in the coming months. (Wall Street Journal.) semantics pragmatics/discourse vagueness more well-defined concepts intended run, bond, market (NLP: WSD) ambiguity literal content specific interpretation intended metaphoric interpretation stellar, stumbling
11 Semantics-oriented NLP applications Machine translation: The translation of a text segment should mean the same as the original (to emphasize linguistic meaning) or should convey the same content (to emphasize speaker meaning). Information extraction is to extract components of the information conveyed by a text. Question answering is extraction combined with inference of an answer to a given question. Text classification, in typical cases, relates to the meanings of the texts being classified.
12 Semantics and generation Generation: semantic representation NL. Less challenging than analysis the structure of the input is under control. Needed in e.g. dialogue systems. Interlingua semantic representation in machine translation: Analysis: source language interlingua. Generation: interlingua target language. Would be economic if many languages are involved. The idea has not proved very successful so far.
13 Reference Reference is very important what statements are about. Referring expressions are very common. Reference is a discourse phenomenon. Resolving reference is a crucial step in e.g. extraction, e.g.in sentiment analysis translation, e.g. to get agreement right English it vs French il/elle vs Swedish den/det.
14 Reference, an example The portrait, comments, and updates provide constant reminders of the existence of friends. The content is not all that important, but the effect is that we perceive our Facebook friends as closer than other acquaintances who are not on Facebook. (From a uu.se press release.) Referring expressions are interpreted under coherence assumptions: The content of a Facebook page the effect of Facebook interaction with people we who are (potential) users of Facebook
15 Kinds of referring expressions Indefinite noun phrases. E.g. a book. Introduce new entities. Pronouns. E.g. he. Typically coreferent with a previous referring expression (antecedent). Names. E.g. Mats Dahllöf/Mats. Demonstrative. E.g. this room. Other definite noun phrases. E.g. the first chapter. Reference to somehow known entity, often previously mentioned.
16 Named entity recognition (NER) To identify expressions being used as names. (What characterizes a name?) Also to identify what kind of name it is: E.g. of a person, or a place, or a stretch of time, or a chemical compound, or a gene, etc. State-of-the-art NER systems for English produce near-human performance. For example, the best system entering MUC-7 scored 93.39% of F-measure while human annotators scored 97.60% and 96.95% (Wikipedia, Named-entity recognition).
17 Anaphora and deixis resolution Pronouns (they), pronominal adverbs (there, then), and definite NP s refer to entities by means of contextually given information. E.g. by referring to previously mentioned referents anaphora. E.g. by reference based on the participants, time, and place of the discourse deixis (e.g. I, you, here, yesterday). Anaphora and deixis resolution is much more challenging task than NER. The reference of name-like graph words is much more predictible. Compare Barack Obama and he.
18 Sentiment analysis an extraction task What views do people express in blogs and reviews? That s interesting for politicans and marketing people. Opinions are often expressed in a personal and informal way. E.g. Peter bought me a Baileys marzipan chocolate thing which I washed down with Gluehwein and that, in combination with the bright lights and cheery faces really made me feel warm inside! (From a blog post.) Sentiment analysis: to extract the referent of a sentiment and the polarity positive negative associated with it. E.g. Baileys marzipan chocolate positive.
19 Sentiment analysis, two dimensions Subjectivity: neutral vs polar. Polarity: positive vs negative. Demo (NLTK):
20 Sentiment analysis, two dimensions Demo (NLTK): This movie is tremendously enjoyable, and considering how exotic it is, The Desolation of Smaug is weirdly unassuming. It rattles along, never drags, and is always terrifically likable: open, genial and good-natured. (From a movie review, Subjectivity neutral: 0.1 polar: 0.9 Polarity pos: 0.8 neg: 0.2
21 Opinion definition An opinion is a quintuple (e,a,s,h,t), where e is a target entity. a is an aspect/feature of the entity e. s is the sentiment value of the opinion from the opinion holder h on aspect a of entity e at time t. s is +ve, -ve, or neu, or a more granular rating. h is an opinion holder. t is the time when the opinion is expressed. (Bing Liu)
22 Lexical concepts Words are often grammatically simple, but carry a structured conceptual content. Definitions unpack the content of concepts: friend a person whom one knows well, is loyal to, etc. turkey a kind of animal, a bird, etc. sandwich a kind of food item, contains bread, etc. eat a relation (holding in/of an event) between an organism and a food item, the food is chewed and ingested, etc.
23 Lexical concepts decomposition Componential analysis hen +chicken +female +adult rooster +chicken -female +adult chick +chicken -adult sow +pig +female +adult boar +pig -female +adult piglet +pig -adult
24 Lexical concepts relations Hypernym (superordinate concept): novel book hen chicken bird animal But all the publicity surrounding Franzen s newest novel, Freedom, edged me on to read the first chapter of the book How To Truss Your Chicken [... ] Catch the wings with the string, pulling them close to the sides of the bird. from blog posts.
25 Lexical concepts relations Part meronym book chapter bird wing; bird side; But all the publicity surrounding Franzen s newest novel, Freedom, edged me on to read the first chapter of the book How To Truss Your Chicken [... ] Catch the wings with the string, pulling them close to the sides of the bird. from blog posts.
26 Synonymy Synonymy holds between two words (word tokens) which express the same or similar concepts. Unsupervised detection of synonymy can be based on The Distributional Hypothesis: words with similar distributions have similar meanings. Random Indexing is a method here. Synonymy knowledge useful in e.g. translation, text classification, and information extraction. Also query expansion in retrieval.
27 Formalizing meaning (see J&M Ch. 17) Linguistic content has at least to a certain degree a logical structure that can be formalized by means of logical calculi meaning representations. The representation languages should be simple and unambiguous in contrast to complex and ambiguous NL. Logical calculi come with accounts of logical inference. They are useful for reasoning-based applications. Meaning formalization faces far-reaching conceptual and computational difficulties.
28 Compositionality (J&M Ch. 18) Linguistic content is compositional: Simple expressions have a given (lexical) meaning; the meaning of complex expressions is determined by the meanings of their constituents. People produce and understand new phrases and sentences all the time. (NLP must also deal with these.) Compositionality is studied in detail in compositional syntax-driven semantics. Work in this field is typically about hand-coded rule systems for small fragments of NL.
29 Compositional aspects I ate a turkey sandwich. a: some item was thus involved. Logic: existential quantification. This item was a turkey sandwich. turkey sandwich: a kind of sandwich somehow being of a turkey kind.
30 Predicates and arguments Predicates when associated (through the argument roles) with arguments give us propositions, which are true or false, or at least say something about how things are. Many tasks in information extraction are about finding propositions of a certain kind. Predicate words: verbs, adjectives, nouns, adverbs. Predicate argument analysis in NLP is typically course-grained compared to what formal logical semantics would require.
31 Compositional aspects argument structure I ate a turkey sandwich. Predicate argument structure analysis or semantic role labeling would find: I: The speaker was in relation to ate the eater or first argument. (Pronominal reference resolution is a matter of discourse-related processing.) a turkey sandwich: Was in relation to ate the eaten or second argument. Time could be seen as a third argument expressed as the verb s tense.
32 Syntactic function vs semantic role There are systematic connections between syntactic functions and semantic roles, but these relations also vary: One of the primary difficulties in labeling semantic roles is that one predicate may be used with different argument structures: for example, in the sentences He opened the door and The door opened, the verb open assigns different semantic roles to its syntactic subject. (Gildea and Jurafsky, 2002)
33 Discourse-related aspects I ate a turkey sandwich. I: pronoun resolution (deixis). ate: tense preterite this event took place in the past. Pragmatics: When? How far away in the past would be relevant? Does the context provide a setting for this eating? turkey sandwich: sandwich somehow related to turkey Comtext-sensitive interpretation: sandwich made with turkey meat as an ingredient.
34 Compositional semantics in NLP I ate a turkey sandwich. Just predicate argument structure (more basic): predicate: ate eater: I eaten: a turkey sandwich NLP: semantic role labelling. Predicate logic (something like): x(ate(i, x) Sandwich(x) ContainsTurkey(x)) (Referent of I: pronoun resolution.)
35 Predicate argument data If we want to do supervised predicate argument analysis, we need training data. Would unsupervised methods be possible when it comes to predicate argument analysis? Most well-known corpus of this kind: PropBank (Proposition Bank) English (mostly Wall Street Journal) annotated with the predicate-argument structure of the verbs. (Also other languages.) (Martha Palmer, and co-workers, 2002.) There is also a NomBank with noun annotation.
36 PropBank semantic roles neutral semantic roles, as opposed to verb-specific ones E.g. give: Arg0: giver Arg1: thing given Arg2: entity given to Rather than giver, gift, recipient. (PropBank does not give us propositions where the predicate comes from e.g. a noun or an adjective.)
37 System: Semantic role labeler, Lund joint syntactic-semantic analysis bottom-up projective parser using pseudo-projective transformations the semantic model uses global inference mechanisms on top of a pipeline of classifiers Richard Johansson and Pierre Nugues, 2008, Dependency-based Syntactic-Semantic Analysis with PropBank and NomBank, CoNLL 2008: Proceedings of the 12th Conference on Computational Natural Language Learning,
38 System: Semantic role labeler, Lund Semantic model in three steps: A SRL classifier pipeline that generates a list of candidate predicate-argument structures. A constraint system that filters the candidate list to enforce linguistic restrictions on the global configuration of arguments. A global classifier that rescores [... ] the filtered candidate list.
39 System: Semantic role labeler, Lund Linguistically motivated global constraints CORE ARGUMENT CONSISTENCY. Core argument labels must not appear more than once. DISCONTINUITY CONSISTENCY. If there is a label C-X, it must be preceded by a label X. REFERENCE CONSISTENCY. If there is a label R-X and the label is inside a relative clause, it must be preceded by a label X.
40 System: Semantic role labeler, Lund Crucial to our success was the high performance of the syntactic parser, which achieved a high accuracy. integration of syntactic and semantic analysis is beneficial for both sub-tasks. F1-scores around 0.8.
41 System: Enju Enju is a syntactic parser for English. wide-coverage probabilistic HPSG grammar efficient parsing algorithm provides phrase structures and predicate-argument structures useful for high-level NLP applications, including information extraction, automatic summarization, and question answering
42 Enju output example The company that he runs is small output: ROOT ROOT ROOT ROOT -1 ROOT ROOT is be is be VBZ VB 5 verb_arg12 ARG1 company company is be VBZ VB 5 verb_arg12 ARG2 small small small small JJ JJ 6 adj_arg1 ARG1 company company The the DT DT 0 det_arg1 ARG1 company company that that IN IN 2 relative_arg1 ARG1 company company runs run VBZ VB 4 verb_arg12 ARG1 he he runs run VBZ VB 4 verb_arg12 ARG2 company company
43 LinGO English Resource Grammar broad-coverage, linguistically precise HPSG-based grammar of English semantically grounded in Minimal Recursion Semantics (MRS), which is a form of flat semantic representation developed around the turn of the century.
44 First-order predicate logic flexible, well-understood, and computationally tractable approach to the representation of knowledge [and] meaning (J&M p. 589) expressive verifiability against a knowledge base (related to database languages) inference model-theoretic semantics
45 First-order predicate logic Boolean operators: negation and connectives Existential/universal quantification Individual constants Predicates (taking a number of arguments)
46 When to assume compositionality? Many phrases are lexicalized in a way that gives a non-compositional lexicalized sense and a less plausible compositional reading: hard disc digital memory device part of speech word category green card permanent residence permit (U.S.) This kind of ambiguity is a problem in e.g. machine translation.
47 Major points Logic-based semantics is a theoretical foundation for NLP semantics, but implemented systems are typically more coarse-grained and of a more limited scope. Meaning depends both on literal content and contextual information. This is a challenge for most NLP tasks. Most NLP applications have to be highly sensitive to semantics.
48 Major points Finding and interpreting names and other referential expressions is a central issue for NLP semantics. Disambiguation of polysemous lexical tokens is also a central issue for NLP semantics. Accessing the content of lexical tokens is also useful. Meaning representation involves predicate-argument structure, which captures a basic aspect of NL compositionality.
31 Case Studies: Java Natural Language Tools Available on the Web
31 Case Studies: Java Natural Language Tools Available on the Web Chapter Objectives Chapter Contents This chapter provides a number of sources for open source and free atural language understanding software
More informationINF5820 Natural Language Processing - NLP. H2009 Jan Tore Lønning jtl@ifi.uio.no
INF5820 Natural Language Processing - NLP H2009 Jan Tore Lønning jtl@ifi.uio.no Semantic Role Labeling INF5830 Lecture 13 Nov 4, 2009 Today Some words about semantics Thematic/semantic roles PropBank &
More informationHow the Computer Translates. Svetlana Sokolova President and CEO of PROMT, PhD.
Svetlana Sokolova President and CEO of PROMT, PhD. How the Computer Translates Machine translation is a special field of computer application where almost everyone believes that he/she is a specialist.
More informationCS4025: Pragmatics. Resolving referring Expressions Interpreting intention in dialogue Conversational Implicature
CS4025: Pragmatics Resolving referring Expressions Interpreting intention in dialogue Conversational Implicature For more info: J&M, chap 18,19 in 1 st ed; 21,24 in 2 nd Computing Science, University of
More informationCS 6740 / INFO 6300. Ad-hoc IR. Graduate-level introduction to technologies for the computational treatment of information in humanlanguage
CS 6740 / INFO 6300 Advanced d Language Technologies Graduate-level introduction to technologies for the computational treatment of information in humanlanguage form, covering natural-language processing
More informationBuilding a Question Classifier for a TREC-Style Question Answering System
Building a Question Classifier for a TREC-Style Question Answering System Richard May & Ari Steinberg Topic: Question Classification We define Question Classification (QC) here to be the task that, given
More informationModule Catalogue for the Bachelor Program in Computational Linguistics at the University of Heidelberg
Module Catalogue for the Bachelor Program in Computational Linguistics at the University of Heidelberg March 1, 2007 The catalogue is organized into sections of (1) obligatory modules ( Basismodule ) that
More informationIntro to Linguistics Semantics
Intro to Linguistics Semantics Jarmila Panevová & Jirka Hana January 5, 2011 Overview of topics What is Semantics The Meaning of Words The Meaning of Sentences Other things about semantics What to remember
More informationParaphrasing controlled English texts
Paraphrasing controlled English texts Kaarel Kaljurand Institute of Computational Linguistics, University of Zurich kaljurand@gmail.com Abstract. We discuss paraphrasing controlled English texts, by defining
More informationSearch and Data Mining: Techniques. Text Mining Anya Yarygina Boris Novikov
Search and Data Mining: Techniques Text Mining Anya Yarygina Boris Novikov Introduction Generally used to denote any system that analyzes large quantities of natural language text and detects lexical or
More informationParsing Software Requirements with an Ontology-based Semantic Role Labeler
Parsing Software Requirements with an Ontology-based Semantic Role Labeler Michael Roth University of Edinburgh mroth@inf.ed.ac.uk Ewan Klein University of Edinburgh ewan@inf.ed.ac.uk Abstract Software
More informationOpen Domain Information Extraction. Günter Neumann, DFKI, 2012
Open Domain Information Extraction Günter Neumann, DFKI, 2012 Improving TextRunner Wu and Weld (2010) Open Information Extraction using Wikipedia, ACL 2010 Fader et al. (2011) Identifying Relations for
More informationIntroduction to formal semantics -
Introduction to formal semantics - Introduction to formal semantics 1 / 25 structure Motivation - Philosophy paradox antinomy division in object und Meta language Semiotics syntax semantics Pragmatics
More informationDomain Adaptive Relation Extraction for Big Text Data Analytics. Feiyu Xu
Domain Adaptive Relation Extraction for Big Text Data Analytics Feiyu Xu Outline! Introduction to relation extraction and its applications! Motivation of domain adaptation in big text data analytics! Solutions!
More informationSemantic analysis of text and speech
Semantic analysis of text and speech SGN-9206 Signal processing graduate seminar II, Fall 2007 Anssi Klapuri Institute of Signal Processing, Tampere University of Technology, Finland Outline What is semantic
More informationArchitecture of an Ontology-Based Domain- Specific Natural Language Question Answering System
Architecture of an Ontology-Based Domain- Specific Natural Language Question Answering System Athira P. M., Sreeja M. and P. C. Reghuraj Department of Computer Science and Engineering, Government Engineering
More informationTowards Automatic Animated Storyboarding
Proceedings of the Twenty-Third AAAI Conference on Artificial Intelligence (2008) Towards Automatic Animated Storyboarding Patrick Ye and Timothy Baldwin Computer Science and Software Engineering NICTA
More informationLearning Translation Rules from Bilingual English Filipino Corpus
Proceedings of PACLIC 19, the 19 th Asia-Pacific Conference on Language, Information and Computation. Learning Translation s from Bilingual English Filipino Corpus Michelle Wendy Tan, Raymond Joseph Ang,
More informationAn Overview of a Role of Natural Language Processing in An Intelligent Information Retrieval System
An Overview of a Role of Natural Language Processing in An Intelligent Information Retrieval System Asanee Kawtrakul ABSTRACT In information-age society, advanced retrieval technique and the automatic
More informationOverview of the TACITUS Project
Overview of the TACITUS Project Jerry R. Hobbs Artificial Intelligence Center SRI International 1 Aims of the Project The specific aim of the TACITUS project is to develop interpretation processes for
More informationOutline of today s lecture
Outline of today s lecture Generative grammar Simple context free grammars Probabilistic CFGs Formalism power requirements Parsing Modelling syntactic structure of phrases and sentences. Why is it useful?
More informationLanguage Meaning and Use
Language Meaning and Use Raymond Hickey, English Linguistics Website: www.uni-due.de/ele Types of meaning There are four recognisable types of meaning: lexical meaning, grammatical meaning, sentence meaning
More informationA. Schedule: Reading, problem set #2, midterm. B. Problem set #1: Aim to have this for you by Thursday (but it could be Tuesday)
Lecture 5: Fallacies of Clarity Vagueness and Ambiguity Philosophy 130 September 23, 25 & 30, 2014 O Rourke I. Administrative A. Schedule: Reading, problem set #2, midterm B. Problem set #1: Aim to have
More informationA chart generator for the Dutch Alpino grammar
June 10, 2009 Introduction Parsing: determining the grammatical structure of a sentence. Semantics: a parser can build a representation of meaning (semantics) as a side-effect of parsing a sentence. Generation:
More informationSpecial Topics in Computer Science
Special Topics in Computer Science NLP in a Nutshell CS492B Spring Semester 2009 Jong C. Park Computer Science Department Korea Advanced Institute of Science and Technology INTRODUCTION Jong C. Park, CS
More informationCINTIL-PropBank. CINTIL-PropBank Sub-corpus id Sentences Tokens Domain Sentences for regression atsts 779 5,654 Test
CINTIL-PropBank I. Basic Information 1.1. Corpus information The CINTIL-PropBank (Branco et al., 2012) is a set of sentences annotated with their constituency structure and semantic role tags, composed
More informationOverview of MT techniques. Malek Boualem (FT)
Overview of MT techniques Malek Boualem (FT) This section presents an standard overview of general aspects related to machine translation with a description of different techniques: bilingual, transfer,
More informationNatural Language Database Interface for the Community Based Monitoring System *
Natural Language Database Interface for the Community Based Monitoring System * Krissanne Kaye Garcia, Ma. Angelica Lumain, Jose Antonio Wong, Jhovee Gerard Yap, Charibeth Cheng De La Salle University
More informationL130: Chapter 5d. Dr. Shannon Bischoff. Dr. Shannon Bischoff () L130: Chapter 5d 1 / 25
L130: Chapter 5d Dr. Shannon Bischoff Dr. Shannon Bischoff () L130: Chapter 5d 1 / 25 Outline 1 Syntax 2 Clauses 3 Constituents Dr. Shannon Bischoff () L130: Chapter 5d 2 / 25 Outline Last time... Verbs...
More informationRethinking the relationship between transitive and intransitive verbs
Rethinking the relationship between transitive and intransitive verbs Students with whom I have studied grammar will remember my frustration at the idea that linking verbs can be intransitive. Nonsense!
More informationTowards Robust High Performance Word Sense Disambiguation of English Verbs Using Rich Linguistic Features
Towards Robust High Performance Word Sense Disambiguation of English Verbs Using Rich Linguistic Features Jinying Chen and Martha Palmer Department of Computer and Information Science, University of Pennsylvania,
More informationAutomatic Speech Recognition and Hybrid Machine Translation for High-Quality Closed-Captioning and Subtitling for Video Broadcast
Automatic Speech Recognition and Hybrid Machine Translation for High-Quality Closed-Captioning and Subtitling for Video Broadcast Hassan Sawaf Science Applications International Corporation (SAIC) 7990
More informationNatural Language to Relational Query by Using Parsing Compiler
Available Online at www.ijcsmc.com International Journal of Computer Science and Mobile Computing A Monthly Journal of Computer Science and Information Technology IJCSMC, Vol. 4, Issue. 3, March 2015,
More informationReasoning Component Architecture
Architecture of a Spam Filter Application By Avi Pfeffer A spam filter consists of two components. In this article, based on my book Practical Probabilistic Programming, first describe the architecture
More informationStatistical Machine Translation
Statistical Machine Translation Some of the content of this lecture is taken from previous lectures and presentations given by Philipp Koehn and Andy Way. Dr. Jennifer Foster National Centre for Language
More informationIntroduction. Philipp Koehn. 28 January 2016
Introduction Philipp Koehn 28 January 2016 Administrativa 1 Class web site: http://www.mt-class.org/jhu/ Tuesdays and Thursdays, 1:30-2:45, Hodson 313 Instructor: Philipp Koehn (with help from Matt Post)
More informationIdentifying Focus, Techniques and Domain of Scientific Papers
Identifying Focus, Techniques and Domain of Scientific Papers Sonal Gupta Department of Computer Science Stanford University Stanford, CA 94305 sonal@cs.stanford.edu Christopher D. Manning Department of
More informationAccording to the Argentine writer Jorge Luis Borges, in the Celestial Emporium of Benevolent Knowledge, animals are divided
Categories Categories According to the Argentine writer Jorge Luis Borges, in the Celestial Emporium of Benevolent Knowledge, animals are divided into 1 2 Categories those that belong to the Emperor embalmed
More informationSpeech and Language Processing
Speech and Language Processing An Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition Second Edition Daniel Jurafsky Stanford University James H. Martin University
More informationEAP 1161 1660 Grammar Competencies Levels 1 6
EAP 1161 1660 Grammar Competencies Levels 1 6 Grammar Committee Representatives: Marcia Captan, Maria Fallon, Ira Fernandez, Myra Redman, Geraldine Walker Developmental Editor: Cynthia M. Schuemann Approved:
More informationWhy language is hard. And what Linguistics has to say about it. Natalia Silveira Participation code: eagles
Why language is hard And what Linguistics has to say about it Natalia Silveira Participation code: eagles Christopher Natalia Silveira Manning Language processing is so easy for humans that it is like
More informationAppendix B Data Quality Dimensions
Appendix B Data Quality Dimensions Purpose Dimensions of data quality are fundamental to understanding how to improve data. This appendix summarizes, in chronological order of publication, three foundational
More informationLikewise, we have contradictions: formulas that can only be false, e.g. (p p).
CHAPTER 4. STATEMENT LOGIC 59 The rightmost column of this truth table contains instances of T and instances of F. Notice that there are no degrees of contingency. If both values are possible, the formula
More informationMachine Learning for natural language processing
Machine Learning for natural language processing Introduction Laura Kallmeyer Heinrich-Heine-Universität Düsseldorf Summer 2016 1 / 13 Introduction Goal of machine learning: Automatically learn how to
More informationWhat s in a Lexicon. The Lexicon. Lexicon vs. Dictionary. What kind of Information should a Lexicon contain?
What s in a Lexicon What kind of Information should a Lexicon contain? The Lexicon Miriam Butt November 2002 Semantic: information about lexical meaning and relations (thematic roles, selectional restrictions,
More informationSymbiosis of Evolutionary Techniques and Statistical Natural Language Processing
1 Symbiosis of Evolutionary Techniques and Statistical Natural Language Processing Lourdes Araujo Dpto. Sistemas Informáticos y Programación, Univ. Complutense, Madrid 28040, SPAIN (email: lurdes@sip.ucm.es)
More informationChunk Parsing. Steven Bird Ewan Klein Edward Loper. University of Melbourne, AUSTRALIA. University of Edinburgh, UK. University of Pennsylvania, USA
Chunk Parsing Steven Bird Ewan Klein Edward Loper University of Melbourne, AUSTRALIA University of Edinburgh, UK University of Pennsylvania, USA March 1, 2012 chunk parsing: efficient and robust approach
More informationInformation extraction from online XML-encoded documents
Information extraction from online XML-encoded documents From: AAAI Technical Report WS-98-14. Compilation copyright 1998, AAAI (www.aaai.org). All rights reserved. Patricia Lutsky ArborText, Inc. 1000
More informationParsing Technology and its role in Legacy Modernization. A Metaware White Paper
Parsing Technology and its role in Legacy Modernization A Metaware White Paper 1 INTRODUCTION In the two last decades there has been an explosion of interest in software tools that can automate key tasks
More informationSentence Semantics. General Linguistics Jennifer Spenader, February 2006 (Most slides: Petra Hendriks)
Sentence Semantics General Linguistics Jennifer Spenader, February 2006 (Most slides: Petra Hendriks) Data to be analyzed (1) Maria slaapt. (2) Jan slaapt. (3) Maria slaapt en Jan slaapt. (4) Iedereen
More informationGet the most value from your surveys with text analysis
PASW Text Analytics for Surveys 3.0 Specifications Get the most value from your surveys with text analysis The words people use to answer a question tell you a lot about what they think and feel. That
More informationLogic in general. Inference rules and theorem proving
Logical Agents Knowledge-based agents Logic in general Propositional logic Inference rules and theorem proving First order logic Knowledge-based agents Inference engine Knowledge base Domain-independent
More informationRepresentation and Processing Revisited: Meaning
Chapter 7 Representation and Processing Revisited: Meaning 7.1 Introduction The discussion in previous chapters reinforces the point made in Chapter 3 about the value of syntactic, and shallow semantic
More informationApplying Co-Training Methods to Statistical Parsing. Anoop Sarkar http://www.cis.upenn.edu/ anoop/ anoop@linc.cis.upenn.edu
Applying Co-Training Methods to Statistical Parsing Anoop Sarkar http://www.cis.upenn.edu/ anoop/ anoop@linc.cis.upenn.edu 1 Statistical Parsing: the company s clinical trials of both its animal and human-based
More informationSentence Structure/Sentence Types HANDOUT
Sentence Structure/Sentence Types HANDOUT This handout is designed to give you a very brief (and, of necessity, incomplete) overview of the different types of sentence structure and how the elements of
More informationTOOL OF THE INTELLIGENCE ECONOMIC: RECOGNITION FUNCTION OF REVIEWS CRITICS. Extraction and linguistic analysis of sentiments
TOOL OF THE INTELLIGENCE ECONOMIC: RECOGNITION FUNCTION OF REVIEWS CRITICS. Extraction and linguistic analysis of sentiments Grzegorz Dziczkowski, Katarzyna Wegrzyn-Wolska Ecole Superieur d Ingenieurs
More informationTowards a RB-SMT Hybrid System for Translating Patent Claims Results and Perspectives
Towards a RB-SMT Hybrid System for Translating Patent Claims Results and Perspectives Ramona Enache and Adam Slaski Department of Computer Science and Engineering Chalmers University of Technology and
More informationText Analytics for Competitive Analysis and Market Intelligence Aiaioo Labs - 2011
Text Analytics for Competitive Analysis and Market Intelligence Aiaioo Labs - 2011 Bangalore, India Title Text Analytics Introduction Entity Person Comparative Analysis Entity or Event Text Analytics Text
More informationIntroduction. BM1 Advanced Natural Language Processing. Alexander Koller. 17 October 2014
Introduction! BM1 Advanced Natural Language Processing Alexander Koller! 17 October 2014 Outline What is computational linguistics? Topics of this course Organizational issues Siri Text prediction Facebook
More informationTagging with Hidden Markov Models
Tagging with Hidden Markov Models Michael Collins 1 Tagging Problems In many NLP problems, we would like to model pairs of sequences. Part-of-speech (POS) tagging is perhaps the earliest, and most famous,
More informationThe Role of Sentence Structure in Recognizing Textual Entailment
Blake,C. (In Press) The Role of Sentence Structure in Recognizing Textual Entailment. ACL-PASCAL Workshop on Textual Entailment and Paraphrasing, Prague, Czech Republic. The Role of Sentence Structure
More informationMeaning and communication, swearing and silence
or pragmatics? Two case studies Meaning and communication, swearing and silence Gary Thoms 2/12/14 or pragmatics? Two case studies Outline communication and linguistic meaning semantics and pragmatics
More informationCENG 734 Advanced Topics in Bioinformatics
CENG 734 Advanced Topics in Bioinformatics Week 9 Text Mining for Bioinformatics: BioCreative II.5 Fall 2010-2011 Quiz #7 1. Draw the decompressed graph for the following graph summary 2. Describe the
More informationA + dvancer College Readiness Online Alignment to Florida PERT
A + dvancer College Readiness Online Alignment to Florida PERT Area Objective ID Topic Subject Activity Mathematics Math MPRC1 Equations: Solve linear in one variable College Readiness-Arithmetic Solving
More informationCourse Description (MA Degree)
Course Description (MA Degree) Eng. 508 Semantics (3 Credit hrs.) This course is an introduction to the issues of meaning and logical interpretation in natural language. The first part of the course concentrates
More informationD2.4: Two trained semantic decoders for the Appointment Scheduling task
D2.4: Two trained semantic decoders for the Appointment Scheduling task James Henderson, François Mairesse, Lonneke van der Plas, Paola Merlo Distribution: Public CLASSiC Computational Learning in Adaptive
More informationA Shortest-path Method for Arc-factored Semantic Role Labeling
A Shortest-path Method for Arc-factored Semantic Role Labeling Xavier Lluís TALP Research Center Universitat Politècnica de Catalunya xlluis@cs.upc.edu Xavier Carreras Xerox Research Centre Europe xavier.carreras@xrce.xerox.com
More informationA Survey on Product Aspect Ranking
A Survey on Product Aspect Ranking Charushila Patil 1, Prof. P. M. Chawan 2, Priyamvada Chauhan 3, Sonali Wankhede 4 M. Tech Student, Department of Computer Engineering and IT, VJTI College, Mumbai, Maharashtra,
More informationOnline Latent Structure Training for Language Acquisition
IJCAI 11 Online Latent Structure Training for Language Acquisition Michael Connor University of Illinois connor2@illinois.edu Cynthia Fisher University of Illinois cfisher@cyrus.psych.uiuc.edu Dan Roth
More informationONTOLOGIES A short tutorial with references to YAGO Cosmina CROITORU
ONTOLOGIES p. 1/40 ONTOLOGIES A short tutorial with references to YAGO Cosmina CROITORU Unlocking the Secrets of the Past: Text Mining for Historical Documents Blockseminar, 21.2.-11.3.2011 ONTOLOGIES
More informationThe Prolog Interface to the Unstructured Information Management Architecture
The Prolog Interface to the Unstructured Information Management Architecture Paul Fodor 1, Adam Lally 2, David Ferrucci 2 1 Stony Brook University, Stony Brook, NY 11794, USA, pfodor@cs.sunysb.edu 2 IBM
More informationPresented to The Federal Big Data Working Group Meetup On 07 June 2014 By Chuck Rehberg, CTO Semantic Insights a Division of Trigent Software
Semantic Research using Natural Language Processing at Scale; A continued look behind the scenes of Semantic Insights Research Assistant and Research Librarian Presented to The Federal Big Data Working
More informationISA OR NOT ISA: THE INTERLINGUAL DILEMMA FOR MACHINE TRANSLATION
ISA OR NOT ISA: THE INTERLINGUAL DILEMMA FOR MACHINE TRANSLATION FLORENCE REEDER The MITRE Corporation 1 / George Mason University freeder@mitre.org ABSTRACT Developing a system that accurately produces
More informationAutomated Extraction of Security Policies from Natural-Language Software Documents
Automated Extraction of Security Policies from Natural-Language Software Documents Xusheng Xiao 1 Amit Paradkar 2 Suresh Thummalapenta 3 Tao Xie 1 1 Dept. of Computer Science, North Carolina State University,
More informationTechWatch. Technology and Market Observation powered by SMILA
TechWatch Technology and Market Observation powered by SMILA PD Dr. Günter Neumann DFKI, Deutsches Forschungszentrum für Künstliche Intelligenz GmbH, Juni 2011 Goal - Observation of Innovations and Trends»
More informationCHARTES D'ANGLAIS SOMMAIRE. CHARTE NIVEAU A1 Pages 2-4. CHARTE NIVEAU A2 Pages 5-7. CHARTE NIVEAU B1 Pages 8-10. CHARTE NIVEAU B2 Pages 11-14
CHARTES D'ANGLAIS SOMMAIRE CHARTE NIVEAU A1 Pages 2-4 CHARTE NIVEAU A2 Pages 5-7 CHARTE NIVEAU B1 Pages 8-10 CHARTE NIVEAU B2 Pages 11-14 CHARTE NIVEAU C1 Pages 15-17 MAJ, le 11 juin 2014 A1 Skills-based
More informationPredicate logic Proofs Artificial intelligence. Predicate logic. SET07106 Mathematics for Software Engineering
Predicate logic SET07106 Mathematics for Software Engineering School of Computing Edinburgh Napier University Module Leader: Uta Priss 2010 Copyright Edinburgh Napier University Predicate logic Slide 1/24
More informationThe compositional semantics of same
The compositional semantics of same Mike Solomon Amherst College Abstract Barker (2007) proposes the first strictly compositional semantic analysis of internal same. I show that Barker s analysis fails
More informationFrom Logic to Montague Grammar: Some Formal and Conceptual Foundations of Semantic Theory
From Logic to Montague Grammar: Some Formal and Conceptual Foundations of Semantic Theory Syllabus Linguistics 720 Tuesday, Thursday 2:30 3:45 Room: Dickinson 110 Course Instructor: Seth Cable Course Mentor:
More informationComputational Linguistics and Learning from Big Data. Gabriel Doyle UCSD Linguistics
Computational Linguistics and Learning from Big Data Gabriel Doyle UCSD Linguistics From not enough data to too much Finding people: 90s, 700 datapoints, 7 years People finding you: 00s, 30000 datapoints,
More informationSyntax: Phrases. 1. The phrase
Syntax: Phrases Sentences can be divided into phrases. A phrase is a group of words forming a unit and united around a head, the most important part of the phrase. The head can be a noun NP, a verb VP,
More informationCS510 Software Engineering
CS510 Software Engineering Propositional Logic Asst. Prof. Mathias Payer Department of Computer Science Purdue University TA: Scott A. Carr Slides inspired by Xiangyu Zhang http://nebelwelt.net/teaching/15-cs510-se
More informationHow To Make Sense Of Data With Altilia
HOW TO MAKE SENSE OF BIG DATA TO BETTER DRIVE BUSINESS PROCESSES, IMPROVE DECISION-MAKING, AND SUCCESSFULLY COMPETE IN TODAY S MARKETS. ALTILIA turns Big Data into Smart Data and enables businesses to
More informationSentiment Analysis. D. Skrepetos 1. University of Waterloo. NLP Presenation, 06/17/2015
Sentiment Analysis D. Skrepetos 1 1 Department of Computer Science University of Waterloo NLP Presenation, 06/17/2015 D. Skrepetos (University of Waterloo) Sentiment Analysis NLP Presenation, 06/17/2015
More informationText Mining - Scope and Applications
Journal of Computer Science and Applications. ISSN 2231-1270 Volume 5, Number 2 (2013), pp. 51-55 International Research Publication House http://www.irphouse.com Text Mining - Scope and Applications Miss
More informationMATRIX OF STANDARDS AND COMPETENCIES FOR ENGLISH IN GRADES 7 10
PROCESSES CONVENTIONS MATRIX OF STANDARDS AND COMPETENCIES FOR ENGLISH IN GRADES 7 10 Determine how stress, Listen for important Determine intonation, phrasing, points signaled by appropriateness of pacing,
More informationAcademic Standards for Reading, Writing, Speaking, and Listening
Academic Standards for Reading, Writing, Speaking, and Listening Pre-K - 3 REVISED May 18, 2010 Pennsylvania Department of Education These standards are offered as a voluntary resource for Pennsylvania
More informationClustering Connectionist and Statistical Language Processing
Clustering Connectionist and Statistical Language Processing Frank Keller keller@coli.uni-sb.de Computerlinguistik Universität des Saarlandes Clustering p.1/21 Overview clustering vs. classification supervised
More informationInteractive Dynamic Information Extraction
Interactive Dynamic Information Extraction Kathrin Eichler, Holmer Hemsen, Markus Löckelt, Günter Neumann, and Norbert Reithinger Deutsches Forschungszentrum für Künstliche Intelligenz - DFKI, 66123 Saarbrücken
More informationEAST PENNSBORO AREA COURSE: LFS 416 SCHOOL DISTRICT
EAST PENNSBORO AREA COURSE: LFS 416 SCHOOL DISTRICT Unit: Grammar Days: Subject(s): French 4 Grade(s):9-12 Key Learning(s): Students will passively recognize target grammatical structures alone and in
More informationSentiment Analysis: a case study. Giuseppe Castellucci castellucci@ing.uniroma2.it
Sentiment Analysis: a case study Giuseppe Castellucci castellucci@ing.uniroma2.it Web Mining & Retrieval a.a. 2013/2014 Outline Sentiment Analysis overview Brand Reputation Sentiment Analysis in Twitter
More informationPOS Tagsets and POS Tagging. Definition. Tokenization. Tagset Design. Automatic POS Tagging Bigram tagging. Maximum Likelihood Estimation 1 / 23
POS Def. Part of Speech POS POS L645 POS = Assigning word class information to words Dept. of Linguistics, Indiana University Fall 2009 ex: the man bought a book determiner noun verb determiner noun 1
More informationTesting Data-Driven Learning Algorithms for PoS Tagging of Icelandic
Testing Data-Driven Learning Algorithms for PoS Tagging of Icelandic by Sigrún Helgadóttir Abstract This paper gives the results of an experiment concerned with training three different taggers on tagged
More informationCHAPTER 7 GENERAL PROOF SYSTEMS
CHAPTER 7 GENERAL PROOF SYSTEMS 1 Introduction Proof systems are built to prove statements. They can be thought as an inference machine with special statements, called provable statements, or sometimes
More informationLINGSTAT: AN INTERACTIVE, MACHINE-AIDED TRANSLATION SYSTEM*
LINGSTAT: AN INTERACTIVE, MACHINE-AIDED TRANSLATION SYSTEM* Jonathan Yamron, James Baker, Paul Bamberg, Haakon Chevalier, Taiko Dietzel, John Elder, Frank Kampmann, Mark Mandel, Linda Manganaro, Todd Margolis,
More informationAn NLP Curator (or: How I Learned to Stop Worrying and Love NLP Pipelines)
An NLP Curator (or: How I Learned to Stop Worrying and Love NLP Pipelines) James Clarke, Vivek Srikumar, Mark Sammons, Dan Roth Department of Computer Science, University of Illinois, Urbana-Champaign.
More informationEnglish Grammar Checker
International l Journal of Computer Sciences and Engineering Open Access Review Paper Volume-4, Issue-3 E-ISSN: 2347-2693 English Grammar Checker Pratik Ghosalkar 1*, Sarvesh Malagi 2, Vatsal Nagda 3,
More informationChapter 8. Final Results on Dutch Senseval-2 Test Data
Chapter 8 Final Results on Dutch Senseval-2 Test Data The general idea of testing is to assess how well a given model works and that can only be done properly on data that has not been seen before. Supervised
More informationBACHELOR OF BUSINESS ADMINISTRATION Program: Goals and Objectives
BACHELOR OF BUSINESS ADMINISTRATION Program: Goals and Objectives 1. A Seidman BBA graduate will be an effective communicator. He/she will be able to: 1.1 Engage in effective interpersonal dialogue. 1.2
More informationUsing Feedback Tags and Sentiment Analysis to Generate Sharable Learning Resources
Using Feedback Tags and Sentiment Analysis to Generate Sharable Learning Resources Investigating Automated Sentiment Analysis of Feedback Tags in a Programming Course Stephen Cummins, Liz Burd, Andrew
More information