Popular in Course
Popular in ComputerScienence
This 22 page Class Notes was uploaded by Jonas Bartell on Monday October 26, 2015. The Class Notes belongs to CS2731 at University of Pittsburgh taught by Staff in Fall. Since its upload, it has received 93 views. For similar materials see /class/229395/cs2731-university-of-pittsburgh in ComputerScienence at University of Pittsburgh.
Reviews for INTRONATURALLANGPROCSSNG
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 10/26/15
INTRODUCTION TO NATURAL LANGUAGE PROCESSING INTRODUCTION CS 271 ISSP 223173 2003 Introduction Course Overview Administration CS 271 ISSP 223173 2003 Today s Outline introduction 2 Natural Language Processing o The field of Natural Language Processing NLP or Computational Linguistics CL or Human Lan guage Technology HLT is primarin concerned with the creation of computer programs that perform use ful and interesting tasks with human languages eg understanding generation learning o It is secondarin concerned with using computational metaphors to help us come to a better understanding of human language 0 The foundations of the field are in computer science eg Al theory linguistics mathematics and statis tics electrical engineering and psychology 0 Studying NLP involves studying natural languages for mal representations and algorithms for their manipula tion CS 271 ISSP 2230 Fall 2003 Introduction 3 Major Topics of this Course Knowledge of Language 0 words meaningful components of words 0 syntax structure of a sentence 0 semantics explicit meaning of a sentence 0 pragmaticsdiscourse implicitcontextual meaning Methodologies and Tools 0 knowledgebased and statistical 0 state machines rule systems grammars logic search probability automata machine learning and more Applications 0 from hyphenators to intelligent agents Current Research CS 271 ISSP 2230 Fall 2003 Introduction 1 Knowledge of Language 0 Example dialogue from 2001 A Space Odyssey Dave Open the pod bay doors HAL HAL I m sorry Dave I m afraid I can t do that c To participate in such a conversation HAL needs knowl edge about many levels of language words producing contractions plurals syntax questions versus statements word order and grouping semantics meaning of words in isolation and com positionally pragmatics politeness and indirectness discourse between utterance references CS 271 ISSP 2230 Fall 2003 Introduction Applications What makes a computer application a language processing application 0 language processing applications require the use of knowl edge of language CS 271 ISSP 2230 Fall 2003 Introduction 6 Little Applications Little applications typically make use of only a small amount of a single kind of knowledge of language Many of these types of applications are currently in use but they are often nearly invisible 0 line breakers o hyphenators 0 spelling correctors 0 OCR software 0 grammar and style checkers CS 271 ISSP 2230 Fall 2003 Introduction Big Applications Big applications often make use of large amounts and var ied kinds of knowledge of language 0 information retrieval and extraction 0 question answering o dialogue management 0 text summarization 0 machine translation o HAL CS 271 ISSP 2230 Fall 2003 Introduction 8 Some Current Application Scenarios Tutor students in areas such as physics Access the web over the telephone Speak to your appliances Grade essays Generate running commentary for robotic soccer Generate weather reports in multiple languages CS 271 ISSP 2230 Fall 2003 Introduction 9 II Demos Dialogue Systems 0 Why2 AtlasITSPOKE Pitt email to be a subject 0 TOOT NJFUN Litman et al at ATampT 0 JUPITER 1888573TALK Question Answering 0 AnswerBus wwwanswerbuscomnews 0 Ask Jeeves wwwaskcom 0 Opinion Highlighter Pitt Summarization o Newsblaster wwwcscolumbiaedunlpnewsblaster o NewslnEssence wwwnewsinessencecomniecgi Machine Translation 0 Babelfish babelfishaltavistacom Speech Synthesis 0 Festival Text To Speech festvoxorgvoicedemoshtml CS 271 ISSP 2230 Fall 2003 Introduction It II Administration Professor Teaching Assitant Students Textbook Web page Requirements Who should be here CS 271 ISSP 223173 2003 Introduction I I II Dr Diane Litman Affiliations 0 Associate Professor Computer Science Department 0 Research Scientist LRDC 0 Faculty Intelligent Systems Program 0 Member CIRCLE Contact Information 0 5105 Sennott Square 412 6248838 741 LRDC 412 6241261 0 litman cspittedu 0 httpwwwcspittedulitman Office Hours 0 M 1215215 5105 Sennott Square 0 Tu 1000 1200 741 LRDC CS 271 ISSP 2230 Fall 2003 Introduction l2 Litman cont Education 0 1986 PhD Computer Science University of Rochester o Thesis Plan Recognition and Discourse Analysis An Integrated Approach for Understanding Dialogues Employment 0 2001present University of Pittsburgh 0 19852001 Technical Staff Al Principles Research De partment ATampT Labs Research formerly ATampT Bell Laboratories 0 19901992 Assistant Professor Computer Science Columbia University 20002003 Chair North American Chapter of the Associ ation for Computational Linguistics NAACL CS 271 ISSP 2230 Fall 2003 Introduction l3 Litman cont Resea rch 0 Natural Language Processing Spoken Dialogue for Intelligent Tutoring Systems with Prof Vanlehn and LRDC Typicality and Natural Language Learning with M Rotaru Multiple Perspective Question Answering with Prof Wiebe T Wilson Predicting Medical Reasoning Codings in Pathology Protocols with Prof Crowley L lVla other discourse dialogue pragmatics spoken lan guage learning 0 Natural Language Processing Laboratory httpwwwcspittedulitmannlplabhtml o Other Artificial Intelligence knowledge representation plan recognition user mod eling CS 271 ISSP 2230 Fall 2003 Introduction M Teaching Assistant Mihai Rotaru o Doctoral Student Computer Science Department Contact Information 0 5420 Sennott Square 412 6248462 0 mrotaru cspittedu o httpwwwcspittedumrotaru Of ce Hours 0 MW 900 1100 0 Th 11001200 CS 271 ISSP 22301731 2003 Introduction l5 II Students Please introduce yourself CS 271 ISSP 223173 2003 Imxmlmmion II The Text Speech and Language Processing An Introduction to N atu39ral Language Processing Computational Linguis tics and Speech Recognition by Daniel Jurafsky and James H Martin Prentice Hall 2000 0 httpwwwcscoloradoedumartinslphtml 0 should be available from the bookstore 0 also available from online providers Course emphasis will be from a computer science per spective Text will be supplemented with a selection of current re search papers CS 271 ISSP 2230 Fall 2003 Immduction l7 Course Web Page U RL 0 wwwcspitteduitmancoursescs2731cs2731html SyHabus 0 topics 0 readings o assignments 0 lecture notes 0 announcements 0 NOTE involves viewingprinting postscript pdf ppt etc CS 271 ISSP 223173 2003 Introduction l8 Requirements Readings before class Homeworks 0 problem sets 0 writing programs 0 using programs Exams 0 midterm 0 final P roject Leading and participating in research paper discussions CS 271 ISSP 223173 2003 Introduction If Prerequisites An interest in language and 0 ability to write and use programs 0 background in computer science linguistics or speech 0 some basic exposure to logic probability algorithm anal ysis andor programming language theory would be helpful CS 271 ISSP 2230 Fall 2003 Introduction 20 H For Next Time Read Chapter 1 0 available online Get a CS account 0 httpwwwcspitteduuserdbacctapppagehtml Send me email for a class mailing list Assignment 0 try at least one listed demo from each category and be prepared to report back on your experiences Look at Reading List CS 271 ISSP 2230 Fall 2003 Introduction 2 Survey Name Email Department Year Undergraduate Major Relevant Courses Relevant Research Experience Methodologies state machines rule systems grammars logic search probability automata machine learning etc Programming Languages Operating Systems Goals CS 271 ISSP 2230 Fall 2003 Introduction
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'