|
INTRODUCTION
TO NATURAL LANGUAGE PROCESSING (CS 2731 / ISSP 2230), Fall 2013
|
|
| Instructor |
Diane Litman |
| When & Where | Tuesday and Thursdays 2:30-3:45, SENSQ 5313 |
| Office Hours | After class or by appointment |
| Description | This course provides an introduction to the field of Natural
Language Processing (NLP) - the creation of computer programs that
can understand, generate, and learn natural language. Natural
language understanding will be used as a vehicle to introduce
three major subfields of NLP: syntax, semantics, and
pragmatics. The course will introduce both knowledge-based and
statistical methods for NLP, and will illustrate the use of such
methods in a variety of application areas.
Prerequisites: CS 1501 OR consent of the instructor Text: Speech and Language Processing by Jurafsky and Martin, Second Edition (errata). |
| Required Work | Homeworks (35%): written and programming Exams (35%): midterm and final Group Course Project (20%): presentation and written report Supplemental Research Papers (10%): leading discussion and class participation Late Penalty: For assignments that may be accepted late, the penalty is 10% per day up to 5 days including Saturday, Sunday, and holidays. Assignments are due by 11:59pm. |
|
Date/Topic
|
J&M Readings
|
Assignments and Other Readings |
| August 27 Introduction |
Ch 1 | NB (CHI 2012) By 11:59 PM 9/2, use the assignment tool in CourseWeb/Blackboard to enter JUST THE NAME of something that you will demo or discuss in class on 9/5. Your demo should showcase one NLP application that you have used, highlighting its strengths and weaknesses. Your presentation should be no longer than 1-2 minutes, and done without slides! By 11:59 PM 9/2, use the assignment tool to provide the links for two SHORT conference papers (representing two different chapters selected from J&M chapters 4, 5, 12, 13, 14, 19, 20, 21, 22, 23, or 24) that you would be willing to present later in the semester. Make sure that you also explicitly state the chapter # in Jurafsky & Martin that each paper is associated with. Pick a paper by using the ACL Anthology to search ACL, EACL, NAACL, EMNLP, COLING conference proceedings, or by searching the textbook bibliography. Short papers are usually 4 pages, and sometimes are called posters and/or are in the companion volumes of the proceedings. I will use this information to make a presentation schedule for the rest of the semester. |
| August 29 Regular Expressions & Automata Guest Lecturer: |
Ch 2 | Ken Church's tutorial Unix for Poets, pages 1-19 |
| September 3, 5 Words & Transducers Guest Lecturer (9/3): Everyone (9/5): |
Ch 3 | |
| September 10, 12 Everyone: Demos (continued) |
Ch 4 (through 4.6) | Here are the final list of conference paper readings and the procedures for leading/supporting the associated class discussions. |
| September 17, 19, 24 Part-of-Speech Tagging |
Ch 5 Optional: relevant sections of Ch 6 |
9/17: HW1 Assigned
9/19: Ch 4 Reading (Chang and Wang) Grammar Rock for Pronouns |
| September 24, 26 Formal Grammars of English |
Ch 12 | 9/24: Ch 5 Reading (Jabbari and Xu)
9/26: Ch 5 Reading (Murrugarra Llerena and Myers) |
| October 1, 3, 8 Syntactic Parsing |
Ch 13 | 10/3: Ch 12 Reading (Parenti and Xue) 10/4: HW2 Assigned |
| October 8, 10 Statistical Parsing |
Ch 14 | 10/10: Ch 13 Reading (Li and Li) |
| October 15 Fall Break |
  | No class |
| October 17, 22 Lexical Semantics |
Ch 19 | 10/17: Ch 14 Reading (Xue and Yao)
10/17: Project Assigned |
| October 22, 24 Computational Lexical Semantics |
Ch 20 | 10/24 Midterm Review
10/25: Monitored Withdrawal Deadline |
| October 29 | Midterm Exam | Through Chapter 14 NO MAKEUPS |
| October 31, Nov 5 Computational Lexical Semantics |
Ch 20 | 10/31: Ch 19 Reading (Falakmasir and Yan) |
| Nov 5, 7, 12 Computational Discourse |
Ch 21 | Google's
Ngram Viewer Goes Wild
11/7: Ch 20 Reading (Hosseini and Thomas) |
| Nov 14, 19 Information Extraction |
Ch 22.1 and 22.2 ONLY | 11/14: Ch 21 Reading (Malakouti and Rahami) 11/14: Project Preliminary Evaluation due 11/14: HW3 assigned Interesting application of prior chapters: What Would I Say? (or, how to turn yourself into a facebook bot) |
| Nov 19, 21 Question Answering |
Ch 23.2 ONLY | 11/19: Ch 22 Reading (Chen and Lu)
11/21: Ch 23 Reading (Knittel and Luo) Preliminary Evaluation Results IBM to Announce More Powerful Watson via the Internet Watson paper, Watson and DeepQA homepages, documentary |
| Nov 21, 26, Dec 3, 5 Dialogue and Conversational Agents Part1, Part2, Part3 |
Ch 24 | 11/26: Ch 23 Reading (Chen and Shi)
Due (11/26): HW3 12/3: Ch 24 Reading (Liu and Savelka) Due (12/5): Project Final Evaluation |
| December 10 | (class will start early at 1:30) | Project Presentations/Reports |
| December 12 |
Final Exam (class will start early at 2pm) |
Chapters 19-24 NO MAKEUPS Google announces release of many new annotated datasets If you did well on the final project, you might think about trying to write a paper for the ACL student workshop or BEA workshop. Both are in Baltimore and many departments provide some travel support, which might be enough to cover the expenses for something close by. Watson on Broadway |
Acknowledgements: Some of the materials used in this course borrow from the NLP courses of Steven Bird, Julia Hirschberg, Dan Jurafsky, Chris Manning, James Martin, Johanna Moore, Dragomir Radev, Philip Resnick, Ellen Riloff.