Introduction to Programing for Linguists
Prof. Ray C. Dougherty
Tentative syllabus. The final syllabus will be available by September 1996.
Reading List: Chomsky, Noam. 1966. Language and Mind. New York: Harcourt Brace Dougherty, Ray C. 1994. Natural Language Computing: An English Generative Grammar in Prolog. Hillsdale, N.J.: Lawrence Erlbaum Inc. Gazdar, Gerald and Mellish, Chris. 1989. Natural Language Processing in Prolog: An Introduction to Computational Linguistics New York: Addison Wesley Inc.
All supplemental readings, and all Internet materials, are on reserve in Bobst Library and in the Linguistics Department Library, 719 Broadway, Fifth Floor. Weekly readings are assigned in Dougherty and Gazdar & Mellish.
Check the Academic Computing Facility for lab hours
Each student will complete the course project, which will be given out at the first class meeting and also is described in some detail on the page:
The class will be held in a computerized classroom. Many of the lectures will include computer screens that will be projected onto an overhead screen so students can see programs being executed.
|Introduction to Minimalist Grammar and Logical Programming|
(a) This course (b) The NYU Academic Computing Facility (c ) Dougherty, Preface, Introduction: What is Computational Linguistics? How to Use this Book, pp. I-xlvi. (d) Dougherty, Ch. 1, Natural Intelligence Linguistics and Prolog, pp. 1-23
(a) http://www.nyu.edu/pages/linguistics/ling.html (b) Dougherty, Ch2., How to Read and Write in Prolog, pp. 24-51 (c) Dougherty, Ch3., How to Load, Run, and Edit a Prolog Program, pp. 52-82. (d) Gazdar and Mellish, Introduction, pp. 1-20
(a) http://www.nyu.edu/pages/linguistics/anlcbk.html (b) http://www.nyu.edu/pages/linguistics/parsers.html (c) Dougherty, 7.1. The Levels of Human Language Structure, pp. 161-179 (d) Dougherty, 7.2. Morphological Parsers, pp. 180-194.
|Part 1 of Project: (PF,LF) Pairs as Primary Data
Turn in your paper at the fourth class meeting
(a) http://www.nyu.edu/pages/linguistics/marc/parsshk.html (b) Dougherty, 7.3. Recursion: Affixes on the Affixes, pp. 195-215 (c) Dougherty, 7.4. Regular and Irregular Morphology, pp. 216-228 (d) Dougherty, 7.5. The Minimalist Framework, pp. 229-238
(a) http://www.nyu.edu/pages/linguistics/courses/g611830/german.html (b) http://www.nyu.edu/pages/linguistics/parsers.html (c) Dougherty, 8.1. Syntax: Representations and Parsers, pp. 239-253 (d) Dougherty, 8.2. Rule Governed Creativity: Derivations, pp. 254-265 (e) Gazdar and Mellish, 5.1. A Simple Parsing Problem, pp. 143-144 (f) Gazdar and Mellish, 5.2. Bottom-up Parsing, pp. 144-151 (g) Gazdar and Mellish, 5.3. Top-down Parsing, pp. 152-156
|Make an Appointment with the Proffessor to Discuss the Project
Office Hours or By Appointment:
(a) http://www.nyu.edu/pages/linguistics/ling.html (b) Dougherty, 8.3. Parsers Assign Structure to an Ordered String, pp. 266-275 (c) Dougherty, 8.4. Top Down and Bottom Up Parsing, pp. 276-302 (d) Gazdar and Mellish, 5.5. Comparing Strategies, pp. 165-166 (e) Gazdar and Mellish, 5.6. Breadth-first and Depth-first Search, pp. 166-168 (f) Gazdar and Mellish, 5.7. Storing Intermediate Results, pp. 168-169 (g) Gazdar and Mellish, 5.8. Ambiguity, pp. 169-173 (h) Gazdar and Mellish, 5.9. Determinism and Lookahead, pp. 174-179
(a) http://www.nyu.edu/pages/linguistics/courses/g611830/lecture7.html (b) Look up 'Recursion' in Dougherty and read the references. (c) Look up 'Recursion' in Gazdar and Mellish and read the references.
|Part 2 of Project: Possible Notations at the Level of PF
Turn in your paper at the eighth class meeting
(a) http://www.nyu.edu/pages/linguistics/courses/g611830/lecture8.html (b) Dougherty, Ch 4. Tables of Data as Prolog Facts and Relations, pp. 83-110 (c) Peirce, C.S. How to Make Our Ideas Clear, on reserve in the library.
At this point, you have learned (or at least been exposed to) enough Prolog tools to complete the course project. Dougherty, Natural Language Computing, does not discuss how one might optimally represent features and complex node types in Prolog. Essentially, all feature structures (and node labels) are treated as list structures. For those who want to see alternatives, we will start to examine some of the techniques and procedures presented in Gazdar and Mellish, Natural Language Processing in Prolog. You should regard G&M as a cookbook of techniques, sort of like an encyclopedia of Prolog processes and data structures, that can more or less match the structures one finds in human languages.
(a) http://www.nyu.edu/pages/linguistics/courses/g611830/lecture9.html (b) Dougherty, 8.5. Horizontal Appends: Complement Structures, pp. 303-318 (c) Dougherty, 8.6, Vertical Appends: Selection Restrictions, pp. 319-327 (d) Gazdar and Mellish, 4.1. Grammar as Knowledge Representation, pp. 100-103 (e) Gazdar and Mellish, 4.2. Words, Rules and structures, pp. 104-109 (f) Gazdar and Mellish, 4.3. Representing Simple Grammars in Prolog, pp. 110-114
(a) http://www.nyu.edu/pages/linguistics/courses/g611830/lecture10.html (b) Dougherty, Ch. 5. How Prolog Backtracks in Searches, pp. 111-140 (c) Gazdar and Mellish, 4.4. Subcategorization and the Use of Features, pp. 115-126 (d) Gazdar and Mellish, 4.5. Definite Clause Grammars, pp.127-131 (e) Gazdar and Mellish, 4.6. Classes of Grammars and Languages, pp. 132-142
|Part 3 of Project: Definig the Pairing for the Calendar and PF Notations
Turn in your paper at the fourth class meeting
(a) http://www.nyu.edu/pages/linguistics/courses/g611830/lecture11.html (b) Gazdar and Mellish, 8.1. Semantics: Compositionality, pp. 280-281 (c) Gazdar and Mellish, 8.2. Meaning as Reference, pp. 283-287 (d) Gazdar and Mellish, 8.3. Translation to a Meaning Representation Language, pp. 288-292 (e) Gazdar and Mellish, 8.4. A Database Query Language, pp. 290-293
(a) http://www.nyu.edu/pages/linguistics/courses/g611830/lecture12.html (b) Gazdar and Mellish, 8.5. Computational Semantics as Feature Instantiation, pp. 293-294 (c) Gazdar and Mellish, 8.6. Transitive Verbs and Quantification, pp. 295-301 (d) Gazdar and Mellish, 8.7. Ambiguity, Preferences, and Timing, pp. 301-302 (e) Gazdar and Mellish, 8.8. Building Semantic Checking in the Grammar, pp. 303-308
|Part 4 of Project: Defining the Pairing for the Clock and PF Notations
Turn in your paper at the thirteenth class meeting
(a) http://www.nyu.edu/pages/linguistics/courses/g611830/lecture13.html (b) Gazdar and Mellish, 7.1. Feature-Theoretic Syntax , pp. 218-220 (c) Gazdar and Mellish, 7.2. Feature Structures as Graphs, pp. 221-227 (d) Gazdar and Mellish, 7.3. Feature Structures in Prolog, pp. 228-229 (e) Gazdar and Mellish, 7.4. Subsumption and Unification, pp. 230-237 (f) Gazdar and Mellish, 7.5. The Status of Rules, pp. 238-239
(a) http://www.nyu.edu/pages/linguistics/courses/g611830/lecture14.html (b) Gazdar and Mellish, 7.6. Implementing PATR in Prolog, pp. 240-247 (c) Gazdar and Mellish, 7.7. Chart Parsing with Feature-Based Grammars, pp. 248-255 (d) Gazdar and Mellish, 7.8. Representation of Lexical Knowledge, pp.256-269 (e) Gazdar and Mellish, 7.9. Implementing a Lexicon in Prolog, pp. 270-272 (f) Gazdar and Mellish, 7.10. DAG's Versus Terms, pp. 273-275
|Part 5of Project: Turn in the Final Polog Program