SIMS 296a-4: Text Data Mining Schedule and Lectures

Overview	Lectures and Readings Go directly to the current week. We are using the following text for background: Foundations of Statistical Natural Language Processing by Manning and Schuetze, MIT Press 1999. August 23: Reading: Textbook Chapter 1 (Introduction) Reading: Untangling Text Data Mining. Discussion Leader: Marti Hearst What is Text Data Mining? (LINDI Intro) Discussion Leader: Marti Hearst Intro to Computational Linguistics August 31: Reading: Textbook Chapter 3 (Intro to Linguistics) Discussion Leader: Ronald Sprouse, Intro to Linguistics Discussion Leader: Hao Chen, Current LINDI prototype September 6 Labor Day: No class September 13 Reading: Textbook Chapter 2 (Intro to Probability Theory and Information Theory) Reading: LINDI documents Discussion Leader: Barbara Rosario, Mathematical Foundations September 20 Reading: Textbook Chapter 16 (Text Categorization) Homework: Investigate a project text domain Discussion Leader: Hao Chen September 27 Reading: No reading Homework: Investigate examples within a domain. Discussion Leader: group work October 4 Readings: Textbook Chapter 5: Collocations Allen article introducing mixed-initiative interactions (from IEEE Intelligent Systems, Trends & Controversies, to appear. Just read the first of these three unless you want to read more.) Horvitz article on mixed-initiative interaction (from ACM CHI 99). Hearst paper on computing directionality within senteces (From Text-based Intelligent Systems, edited by Paul Jacobs, Lawrence Erlbaum Associates, 1992.) Homework: none Discussion Leader: Andy Dolbey October 11 Readings: Textbook Chapter 6: Statistical Inference Plus three papers -- get hardcopies outside my office (212 South Hall) Rajeev Agarwal and Lois Boggess, "A Simple but Useful Approach to Conjunct Identification", Proceedings of the 30th Annual Meeting of the Association for Computational Linguistics, 15-21, 1992 Lois Boggess and Rajeev Agarwal and Ron Davis, "Disambiguation of Prepositional Phrases in Automatically Labelled Technical Text", in Proceedings of AAAI 91, 155-159, 1991. Sanda M. Harabagiu and Dan I. Moldovan "Knowledge processing on an extended WordNet", WordNet: An Electronic Lexical Database, Christiane Fellbaum (ed.), MIT Press, 1998. Homework: Each person has a task; we are working with the MESH Scope Notes. See http://www.nlm.nih.gov/mesh/99MBrowser.html Discussion Leader: Jonathan Henke Ngrams over sparse data October 18 Readings: Four papers: Riloff, E. (1993) "Automatically Constructing a Dictionary for Information Extraction Tasks", Proceedings of the Eleventh National Conference on Artificial Intelligence (AAAI-93) , AAAI Press/The MIT Press, pp. 811-816. n Riloff, E. and Schmelzenbach, M. (1998) "An Empirical Approach to Conceptual Case Frame Acquisition", In Proceedings of the Sixth Workshop on Very Large Corpora , 1998. Riloff, E. and Jones, R. (1999) "Learning Dictionaries for Information Extraction by Multi-Level Bootstrapping," In Proceedings of the Sixteenth National Conference on Artificial Intelligence (AAAI-99) , 1999. (More of Ellen Riloff's papers can be found at http://www.cs.utah.edu/~riloff/publications.html) Wendy Lehnert, "Cognition, Computers, and Car Bombs: How Yale Prepared Me for the 90's," in Beliefs, Reasoning, and Decision Making: Psycho-logic in Honor of Bob Abelson, Schank & Langer (Eds.), Lawrence Erlbaum Associates, 199? Homework: Each person has a task. Discussion Leader: Marti Hearst October 25 Readings: Resnik, P. Using Information Content to Evaluate Semantic Similarity in a Taxonomy, IJCAI 95. A longer and more recent version appears in JAIR, 11, 1999. Dolan, William ; Vanderwende, Lucy ; Richardson, Stephen D. Automatically Deriving Structured Knowledge Bases From On-Line Dictionaries In Proceedings of the Pacific Association for Computational Linguistics, April 21-24, 1993, Vancouver, British Columbia. Optional Reading: Ken Church on ngrams, mutual information, t-tests, smoothing, backing off statistics, entropy, binomial distributions, and suffix arrays, applied to lexicography and part of speech tagging. (123 pages long) Ken Church on practical tips on how to implement these kinds of things simply and efficiently using Unix tools. (53 pages long) November 1 Readings: Textbook Chapter 9: HMMs. Discussion Leader: David Blei Homework: Each person has something to do. November 8 Readings: Textbook Chapter 10: Part of Speech tagging. Discussion Leader: Barbara Rosario Homework: Each person has something to do. November 15 Readings: Textbook Chapter 8 (only through 8.4): Lexical Acquisition. S. Soderland, Learning Information Extraction Rules for Semi-structured and Free Text. Machine Learning, 1999. (23 page compressed postscript) Peter Ladkin The X-31 and A320 Warsaw Crashes: Whodunnit? Discussion Leader: Hao Chen Homework: Each person has something to do. November 22 Homework: Writeup goals for end of semester; get work done on those goals.