Semi-Supervised Natural Language Learning (Summer 2006)

This is the homepage for our reading group on semi-supervised natural language learning that met over the Summer of 2006.

Meeting Schedule

• June 14 - Wean 4625 - Unsupervised word sense disambiguation rivaling supervised methods by Yarowsky (1995) (Andy Carlson) PPT slides
• June 21 - Wean 4625 (1pm) - Discovering Word Senses from Text by Pantel, Lin (2002) (Laura Kieras)
• June 28 - No meeting due to ICML
• July 5 - Wean 4625 (1pm) - Prototype-Driven Learning for Sequence Models by Aria Haghighi and Dan Klein - HLT-NAACL 2006 best student paper (Andy Carlson) PPT slides
• July 12 - Wean 8220 (2pm) - Creating Subjective and Objective Sentence Classifiers from Unannotated Texts by Janyce Wiebe and Ellen Riloff (2005) (Jaime Arguello)
• July 19 - Wean 8220 (2pm) - Learning with Local and Global Consistency by Dengyong Zhou, Olivier Bousquet, Thomas Navin Lal, Jason Weston, Bernhard Scholkopf (2003) (Vitor Carvalho)
• July 26 - Wean 4625 (1pm) - No meeting
• August 2 - Wean 4625 (1pm) - The use of unlabeled data to improve supervised learning for text summarization by MR Amini, P Gallinari (2002) (John Elsas) PPT slides
• August 9 - Wean 4625 (1pm) - Graph Based Semi-Supervised Approach for Information Extraction by Hany Hassan, Ahmed Hassan and Sara Noeman (2006) (Mohit Kumar) (Also see Unsupervised Information Extraction Approach Using Graph Mutual Reinforcement by Hany Hassan, Ahmed Hassan, and Ossama Emam)
• August 16 - Wean 4625 (1pm) - Semi-Supervised Text Classification Using EM by Kamal Nigam, Andrew McCallum, and Tom Mitchell (2006) (Presented by Kamal Nigam of Google Pittsburgh)
• August 23 - Wean 4625 (1pm) - Markov Logic Networks by Matt Richardson and Pedro Domingos (2006) (Tom Mitchell)

This meeting schedule is available as a public Google calendar: XML

Relevant AAAI 2006 Papers

• Domingos et al. Unifying Logical and Statistical AI
• Oren Etzioni, Michele Banko, Michael J. Cafarella. Machine Reading
• Marius Pasca et al. Organizing and Searching the World Wide Web of Facts - Step One: the One-Million Fact Extraction Challenge

Relevant HLT-NAACL 2006 Papers

• Aria Haghighi and Dan Klein. Prototype-Driven Learning for Sequence Models
• Yusuke Shinyama and Satoshi Sekine. Preemptive Information Extraction using Unrestricted Relation Discovery
• Alexandre Klementiev and Dan Roth. Named Entity Transliteration and Discovery from Multilingual Comparable Corpora.

Reading List

Tom Mitchell prepared a reading list for the ReadTheWeb course that is a good starting point for our discussion.

Other relevant papers:

• Brin. (1999). Extracting Patterns and Relations from the World Wide Web. pdf
• Collins and Singer. (1999). Unsupervised Models for Named Entity Classification. pdf
• J. Eisner and D. Karakos, Bootstrapping Without the Boot, in Proceedings of the 2005 Conference on Human Language Technology / Empirical Methods in Natural Language Processing (HLT/EMNLP 2005), Vancouver, Canada, October 6-8, 2005.
• R. Jones, R. Ghani, T.M. Mitchell and E. Riloff, Active Learning with Multiple View Feature Sets, ECML 2003 Workshop on Adaptive Text Extraction and Mining , 2003.
• K. Nigam, A. McCallum, and T. Mitchell, Semi-Supervised Text Classification Using EM, in Semi-Supervised Learning, Olivier Chapelle, Bernhard Scholkopf, and Alexander Zien (eds.), MIT Press, 2006.
• Riloff and Jones. (1999). Learning Dictionaries for Information Extraction by Multi-Level Bootstrapping. pdf
• Sarkar. (1999). Applying Co-training methods to Statistical Parsing. pdf
• Rion L. Snow, Daniel Jurafsky, Andrew Y. Ng. Learning syntactic patterns for automatic hypernym discovery

Other Resources

Xiaojin Zhu's Semi-Supervised Learning Literature Survey provides a good overview of semi-supervised learning methods.

Other Information

Please contact Andy Carlson with any questions or suggestions for the reading group.