Information Extraction as a Stepping Stone toward
Story Understanding

Ellen Riloff
Department of Computer Science
University of Utah
Salt Lake City, UT 84112
riloff@cs.utah.edu

Historically, story understanding systems have depended on a great deal of handcrafted knowledge. 
Natural language understanding systems that use conceptual knowledge structures [SA77,
Cul78, Wil78, Car79, Leh81, Kol83] typically rely on enormous amounts of manual knowledge
engineering. While much of the work on conceptual knowledge structures has been hailed as pioneering 
research in cognitive modeling and narrative understanding, from a practical perspective
it has also been viewed with skepticism because of the underlying knowledge engineering bottleneck. 
The thought of building a largescale conceptual natural language processing (NLP) system
that can understand openended text is daunting even to the most ardent enthusiasts. So must we
grit our collective teeth and assume that story understanding will be limited to prototype systems
in the foreseeable future? Or will conceptual natural language processing ultimately depend on a
massive, broadscale manual knowledge engineering effort, such as CYC [LPS86]?
Perhaps the answer lies in the current trends of information extraction research. Information
extraction (IE) is a form of natural language processing in which certain types of information must
be recognized and extracted from text. Although IE systems are typically designed for a single domain, 
there is a great deal of interest in building systems that are easily portable to new domains.
Many researchers are developing methods to acquire the necessary domainspecific knowledge
automatically. The goal is not necessarily to produce a generalpurpose information extraction
system, but to create tools that would allow users to build customized information extraction systems 
quickly.
If the IE researchers are ultimately successful, we could imagine being able to create, quite
literally, an IE system ``du jour'' that is tailored to our interests on any given day. Another possible
future scenario is to assemble a large suite of IE systems, each being a specialist in extracting
information pertaining to its own area of expertise. Collectively, the suite of systems could support
conceptual natural language understanding capabilities, if not for every subject then at least for a
wide variety of subjects.
On the surface, information extraction might appear to be fundamentally different from story
understanding. And there are some important differences between these tasks. But we will argue
that the challenges and difficulties underlying them are largely the same. A corollary of this view
is that automated knowledge acquisition techniques developed for information extraction are likely
to be applicable to story understanding as well. In this chapter, we will explain how research in
information extraction may be a significant stepping stone toward progress in story understanding.
We will discuss emerging trends in the information extraction community, draw parallels between
the two types of natural language understanding, and describe recent research in automated case
frame generation for information extraction.


References
[AB95] Chinatsu Aone and Scott William Bennett. Applying Machine Learning to Anaphora
Resolution. In Working Notes of the IJCAI95 Workshop on New Approaches to Learn
ing for Natural Language Processing, pages 151--157, 1995.
[Car79] J. G. Carbonell. Subjective Understanding: Computer Models of Belief Systems. PhD
thesis, Tech. Rept. 150, Department of Computer Science, Yale University, New Haven,
CT, 1979.
[Car93] C. Cardie. A CaseBased Approach to Knowledge Acquisition for Domain-Specific
Sentence Analysis. In Proceedings of the Eleventh National Conference on Artificial
Intelligence, pages 798--803. AAAI Press/The MIT Press, 1993.
[Cul78] R. E. Cullingford. Script Application: Computer Understanding of Newspaper Stories.
PhD thesis, Tech. Rept. 116, Department of Computer Science, Yale University, New
Haven, CT, 1978.
[DeJ82] Gerald DeJong. An Overview of the FRUMP System. In W. Lehnert and M. Ringle, ed
itors, Strategies for Natural Language Processing, pages 149--177. Lawrence Erlbaum
Associates, 1982.
[DGCN91] Charles P. Dolan, Seth R. Goldman, Thomas V. Cuda, and Alan M. Nakamura. Hughes
Trainable Text Skimmer: Description of the TTS System as Used for MUC3. In Pro
ceedings of the Third Message Understanding Conference (MUC3), pages 155--162,
San Mateo, CA, 1991. Morgan Kaufmann.
[DLW + 91] Kathleen Dahlgren, Carol Lord, Hajime Wada, Joyce McDowell, and Edward P. Stabler 
Jr. ITP: Description of the Interpretext System as Used for MUC3. In Proceedings
of the Third Message Understanding Conference (MUC3), pages 163--170, San Mateo,
CA, 1991. Morgan Kaufmann.
[GSM91] Ralph Grishman, John Sterling, and Catherine Macleod. New York University: Description 
of the Proteus System as Used for MUC3. In Proceedings of the Third
Message Understanding Conference (MUC3), pages 183--190, San Mateo, CA, 1991.
Morgan Kaufmann.
[HL94] P. Hastings and S. Lytinen. The Ups and Downs of Lexical Acquisition. In Proceedings
of the Twelfth National Conference on Artificial Intelligence, pages 754--759. AAAI
Press/The MIT Press, 1994.
[Huf95] Scott B. Huffman. Learning information extraction patterns from examples. In Working
Notes of the IJCAI95 Workshop on New Approaches to Learning for Natural Language
Processing, pages 127--134, 1995.
[Huf96] S. Huffman. Learning information extraction patterns from examples. In Stefan
Wermter, Ellen Riloff, and Gabriele Scheler, editors, Connectionist, Statistical, and
Symbolic Approaches to Learning for Natural Language Processing, pages 246--260.
SpringerVerlag, Berlin, 1996. 22
[KM93] J. Kim and D. Moldovan. Acquisition of Semantic Patterns for Information Extraction
from Corpora. In Proceedings of the Ninth IEEE Conference on Artificial Intelligence
for Applications, pages 171--176, Los Alamitos, CA, 1993. IEEE Computer Society
Press.
[Kol83] J. Kolodner. Maintaining Organization in a Dynamic LongTerm Memory. Cognitive
Science, 7:243--280, 1983.
[LCF + 91a] W. Lehnert, C. Cardie, D. Fisher, E. Riloff, and R. Williams. University of Massachusetts: 
Description of the CIRCUS System as Used for MUC3. In Proceedings of
the Third Message Understanding Conference (MUC3), pages 223--233, San Mateo,
CA, 1991. Morgan Kaufmann.
[LCF + 91b] W. Lehnert, C. Cardie, D. Fisher, E. Riloff, and R. Williams. University of Massachusetts: 
MUC3 Test Results and Analysis. In Proceedings of the Third Message
Understanding Conference (MUC3), pages 116--119, San Mateo, CA, 1991. Morgan
Kaufmann.
[Leh81] W. G. Lehnert. Plot Units and Narrative Summarization. Cognitive Science, 5(4):293--
331, 1981.
[Leh91] W. Lehnert. Symbolic/Subsymbolic Sentence Analysis: Exploiting the Best of Two
Worlds. In J. Barnden and J. Pollack, editors, Advances in Connectionist and Neural
Computation Theory, Vol. 1, pages 135--164. Ablex Publishers, Norwood, NJ, 1991.
[LPS86] D. B. Lenat, M. Prakash, and M. Shepherd. CYC: Using Common Sense Knowledge to
Overcome Brittleness and KnowledgeAcquisition Bottlenecks. AI Magazine, 6:65--85,
1986.
[ML95] Joseph F. McCarthy and Wendy G. Lehnert. Using Decision Trees for Coreference Resolution. 
In Proceedings of the Fourteenth International Joint Conference on Artificial
Intelligence, pages 1050--1055, 1995.
[MSBS91] Christine A. Montgomery, Bonnie Glover Stalls, Robert S. Belvin, and Robert E.
Stumberger. Language Systems, Inc.: Description of the DBG System as Used for
MUC3. In Proceedings of the Third Message Understanding Conference (MUC3),
pages 171--177, San Mateo, CA, 1991. Morgan Kaufmann.
[MUC91] MUC3 Proceedings. Proceedings of the Third Message Understanding Conference
(MUC3). Morgan Kaufmann, San Mateo, CA, 1991.
[MUC92] MUC4 Proceedings. Proceedings of the Fourth Message Understanding Conference
(MUC4). Morgan Kaufmann, San Mateo, CA, 1992.
[MUC93] MUC5 Proceedings. Proceedings of the Fifth Message Understanding Conference
(MUC5). Morgan Kaufmann, San Francisco, CA, 1993.
[Ril93] E. Riloff. Automatically Constructing a Dictionary for Information Extraction Tasks.
In Proceedings of the Eleventh National Conference on Artificial Intelligence, pages
811--816. AAAI Press/The MIT Press, 1993.
[Ril96a] E. Riloff. An Empirical Study of Automated Dictionary Construction for Information
Extraction in Three Domains. Artificial Intelligence, 85:101--134, 1996.
[Ril96b] E. Riloff. Automatically Generating Extraction Patterns from Untagged Text. In Proceedings 
of the Thirteenth National Conference on Artificial Intelligence, pages 1044--
1049. The AAAI Press/MIT Press, 1996.
[RL94] E. Riloff and W. Lehnert. Information Extraction as a Basis for HighPrecision Text
Classification. ACM Transactions on Information Systems, 12(3):296--333, July 1994.
[RS95] E. Riloff and J. Shoen. Automatically Acquiring Conceptual Patterns Without an Annotated 
Corpus. In Proceedings of the Third Workshop on Very Large Corpora, pages
148--161, 1995.
[SA77] R. Schank and R. Abelson. Scripts, Plans, Goals and Understanding. Lawrence Erlbaum 
Associates, Hillsdale, New Jersey, 1977.
[SFAL95] S. Soderland, D. Fisher, J. Aseltine, and W. Lehnert. CRYSTAL: Inducing a conceptual 
dictionary. In Proceedings of the Fourteenth International Joint Conference on
Artificial Intelligence, pages 1314--1319, 1995.
[SL94] S. Soderland and W. Lehnert. WrapUp: A trainable discourse module for information
extraction. Journal of Artificial Intelligence Research (JAIR), 2:131--158, 1994.
[Wil78] R. Wilensky. Understanding goalbased stories. PhD thesis, Tech. Rept. 140, Department 
of Computer Science, Yale University, New Haven, CT, 1978.