Information extraction from text; Material

Lecture notes (also available in the course folder on the 4th floor)

1. Lecture 31.1. [PostScript (6 slides/page)] [PowerPoint]

   Lecture based on the article: Grishman: Information Extraction: Techniques and Challenges (for download: see the links on the upper right corner).

   Sample of documents on terrorist attacks (MUC-4).

2. Lecture 25.2. [PostScript (6 slides/page)] [PowerPoint]

3. Lecture 26.2. [PostScript (6 slides/page)] [PowerPoint]

   • Riloff: Automatically Constructing a Dictionary for Information Extraction Tasks (AutoSlog)

   • Soderland et.al; Crystal: inducing a conceptual dictionary

   • Riloff: Automatically Generating Extraction Patterns from Untagged Text (AutoSlog-TS)

   • Riloff, Jones: Learning Dictionaries for Information Extraction by Multi-level Bootstrapping

   • Yangarber, Grishman, Tapanainen, Huttunen: Automatic Acquisition of Domain Knowledge for Information Extraction (ExDisco)

4. Lecture 18.3. [PostScript (6 slides/page)] [PowerPoint]

   • Soderland: Learning Information Extraction Rules for Semi-structured and Free Text (Machine Learning, 1999)

   • Harabagiu et al, FALCON: Boosting Knowledge for Answer Engines

   • Soubbotin et al, Patterns of Potential Answer Expressions as Clues to the Right Answers (Insight Q/A)

Resources

• NIST Message Understanding Conference page