Next: Natural Language Processing
Up: Lent Term 2004: Part
Previous: Database Theory
  Contents
Information Retrieval
Lecturer: Dr S.H. Teufel
No. of lectures: 8
Prerequisite courses: Natural Language Processing, and a basic encounter with Probability is assumed
Aims
The course is aimed to characterise information retrieval in terms of
the data, problems and concepts involved. The two main formal
retrieval models and evaluation methods are described. The course then
covers problems and standard solutions in information extraction, and
in question answering.
Lectures
- Information retrieval introduction.
Key problems and concepts. Information need. Indexing model. Examples.
- Retrieval models. Vector Space model. Probabilistic model.
- Evaluation methodology. Evaluation and advanced models (relevance
feedback, query expansion).
- Search engines and linkage algorithms.
PageRank and Kleinberg's Hubs and Authorities.
- Information extraction.
Task and evaluation. Lexico-semantic patterns.
- Advanced information extraction methods.
Bootstrapping. Learning.
- Question answering.
Performance criteria and effectiveness measures, test methodology,
established results.
- Outlook and present needs.
Relevance of NLP in IR, IE, QA.
Objectives
At the end of this course, students should be able to
- define the tasks of information retrieval, question
answering and information extraction and differences between them
- understand the main concepts and strategies used in IR, QA, and IE
- appreciate the challenges in these three areas
- develop strategies suited for specific retrieval, extraction or question
situations, and recognize the limits of these strategies
- understand (the reasons for) the evaluation strategies developed for
these three areas
Recommended books
Baeza-Yates, R. & Ribiero-Neto, B. (1999). Modern information
retrieval. Reading, MA: Addison-Wesley and ACM Press.
Spärck Jones, K. & Willett, P. (eds.) (1997). Readings in
information retrieval. San Francisco: Morgan Kaufmann.
Salton, G. & McGill, M.. (1983). Introduction to modern
information retrieval. New York: McGraw Hill.
Next: Natural Language Processing
Up: Lent Term 2004: Part
Previous: Database Theory
  Contents
Christine Northeast
Thu Sep 4 15:29:01 BST 2003