Computing That Serves

New Course for Fall 2005: Statistical Natural Language Processing


CS 601R Section 2, Fall 2005, MWF 10-10:50am taught by Dr. Eric Ringger

This course will provide a thorough introduction to Statistical Natural Language Processing (Stat. NLP) for students who are interested in any of the above problems. We will explore statistical (or empirical) techniques for the automatic analysis of natural (human) language and investigate some of the principal tasks in NLP, including language modeling, word-sense disambiguation, morphological analysis, part-of-speech tagging, named entity tagging, syntactic parsing, semantic interpretation, co-reference resolution, and text classification. We will discuss the linguistic features relevant to each task, how to design statistical models that can accommodate those features, and how to estimate parameters for such models. Each of these tasks is relevant to one or more of the problems listed in the teaser above, and we will explore many of those problems throughout the course.

Desired prerequisites include familiarity with basic probability and statistics and a desire to solve tough natural language problems. Some prior experience with linguistics is helpful but not required. In addition to learning the issues and techniques of statistical NLP, this course aims to help the student build real tools, read current research papers in the field, and see where the opportunities for research await.