CS 6840 - Natural Language Processing

• Students will be able to build language models for authorship attribution and language detection.
• Students will be able to implement the probabilistic CKY algorithm for unlexicalized syntactic parsing.
• Students will be able to use existing NLP packages to train and evaluate HMMs for tagging.
• Students will develop a theoretical understanding of two major probabilistic approaches to Part of Speech tagging: Hidden Markov Models and Maximum Entropy Models.
• Students will gain a basic understanding of lexical semantics and machine learning algorithms for word sense disambiguation.
• Students will gain an understanding of Markov chains, language models, and associated smoothing techniques.
• Students will gain an understanding of Phrase Structure Grammars, Dependency Grammars, and Combinatory Categorial Grammars.
• Students will gain an understanding of basic text processing stepts such as sentence segmentation, tokenization, and stemming.
• Students will gain an understanding of deterministic and learning-based algorithms for coreference resolution.
• Students will gain an understanding of formal meaning representations based on lambda calculus and first order predicate logic.
• Students will gain an understanding of major categories of machine learning approaches for semantic parsing.
• Students will gain an understanding of semantic roles, frames, and major approaches to semantic role labeling.
• Students will gain an understanding of the major types of ambiguity in natural language processing.
• Students will gain an understanding of unlexicalized and lexicalized syntactic parsing.
• Students will understand the significance of NLP in solving practical problems such as Information Retrieval, Information Extraction, and Question Answering.