Natural Language Processing

Syllabus

Introduction: Word statistics, Zipf's Law, corpora.
Mathematical Foundations: Elementary probability, Bayesian statistics, essential information theory.
Linguistic Essentials: Part of speech, phrase structure, semantics and pragmatics.
Corpus Based Work: corpora, text preprocessing.
Collocations: frequency, mean and variance, hypothesis testing.
Statistical Inference: n-gram models, statistical estimators, combining estimators.
Word Sense Disambiguation: Supervised disambiguation, dictionary based disambiguation, unsupervised disambiguation.
Markov Models: Markov models, Hidden Markov Models, HMM implementation, HMM applications, part-of-speech tagging.
Probabilistic Context Free Grammars: Features of PCFG, questions for PCFG, inside and outside probabilities.
Information Retrieval: background, vector space model term distribution models, LSI, web search engines.
Text Categorization: decision trees, maximum entropy modeling, perceptrons, k-nearest neighbor classification, naive Bayes.
Clustering: Hierarchical clustering, non-hierarchical clustering.
Advanced topics: Statistical alignment, machine translation, text summarization, question answering.

List of exercise submissions in csv format and pdf.
An equations-sheet that would be attached to the exam can be seen here.

Jane Austen's works, zipped
list of stopwords stop_words_sorted.txt
shakespeare's tragedies & comedies shakes6.zip
porter stemmer ANSI C porter.c porter_thread_safe.c

NLTK Natural Language Toolkit - a comprehensive toolkit for NLP implemented in python.
Evgeniy Gabrilovich a site with lots of internet resources for NLP.
A site with all papers related to Machine Translation.
companion website for the book
statistical NLP resources
software tools for NLP
Text Categorisation: A Survey nice tutorial about text categorization
Machine Learning in Automated Text Categorization detailed introduction.