Natural Language Processing (202-2-5211)

News:

1. May 15: Assignment 1 is available
2. May 26: Added information on reading the Treebank files in Scheme in this page
3. May 28: Added information on computing KL divergence in this page
4. June 23: Assignment 2 is available
5. June 23: Notes on summarization added.
6. June 23: Notes on text clustering added.

1. General Intro to NLP - Linguistic Concepts
2. Parts of speech Tagging
3. Context Free Grammars Parsing
4. Automatic Text Summarization

1. Language Models and n-grams -- Statistical Models of Unseen Data (Smoothing)
2. Information Extraction / Named Entity Recognition (see Assignment 2)
3. Using Machine Learning Tools: Classification, Sequence Labeling / Supervised Methods / SVM (see Assignment 2)
4. Compositional Semantic from CFG Parsing
5. Sentence Simplification

• Assignment 1: Tagging and Context Free Grammars (CFG and PCFG)
• Assignment 2: Chunking and Named Entity Extraction
• Assignment 3: Summarization: Individual topics to be arranged including
  • Implementation of KL-Sum, Sum-Basic or Lex-Rank and test on DUC2007 dataset.
  • Preparation of a Hebrew dataset for summarization (similar to DUC2007) and implementation of a Hebrew summarizer.

• NLTK Installation: Nltk is a Python based toolkit with wide coverage of NLP techniques - both statistical and knowledge-based.
• SISC Scheme Interpreter: we use Scheme examples to demonstrate algorithms in parsing, generation and some semantic analysis. This interpreter is very small and convenient to use on any platform supporting Java (full version is 2.4MB with full doc - jar is 300KB).

• Bob Carpenter's Computational Linguistics Syllabus - Excellent collection of references in the field.
• The LingPipe Java library suite - by Bob Carpenter. Information extraction and data mining tools.
• The NLTK Toolkit - Python toolkit with thorough tutorial on all Natural Language topics.
• Introduction to Natural Language Processing, by Steven Bird, Ewan Klein and Edward Loper, 2009, distributed on the NLTK site.