Syllabus - details

Book by book

Manning and Schütze: Foundations of Statistical Natural Language processing (FSNLP):

• Ch 1
• Ch 2:
  • Sec 2.1 except 2.1.10,
  • the essencials from sec 2.2 up to and including 2.2.3
• Ch 3 is considered known background and should be studied by they who lack this background
• Ch 4
• Ch 5, except 5.3.4
• Ch 6:
  • Introduction
  • Sec 6.1
  • Sec 6.2 up to (but not including) Sec. 6.2.4
• Ch 7, except 7.3-7.4
• Ch 8:
  • Sec 8.1
  • Sec 8.5
• Ch 14:
  • Sec 14.2
    • Introduction+
    • 14.2.1
• Ch 15:
  • Sec 15.1-15.2
• Ch 16:
  • Introduction (up to but not including Sec 16.1)
  • Sec 16.2
  • Sec 16.4

Nivre’s web course: Statistical Natural Language Processing (NW)

• Lect. 1-4

Bird, Klein and Loper: Natural Language Processing with Python (NLTK)

• Ch 1: Sec 1.1-1.3, 1.5
• Ch 2: Sec 2.1-2.2
• Ch 3: Sec 3.1-3.2
• Ch 6: Everything except Sec 6.4

Manning, Raghavan and Schütze: Introduction to Information Retrieval (IIR):

• Ch 13, except 13.2.1
• Ch 14:
  • Introduction
  • Sec 14.1-14.3

Jurafsky and Martin: Speech and Language Processing (J&M)

• Ch 6: Sec. 6.6-6.8 (except 6.6.4)
• Ch 20: Sec 20.7

By subject

Basics: ”Working with texts”

• FSNLP:Ch 1, Ch 4
• NLTK: Ch 1, 2.1, 3.1-3.2
• Slides from lecture 22 Aug

Probability theory

• FSNLP: Sec 2.1 except 2.1.10
• Nivre’s web course: Lect. 1-3

Main concepts of Entropy

• FSNLP 2.2-2.2.3 (We do not expect all details here, but you should know formulas 2.26 and 2.36 from FSNLP and have some ideas about why entropy is an essential concepts.)

Statistics and inference

• FSNLP 5.1-5.3.3
• Nivre’s web course, lect. 4
• Slides from lecture 12 Sept.
• Could be useful to consider other sources as well

Collocations

• FSNLP:Ch. 5, except 5.3.4

Methodology, evaluation, smoothing

• FSNLP:
  • Ch 6  up to (but not including) Sec. 6.2.4
  • Sec 8.1
  • Ch 16: Introduction (up to 16.1)

Naïve Bayes classification and word sense disambiguation

• FSNLP Ch. 7, except 7.3-7.4
• Manning, Raghavan, Schütze, IIR, Ch. 13
• NLTK 6.1-6.3, 6.5

Vector space semantics and IR

• FSNLP Sec 8.1, 8.5, 15.1-2
• J&M, Sec. 20.7
• Slides from lecture 14 Nov.

Vector space classification: Rocchio and k nearest neighbors

• FSNLP 16.4
• IR 14-14.3
• Slides from lecture 14 Nov.

Vector space flat clustering: k means

• FSNLP 14.2: intro+14.2.1
• Slides from lecture 14 Nov.

Linear classifiers, logistic regression, maximum entropy classifiers and tagging

• FSNLP, Ch. 16.2
• Jurafsky&Martin, Sec. 6.6-6.8 (except 6.6.4)
• Ratnaparkhi 1996
• NLTK, sec. 6.6
• Slides from lecture 21& 28 Nov.