Week 6

Lecture 7: Word semantics and Embeddings 27 Sept.

Presentation

Recordings

• First hour: Word semantics, word vectors, tf-idf
• Second hour: Word embeddings

Mandatory reading

Jurafsky and Martin, Speech and Language Processing, 3. ed. (edition of Sept. 2021!)

• Ch. 3, sec 3.3 "Sampling sentences from a language model"
• Ch. 6, "Vector Semantics and Embeddings", everything except
  • Not section 6.6 Pointwise Mutual Information (PMI)
  • Some of the technical formulas in se. 6.8
• Ch. 18, "Word Senses and Word Net"
  • Sec. 18.0-18.3

Recommended reading

For they who need a brush-up on simple algebra and geometry, we recommend the first two sections, page 1-17, of our own

• Compendium on Geometry and linear algebra for IN3050/IN4050

Lab-session 4, Thursday 30 September at Sed

• Program: Weekly exercise set 3