Foundations of Natural Language Processing 38°

Coordinator: Andrea Galassi

Room: Online via Teams and CIRSFID-ALMA AI

Objective: 10 hours of lecture + 14 autonomous practices (Wednesday)

Ects: 3

LECTURES: 

• 03 May 2023 - from 14.30 to 16.30
  Lesson 1: 
  • Introduction and motivation
  • Basic pre-processing (tokenization, lemmatization, stemming)
  • Regular Expressions
• 17 May 2023 - from 14.30 to 16.30
  Lesson 2:
  • Basic representation methods (BoW, n-grams, ...)
  • Distance metrics
  • Basic tasks and methods (classification, clustering, topic modelling)
• 31 May 2023 - from 14.30 to 16.30
  Lesson 3: 
  • Advanced representation methods: word embeddings (word2vec, GloVe)
  • Advanced tasks (sequence tagging, NER)
  • Constituency and dependency grammars
• 7 June 2023 - from 14.30 to 16.30
  Lesson 5:
  • Brief introduction to recurrent architectures
  • Language models
  • Sequence to sequence tasks (summarization, question answering, machine translation)
• 15 June 2023 - from 14.30 to 16.30
  Lesson 6: 
  • Contextual and domain-specific embeddings (ELMO, BERT, Sentence-BERT, LegalBERT...)
  • State-of-the-art applications