Semester: Spring 2017
When and where: From 24th February till 26th May 2017 on 
  • Friday (2.30pm-5.15pm), aula magna, Viale Regina Elena, 295b, building C (next to Unitelma), ground floor

Contact information

Instructor: Prof. Roberto Navigli
Office: room G24, viale Regina Elena, 295 (pal. G)
Phone number: 06 49255161
Email: surname chiocciola di plus uniroma1 plus it (if you are a human being, please replace plus with . and chiocciola with @)
Tutors: Tommaso Pasini and Valentina Pyatkin

Basic information

The Natural Language Processing course introduces a field of Artificial Intelligence which deals with the automatic processing of natural language. The course is taught in English. The student will understand the theoretical and practical fundamentals of how to process natural language automatically at the different levels of morphology, part-of-speech tagging, syntax, semantics, discourse and dialogue. Machine translation and other applications will also be introduced.

The course is currently in the curriculum of both the Laurea Magistrale in Informatica, Master Degree in AI and Robotics and the Laurea Magistrale in Ingegneria Informatica.


 Suggested books


Note that this is new as of 2016: There will be a set of homeworks for attending students which will provide 50% of the final grade and a lightweight course project (50%). Homeworks are valid until the whole month of February 2017. Non-attending students will have to work on a more complex project (about double amount of work).

Project delivery can be done at any time (between 1st June and 30th September 2017 and 1st  January and 17th February 2018) using the following link [TO BE ACTIVATED]. The project can be presented in any form (written report, PowerPoint presentation, etc.), but needs to include a clear quantitative evaluation of its performance. Evaluation is done every 20-30 days, with discussion meetings by appointment. 

Course outline

  • Introduction to Natural Language Processing
  • Finite-state automata and transducers
  • Computational morphology
  • N-gram language models; smoothing; interpolation; backoff
  • Part-of-speech tagging  (including multilingual POS tagging)
  • Syntactic parsing: rule-based parsing; CYK algorithm; Earley's algorithm
  • Computational semantics and lexical semantics
  • Computational lexicons: WordNet
  • Multilingual semantic networks: BabelNet
  • Word Sense Disambiguation and Induction; Entity Linking
  • Neural networks, word embeddings and deep learning
  • Semantic parsing
  • Statistical Machine Translation

No comments:

Post a Comment