| Kursusnavn (dansk): | Natural Language Processing |
| Kursusnavn (engelsk): | Natural Language Processing |
| Semester: | Forår 2017 |
| Udbydes under: | Bachelor i softwareudvikling (bswu) |
| Omfang i ECTS: | 7,50 |
| Kursussprog: | Engelsk |
| Kursushjemmeside: | https://learnit.itu.dk |
| Min. antal deltagere: | 15 |
| Forventet antal deltagere: | 0 |
| Maks. antal deltagere: | 30 |
| Formelle forudsætninger: | There are no formal prerequisites for this introductory course.
However, we do assume that, before enrolling in the course, our students already have at least a basic knowledge of:
* arithmetics,
* probability theory, and
* computer programming.
To facilitate participation in the course, in the introductory sessions, we provide a short overview of these topics as well. |
| Læringsmål: | After the course the student should be able to:
* identify the fundamental problems in NLP
* describe the standard approaches to solving these problems
* create and apply statistical NLP models by using off-the-shelf tools
* evaluate statistical models empirically
* analyse the more advanced approaches to solving fundamental NLP tasks |
| Fagligt indhold: | The course is an introduction to natural language processing (NLP), a field of artificial intelligence where we build machines capable of understanding human languages.
To this end, we typically use various machine learning (ML) techniques.
The course aims at equipping the students with the basic knowledge and skill sets necessary for applying basic NLP in research and development.
The topics include:
* introduction to NLP as applied ML
* review of probability theory
* generative statistical models: n-grams and smoothing
* part-of-speech tagging
* speech recognition and synthesis
* phrase-based parsing
* dependency parsing
* statistical machine translation
* cross-domain and cross-lingual processing
* deep learning with neural networks for NLP
* applications of NLP (e.g., sentiment analysis, computational social science) |
| Læringsaktiviteter: | There are 14 weeks of teaching activities:
* 12 weeks of lectures and exercises (2 x 12 sessions)
* 2 weeks of group work
The lectures provide an outline of NLP as applied ML, and as a field of artificial intelligence. This includes the theoretical frameworks, the fundamental problems, and the standard approaches to solving these problems.
The exercise sessions provide hands-on experience with applying and evaluating NLP tools on multilingual data. They illustrate the concepts provided in the lectures, and facilitate the adoption of an empirical approach to creating and evaluating statistical models.
The course also includes 3-4 homework assignments. The group work sessions are aimed at discussing these assignments through group presentations. |
| Obligatoriske aktivititer: | Der er ingen obligatoriske aktiviteter. Vær venlig KUN at ændre denne tekst når der er obligatoriske aktiviteter.
There are no mandatory activities. Please, change this text ONLY when there are mandatory activities |
| Eksamensform og -beskrivelse: | A11: Skriftlig eksamen (stedprøve) med adgang til internet, skriftlige og trykte hjælpemidler., (7-scale, external exam) Written exam on premises. Exam duration is 4 hours.
Students are allowed to consult the literature.
The exam includes multiple-choice questions, and essay-form problem-solving questions.
|
| Litteratur udover forskningsartikler: | Manning, Schuetze: Foundations of Statistical Natural Language Processing. MIT Press, 1999.
Jurafsky, Martin: Speech and Language Processing. Prentice Hall, 2008. |
| | |
Kursusoversigt