NLP Foundations (RM)

2019-2020

Course Objective

The ability to:
1. find out what a specific NLP task consists of (what the goal of the
task is, what the evaluation data looks like) and explain this to others
2. identify what (linguistic) information is relevant for addressing a
specific task and what analyses are involved in obtaining this
information
3. understand core algorithms and methods used on their fields and
reflect on the strengths and weaknesses that follow from their working
4. find out what the state-of-the-art methods and open issues for a
specific task are by reading and reflecting on appropriate research
papers
5. design systems for various NLP tasks for specific languages (and
implement them given availability of appropriate resources) and find and
apply available resources (annotated data, dictionaries/ontologies,
tools)
6. carry out appropriate evaluations of NLP technologies and reflect on
the outcome of these evaluations.
7. describe their questions, insights, findings and proposals in
academic writing style

Course Content

Natural Language Processing (NLP) is a highly dynamic research field
that
mainly operates on the interface between linguistics and computer
science. In order to get computers to deal well with natural language,
it is important to understand both how language works and how
computational methods work. Computational linguists work on this
interface and have developed methods and technologies for language
analysis.
This course covers technologies and computational models for core
domains of natural language processing (morphology, syntax and
(semantic) parsing, semantics) as well as at least one more applied NLP
task. The main focus of the course is to learn how to analyze various
NLP tasks: what information is relevant? What linguistic properties need
to be dealt with? How can this information be identified and used while
creating computational models and tools? Students are trained to find,
understand and work with the latest developments in this sometimes
rapidly advancing field. The course includes practical components where
we learn to find (more or less) ready-to-use tools that can be used for
various NLP tasks, apply them and reflect on the consequences of
underlying technologies on the performance of the tools.

The course is split up in two components: one 6 ECTS component taught in
period 4 and a 3 ECTS component taught in period 5 which offers the
opportunity to either dive deeper into one of the core technologies
covered in the course or to investigate an application that makes use of
these technologies. It is possible to only follow the 6 ECTS component:
if you are certain that you will only follow the first 6 ECTS you may
want to opt for the non-Research Master variant of this course. It is
more applied, but dives less deep in the theory.

Teaching Methods

During Period 4, there are two lectures a week. In the first lectures,
the main technology and knowledge will be covered. The second lecture is
typically more interactive and reflective in nature. Technical skills
and deeper knowledge are obtained through exercises. During Period 5,
there is one meeting per week where supervision on the final project is
provided.

Method of Assessment

Students hand in a portfolio of (revised) exercises that cover technical
skills and deeper reflective skills. An exam (possibly take-home) is
used to test understanding of fundamental concepts. The third component
consists of carrying out an experiment and writing a short scientific
report. The report and code will be graded (as one component). The three
individual components (portfolio, exam, report + code) need to receive
at least a passing grade (5.5). The grade for the entire course is
formed by the average of the three components.

Target Audience

Research Master Linguistics (specialization Human Language
Technologies). Also accessible for students with comparable background.
Students with a computer science background and limited linguistic
background are recommended to follow the NLP technology course that is
part of the Masters Artificial Intelligence in Period 5. It is not
allowed to get credits for both the first 6 credits of this course and
the NLP technology course (due to overlap).

Custom Course Registration

Students who have completed a similar course as the foundation course successfully may take part in the 3ECTS course independently. They need to contact the teacher before the end of period 4.

Recommended background knowledge

Programming (Python) and Linguistic Foundations (please contact the
teacher in advance in case of doubt).

General Information

Course Code L_AAMPLIN022
Credits 9 EC
Period P4+5
Course Level 500
Language of Tuition English
Faculty Faculty of Humanities
Course Coordinator dr. R. Morante Vallejo
Examiner dr. R. Morante Vallejo
Teaching Staff dr. R. Morante Vallejo

Practical Information

You need to register for this course yourself

Last-minute registration is available for this course.

Teaching Methods Seminar