Machine Learning for NLP

2019-2020

Course Objective

This course provides students with the means to conduct NLP research
using machine learning. Students will learn: a) what the main machine
learning technologies used in Natural Language Processing are b) how
they work and how they can be used c) the methodologies for using these
technologies in NLP research and d) how to represent linguistic data and
what the impact is of choices in data representation. By the end of this
course, students will be able to (1) name and describe the (working of)
main machine learning technologies in NLP, (2) apply these technologies
to specific NLP tasks (3) design a research environment where machine
learning is used to solve an NLP problem, and (4) interpret and analyze
evaluation results from machine learning experiments.

Course Content

Machine learning is a dynamic and active research field. The main goal
of machine learning is to develop systems which can automatically solve
different problems without being specifically programmed, i.e. by
learning from the data. In this course, we will focus on the use of
machine learning as a methodology for solving NLP tasks (e.g.
pos-tagging, syntactic parsing, information extraction). We cover both
`traditional' machine learning methods as the latest deep learning
approaches. Representation of language as data plays a prominent role in
this course.
Particular attention will be paid to the methodologies for using machine
learning in NLP research. We will cover the experimental setup, running
existing packages on new tasks and evaluation of overall results as well
as error analysis. The course covers practical skills that can be
useful in industry as well as in academia.
The course can be followed by any student with sufficient linguistic and
programming knowledge.

Teaching Methods

2 lectures of 2 hours. One focuses on machine learning algorithms, the
other on linguistic properties and practical aspects.

Method of Assessment

Students hand in a portfolio of exercises carried out during the course
(50%) and take a final exam (50%). Both components need to receive a
passing grade in order to pass the course (at least 5.5).

Entry Requirements

Programming (Python) and Linguistic foundations

Literature

TBA

Target Audience

This course is specifically designed for students in the Text Mining
1-year master. It can also be followed by Computer Science students
(among others Business Analytics and AI).

General Information

Course Code L_AAMPLIN024
Credits 6 EC
Period P2
Course Level 500
Language of Tuition English
Faculty Faculty of Humanities
Course Coordinator dr. A.S. Fokkens
Examiner dr. A.S. Fokkens
Teaching Staff

Practical Information

You need to register for this course yourself

Last-minute registration is available for this course.

Teaching Methods Seminar
Target audiences

This course is also available as: