The course introduces students to the current AI methods of modelling natural language and meaning it expresses.
In this course we will discuss ways of representing meaning of words, sentences and conversations with computational methods, both top-down rule-based representations and data-driven representations learned by machine learning. We will contrast them with each other, examine how we can draw inferences or reason with them computationally, and how can they be applied in different language technology tasks and applications.
• natural language meaning, inference, ambiguity and similarity
• formal rule-based representations and techniques
• distributional vector-space models estimated from data
• machine learning-based distributed and contextualised word embeddings and language models
Prerequisites and selection
Admission to the course requires general entry requirements for second-cycle education and a successful completion of a course in
programming, 7. 5 credits and
machine learning, 7. 5 credits
or courses giving equivalent skills and knowledge
Assessment of qualifications
Assessment is based upon the number of credits from previous university studies and the fit of the candidates academic background to the course pre-requisities.
To access their eligibility for the course, candidates are requested to fill in a summary sheet and upload it to the admissions system.
Summary sheet for application
Selection is based upon the number of credits from previous university studies, maximum 165 credits.