This course is part of the track Learning and Computation.

This course will focus on modern machine learning based and statistical systems and models for processing natural language. The course will cover advanced formalisms and techniques (both neural and statistical, as appropriate) for predicting syntactic structure, semantic structure, as well as models for machine translation. A portion of the course will be devoted to applications of language processing that are most commonly used in industry (e.g. models of sentiment analysis), and some unsupervised techniques like topic modelling may be included. Together these parts will provide students with an overview of the theory and practice in current data-driven natural language processing, and provide the building blocks for structure processing, language understanding and machine translation systems.