Current 530 Courses

2018-2019:

Term 1

530A: The Sound Pattern of Icelandic –Gunnar Hansson

This seminar will explore the sound system of Icelandic, broadly construed: its phonetics and phonology, and the phonology-phonetics and phonology-morphology interfaces. The main focus will be on those aspects of the Icelandic sound system that are typologically unusual, of special theoretical interest, or lend themselves to methodological approaches (e.g. experimental or computational/corpus-based) that have not previously been applied. A small sample of phenomena of potential interest includes: preaspirated stops, short (monomoraic) diphthongs, voiceless sonorants, prestopped nasals and laterals, excrescent stops, and various other segmental alternations (umlaut, hardening, vowel epenthesis/syncope, cluster simplification, diphthongization, palatalization, deaspiration, and more). Many of these sound patterns interact with inflectional and derivational morphology, giving rise to intriguing interface effects: paradigm gaps, phonologically-conditioned allomorphy, levelling (paradigm uniformity), and probabilistic correlations between stem shape and inflectional class, to name a few. Students will pursue original research projects, individually and/or in groups.

 

 

530F: Natural Language Processing with Deep Learning–Muhammad Abdul-Mageed

Natural language processing (NLP)is the field focused at teaching computers to understand and generate human language. Dialog systems where the computer interacts with humans, such as the Amazon Echo, constitute an instance of both language understanding and generation, as the machine attempts to identify the meaning of questions and generate meaningful answers. Recent advances in machine learning, especially in Deep learning, a class of machine learning methods inspired by information processing in the human brain, have boosted performance on several NLP tasks. Deep learning of natural language is in its infancy, with expected breakthroughs ahead. Solving NLP problems directly contributes to the development of pervasive technologies with significant social and economic impacts and the potential to enhance the lives of millions of people. Given the central role that language plays in our lives, advances in deep learning of natural language have implications across almost all fields of science and technology, as well as many other disciplines like linguistics, as NLP and deep learning are instrumental for making sense of the ever-growing data collected in these fields. This course provides a graduate-level introduction to Natural Language Processing with Deep Learning. The goal of the course is to familiarize students with the major NLP problems and the primary deep learning methods being developed to solve them. This includes problems at various linguistic levels (e.g., word and sub-word, phrase, clause, and discourse). Methodologically, this involves unsupervised, distributed representations and supervised deep learning methods across these linguistic levels. The course also includes providing an introductory level of hands-on experience in using deep learning software as well as opportunities to develop advanced solutions for NLP problems in the context of a final project.

 

Term 2

530B: Sociolinguistics in Social Media–Julian Brooke

This is a project class focused on corpus linguistics in general, and sociolinguistic analysis of social media in particular. We will review the growing body of relevant research from linguistics, computer science, and psychology, and students are expected to formulate novel research hypotheses involving one or more sociolinguistic factors (e.g. gender, age, social background, etc.) and linguistic variables which can be identified with minimal human intervention. Using the Python programming language, each student will collect appropriate documents from the internet, process the data, and carry out a statistical analysis, employing machine learning models where appropriate.

Prerequisite: LING 447G – Python for Linguists. For questions about registering in this course, please contact Will Sarmiento.

 

530C: TBA–Rose-Marie Dechaine