Why Linguistics is Hard for Computers

Author: Anne Daly


*This talk was filmed on Tuesday, April 18 2017.

Summary: Computers are able to easily surpass humans at many tasks such as adding large amounts of numbers together, but fail on linguistic tasks that humans do without thinking. In this talk, Kenton Murray will give an introduction to Computational Linguistics and explain the many pitfalls that computers encounter when trying to understand language. Murray will give a background on the work that's been done over the last half century to try and tackle this problem, and finish with a short overview of the cutting-edge work being done here at Notre Dame. This talk is an introduction to Computational Linguistics and is aimed at people with an interest in Linguistics, but who do not necessarily have any background in Computer Science.

Kenton Murray is a 3rd year PhD student in the Computer Science and Engineering Department who focuses on Computational Linguistics, and in particular Machine Translation. His work focuses on using deep learning and other machine learning techniques on linguistically diverse languages. Prior to Notre Dame, he received a Master's in Language Technologies from Carnegie Mellon University and a Bachelor's in Computer Science from Princeton University.