University Calendar

AIG for Idioms: Towards testing knowledge of English idiomatic expressions with automatic item generation
Thursday, February 23, 2023 4-5PM 
Dr. Michael Flor, Educational Testing Service

Option to join the talk on Zoom: https://montclair.zoom.us/j/84026504566?pwd=VmlCd1c3aXFlZ3hLZFJBcjF5bjZ4UT09

Idioms are ubiquitous in natural languages, very much so in English. Testing English language proficiency is a very common endeavor, both for L1 and L2 (EFL) populations, but there seems to be no specific test of knowledge for English idioms. This talk will present a work in progress, an effort to design a test of knowledge of English idioms. A major part of work is an effort to provide computational linguistic support for automatic generation of items (AIG) for such a test. As idioms are not a fully-solved issue in linguistics, we encounter some of the classic issues in this project as well.   

Bio: Dr. Michael Flor is a Senior Research Scientist in the center for Learning and Assessment Foundations, in the R&D Division of the Educational Testing Service (ETS). Michael is a computational linguist. He earned a PhD in cognitive psychology with specialization in psycholinguistics from Tel Aviv University, Israel.  He joined ETS in 2005. His research focuses on automatic processing of text data for education, combining linguistic, statistical, and cognitive approaches. His work spans both published research and development of applied systems. His work included automatic spelling correction for student essays, automated estimation of text complexity, figurative language detection, measuring lexical cohesion in text, automatic scoring of essays, and automatic question generation. His current research interests include idiomatic expressions and automatic item generation (AIG).