At this week’s MCQLL meeting, Amanda Doucette will be presenting Causal Discovery in the Lexicon: Data, Conditional Independence, and Missingness.

When:
Wednesday, November 13, 12:00–13:00 (Montréal time, UTC-5)
Where:
MCQLL meetings this semester are in hybrid format. We will meet in-person in room 117 of the McGill Linguistics Department, 1085 Dr-Penfield. If you’d like to attend virtually, the Zoom link is here.

All are welcome to attend.

  • Speaker:
    Amanda Doucette
    Title:
    Causal Discovery in the Lexicon: Data, Conditional Independence, and Missingness
    Abstract:

    In this talk, I will present some of my recent work on the causal structure of frequency, word length, phonotactic complexity, and morphological irregularity in the lexicon. First, I will discuss how grapheme-to-phoneme (G2P) methods can be used to extract IPA lexicons from Wikipedia. Then, I will cover some of the challenges in applying causal discovery to real data: conditional independence testing and handling missing data.