NSERC Chair for
Women in Science & Engineering

ICAN-WISE 2018 Feature

Delaney Lothian

Delaney is co-mentored by Dr. Martha White and Dr. Carrie Demmans Epp at the University of Alberta. Together, they are working on developing an intelligent tutor system to help people learn to speak Plains Cree. The project applies human computer interaction and machine learning methods towards an adaptive system to help protect this at-risk language.

Weeks 1-2

Creating a Language Learning Website for a language that’s not well documented means a lot of content development! As well, I have been reading papers on how people learn languages.

Weeks 3-4

The last couple weeks have involved different aspects of back-end website development. This includes database creation and management as well as researching what front-end interactions are worth logging to ensure the system can promote the appropriate material to the learner.

Weeks 5-6

During the previous two weeks I have been working on understanding the structure of Cree Kinship terms, which is highly dependent on who the subject is, their sex, and much more. This is so that I can create appropriate exercises for the learner to understand it simply, as well as the cultural context.
As well, I have been learning how to use and explain copyright, licensing, and informed consent. In my content creation, I have to interact with the people that are giving me their knowledge and recordings, and I have to make sure they understand what I am doing with it, the credit they will receive, and their rights over it.

Weeks 7-8

The last two weeks I had the opportunity to interview my own family to gather recordings of Cree. My family is historically from Lac Ste. Anne which lands within the regional Cree that I am aiming to teach, and I am thrilled to have the ability to document and use parts of the language they remember.

Weeks 9-10

One of the major things I have been working on has been creating “themes” for Cree language learning. This entails finding words related to specific situations (visiting, hunting, shopping, cooking, etc.) so that the learner can learn words that are relevant to personal and daily scenarios.

Weeks 11-12


An example of some of the pairs I have found

In the beginning of this project I discovered quickly is that the collective linguistic knowledge of Cree is not only not-centralized, but not complete. One of these things include “minimal pairs”. A minimal pair is a pair of words that are the exact same but differ by one letter or sound (e.g. sun, run). They are useful for new language learners to train themselves to listen for subtle differences in words. Part of my work these past few weeks has been trying to automate finding minimal pairs within the word set I have so that I can not only use them myself, but hopefully share them is an easily accessible way.

Weeks 13-14


Me standing in front of the WISEST’s student poster at their summer’s end poster presentation

This summer in our lab we had the privilege of welcoming two WISEST (Women in Scholarship, Engineering, Science, and Technology) summer high school students who helped work on projects and conducted their own research under one of my supervisors, Dr. Carrie Demmans Epp. The last few weeks I spent a lot of time with the student brought on to my project, ensuring a smooth hand off and that we met all the goals she had set in the beginning. Being able to mentor her was a rewarding experience and she made a great contribution to the project.