Career Profile

I am a statistician and data scientist who specializes in Natural Language Processing (NLP), with years of experience in training and deploying large language models in academia and industry. I spent fifteen years as a classroom teacher before returning to complete my PhD at Peabody College at Vanderbilt University. I use NLP tools, including traditional ML and transformer models in educational tools and assessment, including predicting human ratings of text quality, question and answer generation, and sequence classification. I am also interested in statistical techniques such as multi-level modeling, structural equation modeling, item response theory, and many-facet Rasch measurement to examine rater effects in research and educational measurement. My secondary research interest is in using deep learning techniques to analyze speech acoustics, specifically in terms of prosody, phonology, and pragmatics of natural speech.

Research Experience

PhD Research Assistant

2022 - Present
Vanderbilt University, Nashville TN

Worked under guidance of Professor Scott Crossley on multiple projects including:

  • Worked on the Feedback Competition, a Kaggle competition to develop an AI essay annotation tool able to divide essays correctly and automatically into their discourse elements. Performed analysis on the corpus data using Python and R.
  • Worked with the lab to develop iTELL, a framework for generating intelligent and interactive textbooks from expository texts provided by content creators.

21st Century World Track Lead for Tools Competition

2022 - Present
The Learning Agency

Worked for The Learning Agency helping to administer the Schmidt Futures funded Tools Competition, a grant competition to develop educational technologies in assessment and learning science as part of a graduate research assistantship at Vanderbilt University.

Graduate Research Assistant

2011 - 2013
Georgia State University, Atlanta GA

Worked under Professor Scott Crossley, helping him with data analysis, transcription, research, and essay rating according to analytic rubrics.

Certifications and Awards

Grand Prize Winner, NAEP Math Automated Challenge

2023 - 2023
NAEP

Worked with a team to develop automated scoring models to predict human scores of constructed responses provided by students explaining their reasoning during the NAEP Math assessment

Montessori Teaching Certificate

2013 - 2016
Association Montessori Internationale
AMI

Completed 9-month (3 summers) in-person Montessori teacher training for elementary grades through the Association Montessori Internationale

Projects

I am involved in a number of NLP and Data Science projects in Peabody College at Vanderbilt University and beyond.

iTELL - A framework for automatically generating intelligent, interactive webapps for adult learners.
MASCoT-CP - I am co-PI on a project to develop multimodal models for the classification of teacher classroom practices. This project is currently a finalist for a $150,000 grant from the Tools Competition
Meta-cognitive Strategies - I am co-PI on a project to classify meta-cognitive strategies from student self-reports in natural language.

Publications

Published papers in journals and conference proceedings

  • Formative Feedback on Student-Authored Summaries in Intelligent Textbooks using Large Language Models.
  • Morris, W., Crossley, S., Holmes, L., Ou, Chaohua, Dascalu, M., & McNamara, D.
    Journal of Artificial Intelligence in Education. 2024
  • Measuring second language proficiency using the English Language Learner Insight, Proficiency and Skills Evaluation (ELLIPSE) Corpus.
  • Crossley, S. A., Tian, Y., Baffour, P., Franklin, A., Kim, Y., Morris, W., Benner, B., Picou, A., & Boser, U.
    International Journal of Learner Corpus Research. 2023
  • Using Transformer Language Models to Validate Peer-Assigned Essay Scores in Massive Open Online Courses (MOOCs).
  • Morris, W., Crossley, S. A., Langdon, H., & Trumbore, A.
    Proceedings of the Thirteenth International Conference on Learning Analytics & Knowledge, Arlington Texas, March 13-17, 2023,
  • Using Transformer Language Models to Provide Formative Feedback in Intelligent Textbooks.
  • Morris, W., Crossley, S., Holmes, L., Ou, C., McNamara, D., Dascalu, M.
    Proceedings of the Twenty-fourth International Conference on Artificial Intelligence in Education (AIED), Tokyo, Japan, July 3–6, 2023.
  • Deidentifying Student Writing with Rules and Transformers.
  • Holmes, L., Crossley, S., Morris, W., Sikka, H., & Trumbore, A.
    Proceedings of the Twenty-fourth International Conference on Artificial Intelligence in Education (AIED), Tokyo, Japan, July 3–6, 2023.
  • Formative Feedback on Student-Authored Summaries in Intelligent Textbooks using Large Language Models.
  • Morris, W., Crossley, S., Holmes, L., Ou, Chaohua, Dascalu, M., & McNamara, D.
    Journal of Artificial Intelligence in Education. 2024
  • iScore: Visual Analytics for Interpreting How Language Models Automatically Score Summaries.
  • Coscia, A., Holmes, L., Morris, W., Choi, J., Crossley, S., & Endert, A.
    Proceedings of ACM Conference on Intelligent User Interfaces. 2024.
  • The english language learner insight, proficiency and skills evaluation (ellipse) corpus
  • Crossley, S., Tian, Y., Baffour, P., Franklin, A., Kim, Y., Morris, W., Benner, M., Boser, U.
    International Journal of Learner Corpus Research 9(2). 2023.
  • Automated scoring of constructed response items in math assessment using large language models
  • Morris, W., Holmes, L., Choi, J.S., Crossley, S.
    International Journal on Educational Data Mining. 2024
  • Plagiarism Detection Using Keystroke Logs
  • Crossley, S., Tian, Y., Choi, J.S., Crossley, S.
    International Journal on Educational Data Mining. 2024

    Skills & Proficiency

    Python

    R

    JavaScript

    HTML5 & CSS