Automatically assessing paraphasias in discourse: Breakthroughs and remaining challenges
This talk will cover our recent efforts to automate assessment of paraphasias in discourse using machine learning. We will discuss methods for automatic paraphasia and target identification. We will cover problems we ran into, what we built to address those issues, and what work is still needed. Alexandra Salem works as research staff for the Department of Medical Informatics and Clinical Epidemiology at the Oregon Health & Science University. Her background is in computer science and her work focuses on automating assessment of language disorders using machine learning and natural language processing.