Considerations and Advances in Applying AI-based Methods to Assessing Writing

Abstract
The ability to convey information through writing is a central component of real-world skills. However, having instructors assess writing can be time consuming, limiting the timeliness and effectiveness of feedback. With the growth of machine learning and Natural Language Processing, AI-based methods to automatically assess writing have been shown to be effective across a number of applications. Indeed, they are now used widely for high stakes testing and some formative assessments. In this talk, I will describe the current state of the field as well as some new advances. This will include a) an overview of the current AI-based approaches to assessing writing b) examples of applying automated scoring to content domains, performance assessments, and large scale data analytics c) some novel human-AI synergistic scoring approaches, and finally d) some implications, considerations and caveats for applying AI-based models in education.

Peter Foltz is Vice President in Pearson's AI and Products Solutions and Research Professor at the University of Colorado’s Institute of Cognitive Science. His work covers artificial intelligence and uses of machine learning and natural language processing for educational and clinical assessments, large-scale data analytics, reading comprehension and writing skills, and 21st Century skills learning, He has served as content lead for framework development for several OECD PISA assessments, including the 2018 Reading Literacy assessment, the 2015 assessment of Collaborative Problem Solving, and a new assessment of reading literacy for developing countries. He has served as guest editor for a number of journals including International Journal of AI in Education and Discourse Processes as well as co-editor of the forthcoming Handbook of Automated Assessment. He previously worked at New Mexico State University, Bell Communications Research, University of Pittsburgh’s Learning Research and Development Center, Yale University, and the Harvard Institute for International Development.

Date: 
Tuesday, March 3, 2020 - 2:00pm
Building: 
2121 Berkeley Way West
Room: 
1215