openTEL is pleased to announce that next openAIED Special Interest Group will have presentations by Wayne Holmes and Suraj Pandey.
Artificial Intelligence in Education. Promise and Implications for Teaching and Learning
Wayne Holmes
Abstract: Artificial intelligence is arguably the driving technological force of the first half of this century, and will transform virtually every human endeavours. Businesses and governments worldwide are pouring enormous sums of money into a very wide array of implementations, and dozens of start-ups are being funded to the tune of billions of dollars. It would be naive to think that AI will not have an impact on education—au contraire, the possibilities there are profound yet, for the time being, overhyped as well. In this presentation, I will be introducing (launching?) my new co-authored book, “Artificial Intelligence in Education. Promise and Implications for Teaching and Learning” which attempts to provide the right balance between reality and hype.
Bio: Wayne Holmes (PhD, University of Oxford) is a Lecturer in Learning Sciences and Innovation, in the Institute of Educational Technology, The Open University. He co-leads openAIED and OU work in Artificial intelligence in Education. He is also a member of the UK’s All Party Parliamentary Group for Artificial Intelligence (Education Taskforce), lead author of Artificial Intelligence in Education. Promise and Implications for Teaching and Learning (2019), and contributor to UNESCO’s Policy Guidelines for Artificial Intelligence in Education.
Key Information for Semantic Text Similarity: The Case of Automatic Short Answer Grading
Suraj Pandey
Abstract: We introduce the notion of key information for semantic text similarity. We then validate it in the context of the Automatic Short Answer Grading (ASAG) task, an increasingly important application of STS.
We hypothesise that each answer to a question has key information which must be present for the answer to be correct. Such key information becomes invaluable to grade answers for questions not seen during training and improving the performance of automated scoring systems. We use a neural model to find structural alignment between a student response and a reference answer. We then supply key information to the model to achieve improved semantic similarity scores. We test the model on the Beetle and the SciEntsBank corpora. The results show that the key information model provides an improvement over state-of-the-art especially for domain adaptability.
Bio: Suraj is a PhD student at the Open University in Computing & Communication working in the field of Natural Language Processing. He is supervised by Alistair Willis and Paul Piwek. His research interest primarily is to teach computers to understand human text. His current project is Semantic Textual Similarity (STS) applied to Automatic Short Answer Grading (ASAG). Previously Suraj did a Master by Research at the University of York in sentiment analysis of reviews.
We look forward to seeing you at 14:00 on 12th March in Meeting Room 1, Jennie Lee.