NTU, Linguistics and Multilingual Studies

Dennis Tay is Professor in Linguistics and Multilingual Studies, School of Humanities, Nanyang Technological University. He has held previous positions in Hong Kong, Singapore, and New Zealand. His research interests include the intersecting areas of cognitive linguistics, metaphor, mental health discourses, and data analytics. He is Co-Editor-in-Chief of Metaphor and the Social World, Associate Editor of Metaphor and Symbol, and Co-Series Editor of the new Bloomsbury Studies in Language and Healthcare.
Machine learning meets metaphor – a case study of metaphor association rules in trauma vs. non-trauma talk
Metaphor analysis offers insight into relationships between language, cognition, and cultural conceptualizations, which are key strands in multilingualism and multiculturalism studies. This analysis can leverage machine learning techniques to attain better results for theory and practice across fields. In this talk, I share my recent application of Association Rule Mining to develop the construct of ‘metaphor assemblages’. These are statistically defined as metaphors that co-occur over multiple discourse units at above-chance level. In the context of mental healthcare activities like psychological interviews and counseling, such assemblages may highlight underexplored (dis)similarities between clients or client groups, beyond the typical clinical focus on isolated metaphors. I compare metaphor associations produced by two groups – one meeting the diagnostic threshold for acute stress disorder (ASD), the other not – relating their first-hand experiences of traumatic events during the 2019 Hong Kong social unrest. Results show i) a shared rule suggesting a highly schematic construal of trauma that nevertheless varies in instantiating details, and ii) several distinct rules that corroborates and lends further insight into the “agentive vs. non-agentive” characteristics exhibited by ASD vs. non-ASD individuals. Implications and future directions will be discussed. If time permits, I will also share my thoughts on pedagogical opportunities that such work may bring to humanities students without a ‘quantitative background’.