SSA QLD Branch October Meeting: clusterBMA: Combine insights from multiple clustering algorithms with Bayesian model averaging

Please join us online for an October Queensland Branch Meeting, on the 5th of October. The seminar will start at 4:00pm, with a branch meeting starting at 5:00pm. Please find details for the seminar below.

TITLE: clusterBMA: Combine insights from multiple clustering algorithms with Bayesian model averaging

SPEAKER: Owen Forbes

TIME: 4:00 PM - 6:00 PM (AEST), Wednesday 5th October 2022

VENUE: Online (Zoom details will be sent with registration)

Please note that the seminar will be recorded and might be put on YouTube or similar platform.

ABSTRACT:

In this talk, I will introduce clusterBMA, a novel Bayesian Model Averaging (BMA) methodology that combines inference across multiple algorithms for clustering of a given dataset. BMA offers some attractive benefits over other existing approaches for ensemble or consensus clustering. Benefits include intuitive probabilistic interpretation of an overall cluster structure integrated across multiple sets of clustering results, flexibility to accommodate various input algorithms, and quantification of model-based uncertainty. We present results from a substantive neuroscience case study, and two simulation studies. This method is implemented in the freely available R package “clusterBMA”, which will be demonstrated during the talk and can be accessed at https://github.com/of2/clusterBMA

SPEAKER'S BIO:

Owen is a final year PhD student under the supervision of Distinguished Professor Kerrie Mengersen at QUT. His research is in applied statistics and neuroscience, studying electrical activity in young people's brains to better understand mental health outcomes and brain development through adolescence. This work contributes to methodology and applications for understanding brain characteristics and mental health through adolescent development, with clinical relevance for risk prediction and early intervention. He is also working on several collaborative applied data science projects in the areas of child health and Indigenous education. As an advocate for mental health action, Owen is also a leader in student-driven wellbeing initiatives, a Peer Support Officer in the ACT State Emergency Service, and a volunteer with Palliative Care ACT. He is passionate about achieving positive societal impacts through data science and research implementation.