Each of Insight's main centres has a long track record of data analytics research. In July 2013 they came together under Science Foundation Ireland as Insight. The size of the centre allows for collaboration on a large scale, which enables the organisation to compete for funding and opportunities at a much higher level.
The Insight Centre for Data Analytics is one of Europe’s largest data analytics research organisations, with 400+ researchers, more than 80 industry partners and over €100 million of funding.
Insight is made up of four main centres at Dublin City University, National University Ireland Galway, University College Cork and University College Dublin as well as a number of affiliated bodies.
Below are the details of potential hosts and their research interests at Insights. Please get in touch if you are interested in travelling to collaborate with Insights, or in hosting ECRs or Students from Insights.
Dr Adrian O'Hagan teaches and conducts research at the interface of the actuarial science and statistics disciplines. He currently supervises two PhD students who work on advanced statistical models for improved general insurance pricing and statistical genetics models with mortality and morbidity applications for actuaries respectively. He is interested in developing and applying advanced statistical techniques to impending actuarial research areas of interest including use of wearable technology data for health insurance pricing and sensor technology data for car insurance pricing.
Michael has interests in Genetics where he's primarily interested in genetics datasets that exhibit population structure. Specifically where individuals may be grouped into sub-populations that have significantly different frequencies of mutations. Michael is interested in modelling and quantifying recombination, mutation, drift, admixture, linkage, etc. He's also interested in developing statistical methods to detect selection and association with phenotypes such as diseases.
His other substantive area of research is in network analysis. He develop methods and models for networks datasets. Previously this work has focussed on methodological research but recently has moved towards specific complex applications such as wildlife conservation projects where we try to determine important underlying patterns in the social networks of endangered animal populations.
Brendan has long standing research interests in areas including mixture modelling, latent variable models, model-based cluttering, network modelling. He is also interested in a variety of applications in bioinformatics, genomics and sociology.
Claire's research interests include latent variable models, mixture models, network models, computational statistics and Bayesian statistics. Her research to date has developed statistical methods which have been applied in a wide range of fields, varying from social science to metabolomics to orthopaedics. Much of her work is computationally intensive and involves working with high-dimensional data sets. Her current research grants support work in the areas of model-based clustering and classification for high-dimensional data of mixed type.
Nial's research interests are in the areas of computational statistics, statistical network analysis, Bayesian statistics, Monte Carlo methods and intractable likelihood problems. He is also interested in applications of these areas including the modelling and analysis of network data and sports analytics.
Michelle's primary research area of interest is functional data analysis. There is an abundance of data that is Functional (i.e., curves, surfaces, trajectories, images). Functional Data Analysis (FDA) (Ramsay and Silverman (2005)) deals with methodology for analysing functions, on a compact interval on the real line. Her research aims at extending FDA techniques such as Predictive/Model Inference, Smoothing and Alignment to the analysis of data in 2 and 3 dimensions defined over complex domains. This involves partitioning the complex domain into a collection of triangular elements using a Delaunay triangulation (Shewchuk, 1996); defining bivariate spline functions on each triangle (Lai and Schumaker, 2007) and attaining statistical inference on the parameters of the spline functions and any parameters of the model. These High-Dimensional Functional Data techniques are used to model agricultural and aquaculture, atmospheric conditions, energy resources and human health and safety.