Dr Katarina Domijan

    Mathematics and Statistics, Hamilton Institute

    Lecturer
    Assistant Professor

    Logic House
    131
    (01) 708 3374

    Biography

    2008-present: Lecturer/Assistant professor in Statistics, Department of Mathematics and Statistics, Maynooth University.
    2004-2008: PhD in Statistics,    Trinity College Dublin, Ireland.
    2002-2004: Statistician, Bioinformatics Mathematics and Statistics Group, AgResearch, New Zealand
    2001-2002: Master of Computing and Mathematical Sciences (MCMS) with first-class honours, University of Waikato, New Zealand.
    2000 Bachelor of Science (BSc) in Mathematics, University of Auckland, New Zealand

    Other


    2021-present: Associate Editor for The  R Journal
    2015-present: Associate Editor for The Journal of Computational and Graphical Statistics
    2014-2016: Member of the International Space Science Institute (ISSI) International Team in Space Sciences: Improving the Reliability of Solar Eruption Predictions to Facilitate the Determination of Targets-of-Opportunity for Instruments with a Limited Field-of-View.
    2013 Member of the organising committee for Conference on Applied Statistics Ireland (CASI) 2013.

    Research Interests

    Bayesian applications to data analysis, multivariate methods, machine learning, kernel methods, variable selection, statistical computing, data visualization, survey data.

    Research Interests

    My research interests lie in applying Bayesian methods of statistical inference to analyze data of complex structure that arise in a variety of applications. In particular, I am interested in Bayesian modelling of classification problems in data with large feature spaces. For me, the most interesting aspects of modelling high-dimensional data are feature selection and model visualization and interpretability.

    Research Projects

      Project Role Funding Body / Program Description Start Date End Date Award (€)
    Socio-economic cost of food hypersensitivity on the island of Ireland PI SafeFood / Consultancy We propose an extensive literature review and examination of available datasets, including obtaining information on the prevalence of medically diagnosed food allergies (FA) and food hypersensitivity (coeliac disease (CD) in this instance), and identification of areas where information/figures etc. are lacking. This initial review will feed into a larger survey study, which will be constructed with careful consideration to the project objectives, to similar surveys studies (and their methods) in the literature, and with a view to compiling data that is currently unavailable. In order to examine the socioeconomic cost (direct, indirect and intangible) of food hypersensitivity, we plan on carrying out two large surveys of medically diagnosed sufferers with both FA and CD, in the Republic of Ireland (ROI) and in Northern Ireland (NI). Theses two studies (one in each jurisdiction) will be subdivided into questionnaires focused on the parents of children & adolescents with FA (<18 years), adults with FA (>18 years), and individuals with CD. It is our plan to recruit a DIT based postgraduate MSc student to complete this study over a 21 month period (with a submission of MSc thesis at 24 months). This project will include: organising focus groups with FA and CD sufferers and meeting/interviewing stakeholders (such as support groups, medics etc). Dr. Nadim Akasheh, (Consultant, St. James’s Hospital) has also agreed to be a collaborator in this study. Data analysis will be conducted under the supervision of another collaborator Dr. Katarina Domijan (Lecturer in Mathematics & Statistics, MU). The collated survey data will be instrumental in the calculation of the socio-economic cost and the household cost (direct & indirect cost) of having a FA, or a child/adolescent with a FA, in the ROI and NI today. Similarly, the socio- economic and household cost can be calculated for individuals living with CD in each jurisdiction. It is expected that the analysed survey data, in combination with the information gathered at focus groups & meetings with stakeholders, will assist in the formulation of recommendations relevant to citizens suffering with food hypersensitivities in Ireland today. In addition, it is our aspiration that these findings will inform policy, risk communication and self-management, educate stakeholders (including the food sector), and ultimately assist in improving the ‘quality of life’ (QoL) of individuals living with food hypersensitivity on the island of Ireland. 02-SEP-19 31-AUG-21 5516.25

    Peer Reviewed Journals

      Year Publication
    2021 'Interactive Slice Visualization for Exploring Machine Learning Models'
    Hurley, CB;O'Connell, M;Domijan, K (2021) 'Interactive Slice Visualization for Exploring Machine Learning Models'. Journal of Computational and Graphical Statistics, [DOI] [Details]
    2021 'Mid infrared spectroscopy and milk quality traits: A data analysis competition at the “International Workshop on Spectroscopy and Chemometrics 2021”'
    Frizzarin M.;Bevilacqua A.;Dhariyal B.;Domijan K.;Ferraccioli F.;Hayes E.;Ifrim G.;Konkolewska A.;Le Nguyen T.;Mbaka U.;Ranzato G.;Singh A.;Stefanucci M.;Casa A. (2021) 'Mid infrared spectroscopy and milk quality traits: A data analysis competition at the “International Workshop on Spectroscopy and Chemometrics 2021”'. Chemometrics and Intelligent Laboratory Systems, 219 [DOI] [Details]
    2021 'Automated Raman Micro-Spectroscopy of Epithelial Cell Nuclei for High-Throughput Classification'
    Kevin O’Dwyer;Katarina Domijan;Adam Dignam;Marion Butler;Bryan M. Hennelly; (2021) 'Automated Raman Micro-Spectroscopy of Epithelial Cell Nuclei for High-Throughput Classification'. Cancers, [DOI] [Details]
    2021 'The effect of audio-visual smiles on social influence in a cooperative Human-Agent Interaction task'
    Ilaria Torre, Emma Carrigan, Katarina Domijan, Rachel McDonnell and Naomi Harte (2021) 'The effect of audio-visual smiles on social influence in a cooperative Human-Agent Interaction task'. ACM Transactions on Computer-Human Interaction, [DOI] [Details]
    2020 'Generalizing Gain Penalization for Feature Selection in Tree-based Models'
    Bruna Wundervald, Andrew Parnell, Katarina Domijan (2020) 'Generalizing Gain Penalization for Feature Selection in Tree-based Models'. Ieee Access, 8 [DOI] [full-text] [Details]
    2019 'Solar Flare Forecasting from Magnetic Feature Properties Generated by the Solar Monitor Active Region Tracker'
    Domijan, K., Bloomfield, D.S. & Pitié, F. (2019) 'Solar Flare Forecasting from Magnetic Feature Properties Generated by the Solar Monitor Active Region Tracker'. Solar Physics, 294 (6) [DOI] [full-text] [Details]
    2019 'A Study of Food Safety Knowledge, Practice and Training among food handlers in Ireland'
    Julie Gruefeldova, Katarina Domijan, Ciara Walsh (2019) 'A Study of Food Safety Knowledge, Practice and Training among food handlers in Ireland'. Food Control, 105 :131-140 [DOI] [full-text] [Details]
    2017 'Conditional visualization for statistical models: An introduction to the condvis package in R'
    O’Connell M.;Hurley C.;Domijan K. (2017) 'Conditional visualization for statistical models: An introduction to the condvis package in R'. Journal of Statistical Software, 81 [DOI] [full-text] [Details]
    2017 'The Persistence of Sperm and the Development of Time Since Intercourse (TSI) Guidelines in Sexual Assault Cases at Forensic Science Ireland, Dublin, Ireland'
    Casey D.;Domijan K.;MacNeill S.;Rizet D.;O'Connell D.;Ryan J. (2017) 'The Persistence of Sperm and the Development of Time Since Intercourse (TSI) Guidelines in Sexual Assault Cases at Forensic Science Ireland, Dublin, Ireland'. Journal of Forensic Sciences, 62 (3):585-592 [DOI] [full-text] [Details]
    2015 'An exploration of EEG features during recovery following stroke - Implications for BCI-mediated neurorehabilitation therapy'
    Leamy D.;Kocijan J.;Domijan K.;Duffin J.;Roche R.;Commins S.;Collins ;Ward T. (2015) 'An exploration of EEG features during recovery following stroke - Implications for BCI-mediated neurorehabilitation therapy'. Journal of NeuroEngineering and Rehabilitation, 11 (1) [DOI] [full-text] [Details]
    2014 'Applications of Raman spectroscopy to the urinary bladder for cancer diagnostics'
    Laura T. Kerr, Katarina Domijan, Ivor Cullen, Bryan M. Hennelly (2014) 'Applications of Raman spectroscopy to the urinary bladder for cancer diagnostics'. Lasers in Medical Science, 3 :193-224 [DOI] [full-text] [Details]
    2011 'Bayesian kernel projections for classification of high dimensional data'
    Domijan, K;Wilson, SP (2011) 'Bayesian kernel projections for classification of high dimensional data'. Statistics and Computing, 21 :203-216 [DOI] [full-text] [Details]
    2006 'Semi-mechanistic modelling in nonlinear regression: a case study (with discussion)'
    Domijan K., Jorgensen M.A. & Reid J.B. (2006) 'Semi-mechanistic modelling in nonlinear regression: a case study (with discussion)'. AUSTRALIAN & NEW ZEALAND JOURNAL OF STATISTICS, 48 :373-391 [full-text] [Details]

    Book Chapters

      Year Publication
    2021 'Visualization-assisted statistical learning'
    Catherine B. Hurley, Katarina Domijan (2021) 'Visualization-assisted statistical learning' In: Handbook of Computational Statistics and Data Science. Hoboken, United States: John Wiley and Sons Ltd. [Details]
    2003 'Modeling thermal destruction of viruses and bacterial cells'
    Domijan K. & Cox N. (2003) 'Modeling thermal destruction of viruses and bacterial cells' In: Collaborations, Designs and Explorations: A Festschrift for Peter Johnstone. New Zealand: Mosgiel, N.Z. : Statistics and Bioinformatics Group, AgResearch. [Details]

    Conference Publications

      Year Publication
    2021 Automated raman cytology for the classification of triple negative breast cancer cell lines
    Kevin O'Dwyer, Katarina Domijan, Adam Dignam, Zhengyuan Tang, Marion Butler, and Bryan M. Hennelly (2021) Automated raman cytology for the classification of triple negative breast cancer cell lines SPIE [Details]
    2019 Comparison of Machine Learning Models in Food Authentication Studies
    Singh, M;Domijan, K (2019) Comparison of Machine Learning Models in Food Authentication Studies 2019 30TH IRISH SIGNALS AND SYSTEMS CONFERENCE (ISSC) [full-text] [Details]
    2019 The Effect of Multimodal Emotional Expression and Agent Appearance on Trust in Human-Agent Interaction
    Ilaria Torre, Emma Carrigan, Rachel McDonnell, Katarina Domijan, Killian McCabe, and Naomi Harte (2019) The Effect of Multimodal Emotional Expression and Agent Appearance on Trust in Human-Agent Interaction . In: Hubert P. H. Shum, Edmond S. L. Ho, Marie-Paule Cani, Tiberiu Popa, Daniel Holden, and He Wang eds. Motion, Interaction and Games (MIG '19) [DOI] [full-text] [Details]
    2016 On the effect of experimental noise on the classification of biological samples using Raman micro-spectroscopy
    Barton S.;Kerr L.;Domijan K.;Hennelly B. (2016) On the effect of experimental noise on the classification of biological samples using Raman micro-spectroscopy Proceedings of SPIE - The International Society for Optical Engineering [DOI] [Details]
    2014 A review of Raman for multicomponent analysis
    Barton S.J., Hennelly B. M., Ward T, Domijan K & Lowry J (2014) A review of Raman for multicomponent analysis Proc. SPIE 9129, Biophotonics: Photonic Solutions for Better Health Care IV, 91290C [DOI] [Details]
    2014 Classification of bladder cancer cell lines using Raman spectroscopy: a comparison of excitation wavelength, sample substrate and statistical algorithms
    Kerr L. T. , Adams A. , O'Dea S. , Domijan K., Cullen I. & Hennelly B. M. (2014) Classification of bladder cancer cell lines using Raman spectroscopy: a comparison of excitation wavelength, sample substrate and statistical algorithms Proc. SPIE 9129, Biophotonics: Photonic Solutions for Better Health Care IV, 91290E [DOI] [Details]
    2005 A Bayesian Method for Automatic Landmark Detection in Segmented Images
    Domijan K. & Wilson S.P. (2005) A Bayesian Method for Automatic Landmark Detection in Segmented Images ICML workshop on Machine Learning Techniques for Processing Multimedia Content [Details]
    2004 Can Density Sorting at Harvest Segregate Mixed Maturity of Feijoa?
    Clark C.J., White A., Woolf A. & Domijan K. (2004) Can Density Sorting at Harvest Segregate Mixed Maturity of Feijoa? Acta Hort. (ISHS) , pp.93-98 [Details]

    Conference Contributions

      Year Publication
    2018 Data Science, Statistics and Visualisation.
    Catherine B. Hurley, Mark O' Connell, Katerina Domijan (2018) Data Science, Statistics and Visualisation. [Oral Presentation], Interactive tools for model exploration, TU Wien [Details]

    Computer Program

      Year Publication
    2019 condvis2: Conditional visualization for statistical models.
    Catherine Hurley, Mark OConnell, Katarina Domijan (2019) condvis2: Conditional visualization for statistical models. Computer Program [Details]
    2018 BKPC: Bayesian Kernel Projection Classifier.
    Katarina Domijan (2018) BKPC: Bayesian Kernel Projection Classifier. Computer Program [Details]
    2016 condvis: Conditional Visualization for Statistical Models.
    Mark O'Connell and Catherine Hurley and Katarina Domijan (2016) condvis: Conditional Visualization for Statistical Models. Computer Program [full-text] [Details]
    2011 tabuSearch: Tabu Search Algorithm for Binary Configurations.
    Katarina Domijan (2011) tabuSearch: Tabu Search Algorithm for Binary Configurations. Computer Program [Details]

    Thesis

      Year Publication
    2009 Bayesian Kernel Classification for High Dimensional Data with Variable Selection.
    Domijan K. (2009) Bayesian Kernel Classification for High Dimensional Data with Variable Selection. Trinity College Dublin: Thesis [Details]

    Electronic Book

      Year Publication
    2018 Notes for ST463/ST683 Linear Models 1.
    Katarina Domijan, Catherine Hurley (2018) Notes for ST463/ST683 Linear Models 1. Electronic Book [Details]

    Reviews

      Journal Role
    Journal Of Computational And Graphical Statistics Associate Editor

    Recent Postgraduates

      Graduation Student Name University Degree Thesis
    2017 Mark O’Connell Maynooth University PhD Conditional Visualisation for Statistical Models