Georgina Cosma received a Ph.D. degree in computer science from the University of Warwick, U.K., in 2008. Her PhD thesis topic was on intelligent information retrieval which involved developing novel approaches to detecting similarity in natural language text and source-code files.
Dr Cosma is currently a Senior Lecturer at the Department of Computer Science, Loughborough University. Her research interests strongly reside in the areas of A.I and data science, feature engineering, machine learning (including Deep Learning) and intelligent information retrieval for tasks such as image retrieval and natural language processing.
Dr Cosma is truly fascinated about developing data science and A.I solutions to real-life problems. She is particularly interested in biomedical predictive modelling and using A.I and data science for achieving social good.
Dr Cosma welcomes collaboration from companies who are seeking solutions to data science problems. She has ample experience in collaborating with companies on the development of machine learning solutions requiring the analysis of large complex and/or multi-modal data, including image, biomedical, sensor, natural language and other types of data.
She is the Principal Investigator of The Leverhulme Trust project grant entitled “Novel Approaches for Constructing Optimised Multimodal Data Spaces” (Research Project Grant No: RPG-2016-252). The project concerns the development of A.I algorithms and methods for overcoming the limitations of Deep Learning when applied to various types of data and tasks.
Dr Cosma has experience in supervising PhD students to completion. If you are a self-funded student interested in undertaking a PhD in Data Science and A.I feel free to send me an email. A list of PhD projects can be found here: https://datascienceplus.blog/research-projects/
For applications for Ph.D. studentships follow the link: http://www.lboro.ac.uk/study/postgraduate/apply/
Dr Cosma's research interests are broadly in the area of Data Science and Machine Learning, and include the following topics:
- Predictive modelling (especially biomedical predictive modelling) and the development of machine learning models which provide reasoning behind predictions.
- Data science (including biomedical data science) which involves proposing new methods and approaches for analysing and modelling biomedical and other data.
- Feature selection using combinatorial methods, for finding combinations of features for building predictive models.
- Feature extraction and fusion of multi-modal data generated from smart environments.
- Deep learning algorithms applied to multi-modal and uni-modal data for human activity recognition.
- Approaches to overcoming the limitations of deep learning methods when applied to noisy and large imbalanced data. Improving the learning capabilities of deep neural networks on learning ‘limited data’.
- Continual/lifelong learning tasks and improving the ability of deep neural networks to learn new information without forgetting.
- Information Retrieval for natural language, source code and image data.
Dr Cosma is an:
- EPSRC ICT Prioritisation Panel member
- EPSRC Healthcare Technologies Investigator Prioritisation Panel member
- Active Member and Reviewer of the Engineering and Physical Sciences Research Council Reviewer College
- Active Reviewer of the Economic and Social Research Council (EPSRC)
- Active Reviewer of the Biotechnology and Biological Sciences Research Council (BBSRC)
- Active Reviewer of the Medical Research Council (MRC)
- Reviewer of Cancer Research UK
Dr Cosma is a reviewer of various journals including the following:
- IEEE Access
- IEEE Signal Processing Letters
- IEEE Transactions on Cybernetics
- Memetic Computing
- Artificial Intelligence in Medicine
- Computational and Mathematical Methods in Medicine
- Applied Soft Computing
- Cancer Biomarkers
- Energy Policy
- Member of the Frontiers Editorial Board - Review Editor on the Editorial Board of Connected Health (specialty section of Frontiers in Digital Health).