Dr Cosma is Associate Professor at the School of Science and Technology.
She is course leader for MRes Computer Science and MRes Electronic Systems courses, and Module Leader for the undergraduate Information Systems Development and the postgraduate Internet Programming modules.
Dr Cosma obtained her PhD in Computer Science from the University of Warwick in July 2008 (EPSRC funded). She joined Nottingham Trent University as Senior lecturer in 2013 and was promoted to Associate Professor in Teaching and Research in October 2018. Dr Cosma has been Associate Fellow at the University of Warwick since 2013.
Dr Georgina Cosma's main research interests are in data science, artificial and computational intelligence in various contexts including risk prediction, behaviour modelling, natural language processing, and smart environments (including smart buildings and cities). She is interested in theoretical and applied research.
Dr Cosma is a member of the Computational Neuroscience and Cognitive Robotics (CNCR) Research Group.
The group is well equipped with state of the art computational and electronics design and test equipment. Of note is a 500 core HPC cluster; and a range of high specification robots, including an iCUB, Sawyer, and Robotniq. Robotics research is facilitated with a 75m2 robot arena with VICON tracking system.
If you are considering to study for MRes or PhD in any of the topics below or other related topics please email Dr Cosma.
Areas of research interest include:
Continual/Lifelong Deep Learning: Training deep neural networks to learn a very accurate mapping from inputs (such as image data, sensor data, text) to outputs (e.g. labels also known as classes) requires large amounts of labelled data. Even when these models are trained, they have limited ability to generalise to conditions which are different to the ones used for training the model. Projects under this topic concern the development of Continual/Lifelong learning algorithms which can learn continuously and adaptively, to autonomously and incrementally develop complex skills and knowledge. Projects include the development of methods for recognising new behaviours in various environments such as smart environments (e.g. cities, homes, healthcare settings), and continuous object recognition.
Deep Feature Selection: Feature selection using Deep Neural Networks has not been well studied, despite its importance which facilitates understanding of data. Projects under this topic concern the development of algorithms which are capable of removing irrelevant features from large unimodal and multi-modal datasets. Projects include detection and analysis of anomalous data in various environments such as smart environments.
Deep learning for noisy imbalanced data: It is a challenging task to train deep learning models on imbalanced data which is commonly generated in real-time scenarios. The complexity escalates when training deep learning algorithms on multi-modal data spaces. Projects include the development of algorithms for classifying imbalanced data obtained from smart environments.
Training Deep Learning algorithms with fewer examples: One-shot and few-shot learning is the act of learning to generalize from one or a few number of training examples per class, respectively. This project concerns the development of k-shot learning algorithms which can learn multi-modal data obtained from various environments using fewer labelled examples. For example, a robot can be taught to learn policies and determine which policy or solution to choose next whilst exploring their environments.
Artificial Intelligence algorithms and reasoning: The emphasis of this project is the design and development of artificial intelligence algorithms which can provide reasoning behind predictions and decisions. Applications where reasoning is particularly needed include biomedical and legal applications.
Multi-modal Information Retrieval: Possible projects include developing algorithms which can optimise retrieval of multi-modal data when given queries in image or textual format, and algorithms for multi-modal cross-lingual (Greek and English) information retrieval systems.
Robot Imitation Learning from Multi-Modal Human Demonstrations: Imitation learning in robots is when a robot is presented with a demonstration of a skill that it should learn to perform on its own.This project will develop a solution to enable robots to exploit and fuse a range of data sources collected during a demonstration. The instructor can be a human or robot instructor performing an action that the robot should imitate. The CNCR Group has several robots that are utilised for imitation learning. One of which is the Sawyer Robot.
Dr Cosma has experience in supervising MRes and PhD students.
PhD - Director of Studies (Main Supervisor)
- Sadegh M. Salesi (in progress) - Novel Computational Intelligence Approaches for Biomedical Predictive Modelling: A focus on Approaches which Provide Reasoning Behind Predictions.
PhD - Co-Supervisor (Second Supervisor)
- Pedro M. Baptista (in progress) - Computational Models of the Retina on Neuromorphic Hardware. Director of Studies is Prof. McGinnity.
- Rowida Alfrjani (completed 2018) - Exploiting Domain Knowledge to Enhance Opinion Mining Using a Hybrid Semantic Knowledge-base Machine Learning Approach. Director of Studies was Dr Taha Osman.
- Olfat M. Mirza (completed 2018) - Style Analysis for Source Code Plagiarism Detection. Dr Cosma co-supervised the student. Director of Studies was Professor Mike Joy from the University of Warwick.
MRes students and selected MSc students
- Gulrukh Turabee (in progress) - Sleep Stage Classification Using EEG Signal Analysis and Deep Learning.
- Salim Sulaiman Maaji (completed 2018) - On-line Voltage Stability Monitoring Using Machine Learning.
- Bhavesh H. Pandya (completed 2018) - Multi-modal Biometric Predictive Modelling using Machine Learning.
- Oluwabunmi T. Oloruntoba (completed MSc. 2018) - A Modifieded Cultural Algorithm for Feature Selection of Biomedical Data.
Opportunities to carry out postgraduate research towards an MPhil / PhD exist and further information may be obtained from the NTU Graduate School
Dr Cosma actively reviews grant proposals for the following funding bodies.
- Member of Review Colleges: EPSRC, ESRC, BBSRC, MRC, and STFC
- EPSRC Panel Member
- Reviewer for Cancer Research UK (CRUK)
- Reviewer for The Leverhulme Trust foundation
Some other Professional memberships:
- Member of IEEE Computer Society and IEEE Computational Intelligence Society
- Member of BCS-IRSG: Information Retrieval Specialist Group
- Member of ACM-SIGIR: Special Interest Group in Information Retrieval
- Member of MEDILINK East Midlands
- Higher Education Academy Fellow
Sponsors and collaborators
Recent grant capture includes
- Principal Investigator on a prestigious 3 year project grant obtained from The Leverhulme Trust. The project is entitled Novel Approaches for Constructing Optimised Multimodal Data Spaces. Duration: October 2016- September 2019. Amount: £115,355
- Principal Investigator on a funded project in the topic of NHS and disease classification Amount: £20,000
- Principal Investigator on a funded project on Machine Learning for Laser Process Real-time control. Amount: £19,000
- Co-Investigator on a funded project on Identifying Prostate Cancer in Patients with low PSA levels using Flow Cytometry data in Collaboration with the John van Geest Cancer Research Centre. Amount: £15,000
- External Grant capture from consultancy projects on data analytics.
A full list of publications can be found via this link: Georgina Cosma's publications
Here is a selection of publications.
Recent journal and conference publications
TAHERKHANI, A., COSMA, G. and MCGINNITY, T.M., 2018. Deep-FS: a feature selection algorithm for deep Boltzmann machines. Neurocomputing, 322, pp. 22-37. ISSN 0925-2312
COSMA, G., MCARDLE, S.E., REEDER, S., FOULDS, G.A., HOOD, S., KHAN, M. and POCKLEY, A.G., 2017. Identifying the presence of prostate cancer in individuals with PSA levels <20 ng ml−1 using computational data extraction analysis of high dimensional peripheral blood flow cytometric phenotyping data. Frontiers in Immunology, 8, p. 1771. ISSN 1664-3224
COSMA, G., BROWN, D., ARCHER, M., KHAN, M. and POCKLEY, A.G., 2017. A survey on computational intelligence approaches for predictive modeling in prostate cancer. Expert Systems with Applications, 70, pp. 1-19.ISSN 0957-4174
TAHERKHANI, A., COSMA, G., ALANI, A.A. and MCGINNITY, T.M., 2019. Activity recognition from multi-modal sensor data using a deep convolutional neural network. In: K. ARAI, S. KAPOOR and R. BHATIA, eds., Intelligent computing. Proceedings of the 2018 Computing Conference, volume 2. Advances in intelligent systems and computing. (857). Chaim: Springer, pp. 203-218. ISBN 9783030011765
SALESI, S., ALANI, ALI A., COSMA, G., 2018. A hybrid model for classification of biomedical data using feature filtering and a Convolutional Neural Network. 2018 Fifth International Conference on Social Networks Analysis, Management and Security (SNAMS), Valencia, 15 10 2018.
MACHADO, P., OIKONOMOU, A., COSMA, G. and MCGINNITY, T.M., 2018. Bio-inspired ganglion cell models for detecting horizontal and vertical movements. In: 2018 International Joint Conference on Neural Networks (IJCNN 2018), Rio de Janeiro, Brazil, 8-13 July 2018.
PANDYA, B., COSMA, G., ALANI, A.A., TAHERKHANI, A., BHARADI, V. and MCGINNITY, T.M., 2018. Fingerprint classification using a deep convolutional neural network. In: 2018 4th International Conference on Information Management (ICIM2018), 25-27 May 2018, Oxford, UK. IEEE, pp. 86-91. ISBN 9781538661482
ALANI, A.A., COSMA, G., TAHERKHANI, A. and MCGINNITY, T.M., 2018. Hand gesture recognition using an adapted convolutional neural network with data augmentation. In: 2018 4th International Conference on Information Management (ICIM2018), 25-27 May 2018, Oxford, UK. IEEE, pp. 5-12. ISBN 9781538661482
MAAJI, S.S., COSMA, G., TAHERKHANI, A., ALANI, A.A. and MCGINNITY, T.M., 2018. On-line voltage stability monitoring using an Ensemble AdaBoost classifier. In: 2018 4th International Conference on Information Management (ICIM2018), 25-27 May 2018, Oxford, UK. IEEE, pp. 253-259. ISBN 9781538661482
OLORUNTOBA O. and COSMA G. (2019) A Modified Cultural Algorithm for Feature Selection of Biomedical Data. Proceedings of the 2018 Computing Conference, volume 2. Advances in intelligent systems and computing.See all of Georgina Cosma's publications...