My 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).
If you are a self-funded student considering to study for a PhD in any of the topics below please email me to discuss before applying. Click here for information on applying for Research Degrees at Loughborough University.
I am recruiting self-funded PhD students who wish to undertake projects in data science and A.I, including projects within the topics below and here:
Topic 1: 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. For a relevant paper on topic 1 see: 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
Topic 2: Deep learning for noisy unbalanced data: It is a challenging task to train deep learning models on unbalanced 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 unbalanced data obtained from smart environments. For relevant papers on topic 2 see: 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 and COSMA, G. and MCGINNITY, T.M., 2019. Feature extraction and classification using leading eigenvectors: applications to biomedical and multi-modal mHealth data. IEEE Access, 7, pp. 107400-107412. ISSN 2169-3536
Topic 3: 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.
Topic 4: Artificial Intelligence algorithms, reasoning and interpretation: The emphasis of this project is the design and development of artificial intelligence algorithms which can provide reasoning behind predictions and decisions. Focus will be on biomedical applications. For relevant papers on topic 4 see: 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 and COSMA, G., ACAMPORA, G., BROWN, D., REES, R.C., KHAN, M. and POCKLEY, A.G., 2016. Prediction of pathological stage in patients with prostate cancer: a neuro-fuzzy model.PLOS ONE, 11 (6), e0155856.ISSN 1932-6203
Topic 5: 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.
Topic 6: Source-code similarity detection using machine learning and A.I: This project involves developing computational intelligence and machine learning models for analysing, classifying and clustering files in large source-code for indexing and similarity detection tasks. For a relevant paper on topic 6 see: COSMA, G. and JOY, M., 2012. An approach to source-code plagiarism detection and investigation using latent semantic analysis. IEEE Transactions on Computers, 61 (3), pp. 379-394. ISSN 0018-9340