My current research aims to investigate whether computational natural language processing (NLP) methods could address health professional’s information needs.• Development of AI-powered clinical solutions.• Collaboration with the National Institute of Health and Care Excellence (NICE) • Research into NLP and ML techniques to facilitate the practice of evidence-based medicine (EBM).• Development of question answering (QA) and information retrieval (IR) systems.• Investigation of human-in-the-loop methods.

Worked on next-generation Samsung devices to incorporate computer vision and speech technology to enhance their capabilities.Applied research in: • Personalised Speech Verification• Privacy-preserving machine learning• Federated LearningAchievements:• Proposed a research project for the SR Robotics competition which was accepted. Led the implementation team.• Wrote and presented R&D project proposals containing both a business and technical case in the areas of Machine Learning and Computer Vision for stakeholders. • Implemented solutions in Python, TensorFlow, KALDI and C++ for audio (ASR) and vision-related use cases involving Machine Learning models such as CNN, RNN and LSTM. • Represented Samsung Research UK at the International Conference for Computer Vision (ICCV 2019), Seoul, Korea.

Modules studied:Deep Learning, Probabilistic Machine Learning, Natural Language Processing, Automatic Speech Recognition, Reinforcement Learning, Machine Translation, Statistical Spoken Dialogue Systems, Speech Synthesis (Text-to-Speech) and Practical OptimisationFinal year project: Overcoming Catastrophic Forgetting in Neural Machine Translation. Researched the effectiveness of strategies including elastic weight consolidation when applied to transformer networks

• Applied machine learning to next generation genetic sequencing.• Implemented models using Python (TensorFlow) and R.• Presented models implemented to the Molecular Epidemiology Unit.

Final year project: Comparison of Machine Learning Algorithms to Classify Higgs Boson to Tau Lepton Decays. The aim of the project was to optimise the classification of Higgs boson to tau lepton decays using machine learning techniques (boosted decision trees, multilayer perceptrons and support vector machines). The algorithms were implemented in C++ using the ROOT data analysis framework.Modules studied relevant to machine learning and research interests: Introduction to C++, Statistical Data Analysis and Mathematical Techniques, Condensed Matter Physics, Statistical Physics, Electromagnetism, Waves and Optics.

• Developed algorithms for the world’s first ever cognitive recruitment engine using natural language processing, machine learning and artificial intelligence.• Coded the algorithms in C# using the word2Vec framework and neural networks.• Was asked to continue to work on the project during the term time on a part time basis (until June 2017).

• Improved the mechanical integrity of FlexOLEDs.• Conducted experiments and developed computer models.• Reviewed the relevant academic research literature and produced a management summary.

• Reviewed and summarised the academic research of a number of universities to identify potential collaborations.• Contracted by Cambridge Display Technologies to continue with the research on a part time basis while studying for a degree.