Research Associate in Heath Text Analytics and Data Science
We are recruiting a post-doctoral natural language processing (NLP) researcher to join our team of NLP specialists, data scientists, clinicians, and engineers, in support of a new project funded by a multi-site UKPRP award for a Violence, Health and Society research consortium. The successful applicant will join a team leading NLP research and methodology development, and their application to electronic health records (EHRs) and other text to solve real-word health research questions. The project in this instance builds on recent NLP to ascertain instances of received, perpetrated, and witnessed violence, applied to mental healthcare EHRs. The post-holder will further develop this work to improve the characterisation of these experiences and to facilitate the application of NLP across text from different administrative sources. The post-holder will collaborate with a multi-disciplinary, multi-agency team to access relevant text, and to define requirements for NLP.
King’s College London, through collaborations with Health Data Research UK (HDR UK: https://www.hdruk.ac.uk), the Centre for Translational Informatics (CTI; https://ctiuk.org/) and the NIHR Maudsley Biomedical Research Centre (BRC; https://www.maudsleybrc.nihr.ac.uk/), is at the forefront of clinical NLP and health data science research in the UK. We pride ourselves on the impact of our research, which has led to the deployment of novel informatics applications and platforms to King’s Health Partners (KHP; https://www.kingshealthpartners.org/) and other NHS sites across London and the UK. These applications are beginning to realise the potential of large-scale analytics and reuse afforded by electronic health records, and the promise of data driven healthcare.
This post will be offered on a fixed-term contract for 2 years with a possibility of a 3 year extension
This is a full-time post - 100% full time equivalent
Communication & Networking:
- Work as part of a multidisciplinary team of researchers, data scientists, social scientists, health scientists, mental health clinicians, health economists, and statisticians, and develop good working relationships within these.
- Communicate complex information both orally and in writing to researchers across multiple specialties, and to the public.
Decision Making, Planning and Problem Solving:
- Participation in the strategic planning of projects.
- Responsible for day-to-day decisions in carrying out projects and will be expected to liaise with line managers on major decisions.
Analysis & Research:
- Contribute to the general direction of research.
- Formulate research questions and devise experimental settings in which to test them.
- Develop and implement computationally efficient algorithms to solve research problems using clinical and health related data.
- Analyse research results, reporting and presenting in a variety of settings and media.
Teamwork, Teaching & Learning Support:
- Collaborate with an interdisciplinary team, participate in team meetings, and keep records of all meetings as required.
- Expected to have a strong desire to learn new technologies and skills and will be supported in doing so.
Sensory/Physical Demands & Work Environment:
- Balance the competing pressures of research and deadlines.
- Be aware of health and safety in the work environment.
- All staff members must strictly follow the College’s Health & Safety regulations.
The above list of responsibilities may not be exhaustive, and the post holder will be required to undertake such tasks and responsibilities as may reasonably be expected within the scope and grading of the post.
Skills, knowledge, and experience
1. PhD or equivalent experience in relevant area
2. Relevant Masters degree (STEM field or equivalent) or demonstrable relevant significant experience
3. Relevant BSc (STEM field or equivalent)
4. Knowledge of a broad range of NLP tasks, including information extraction, clustering, classification, sentiment analysis, topic modelling.
5. Knowledge of machine learning and NLP methods including rule-based, supervised and unsupervised learning, ANN and deep learning, language models, distributed representations.
6. Ability to identify cutting edge research areas in NLP and machine learning, and to work with the team and collaborators to propose and prepare research projects for submission to funders.
7. Excellent knowledge of programming, software design patterns and developing applications in Python
8. Experience of building lexicons, vocabularies or terminologies from corpora and knowledge based resources
9. Experience of defining, selecting, preparing, and labelling corpora for NLP development and evaluation, including leading on the definition and creation of annotations.
10. Ability to work closely with a diverse, interdisciplinary team to deliver value to the research community via publications in top tier venues, participation in program committees, funding bids etc.
11. Ability to work with external and internal collaborators to understand, define and solve project related problems
12. Excellent interpersonal & communication skills
13. The ability to work independently and without direct supervision
14. Willingness to learn new methods and technologies
1. Health data and health informatics related skills and knowledge, including EHR structure and standards, medical coding, knowledge and terminology representation, medical decision support.
2. Knowledge of data management, SQL, relational databases, UNIX, scripting languages, common text formats such as JSON, full text indexing and search
3. Experience in building real-world NLP applications, including requirements gathering, development, evaluation and deployment.
4. Experience with NLP, ML and ANN toolkits and frameworks, such as NLTK, spaCy, SciPy ecosystem, Hugging Face, Keras, TensorFlow, PyTorch
5. Experience of NLP and ML at scale e.g. via deployment on cloud platforms, and technologies such as Azure, Docker
6. Experience of collaborating on systematics literature reviews
7. Knowledge of software development best practices, such as agile methods, use of version control, maintainable code, good documentation, etc.
8. An appreciation of the NHS
9. An appreciation of health data ethics and governance, and a regard for data confidentiality and security at all times
10. Willing to travel in the UK and abroad to attend project meetings and conferences
11. Willing to work outside of normal working hours if requested