Dr Yanhua Xu

Dr Yanhua Xu

HDRC Data Scientist

City of Bradford Metropolitan District Council

I’m a computer scientist specialising in AI applications in health and social care, passionately applying cutting-edge methods to bridge public health and social governance, and enable policy-makers and practitioners to tackle complex challenges more effectively. I focus on leveraging advanced machine-learning techniques, particularly deep-neural networks, to extract actionable insights from complex health datasets.

Currently, I serve as a Data Scientist at the NIHR Health Determinants Research Collaboration (HDRC) in Bradford, hosted by the City of Bradford Metropolitan District Council. In this role, I design, develop and deploy machine-learning solutions for community health assessment, social decision support and risk prediction.

I hold an MSc in AI from the University of Southampton, where I worked under Dr Nicolas Green, and a PhD in AI (Digital Health) from the University of Liverpool, supervised by Prof. Dominik Wojtczak.

Previously, I worked as a Postdoctoral Research Fellow in Urban Data Science at the University of Leeds, contributing to the ERC-funded Data Assimilation for Agent-Based Models (DUST) project under Prof. Nick Malleson.

(edited on Oct 2025)

Interests
  • Deep Learning
  • Natural Language Processing
  • Reinforcement Learning (ongoing)
Education
  • PhD in Artificial Intelligence, 2024

    University of Liverpool

  • MSc in Artificial Intelligence, 2019

    University of Southampton

Recent
Publications

You can also visit my Google Scholar profile. For the most recent versions of my work, please refer to my publications on arXiv or the attached PDF on this website.

Recent
Projects

*
Insight Extraction from Social Work Reflection
Extracting key themes and emotional signals from adult social workers’ reflective comments to identify practice issues and prioritise support for those most in need.
Insight Extraction from Social Work Reflection
Look After Children (LAC)
Understanding and Supporting Looked After Children (LAC) Through Data-Driven Insights.
Look After Children (LAC)
Not in Employment, Education or Training (NEET)
Predicting and Supporting Young People at Risk of NEET.
Not in Employment, Education or Training (NEET)
Footfall Time Series Clustering
Applied unsupervised time-series clustering to identify recurring pedestrian footfall patterns in city-centre data. The project forms part of the DUST research (Data Assimilation for Agent-Based Models).
Footfall Time Series Clustering