Skip to main content
André Bittar

André Bittar

Research Associate

Biography

André is Research Associate in the Department of Biostatistics & Health Informatics at the Institute of Psychiatry, Psychology and Neuroscience. His area of expertise is computational linguistics and natural language processing (NLP). His research, in both academic and industrial settings, has focussed on a range of topics, including anaphora and coreference resolution, temporal information extraction, relation extraction and semantic role labelling, and sentiment analysis.

Since joining King’s in January 2018 his focus has been on the use of NLP and machine learning to further research in mental health. His work in this area includes the predictive modelling of suicide risk, and the identification of suicidality and self-harm in the large quantities of free-text clinical notes stored in electronic health records.

André is also developing his expertise in the computational analysis of first-hand speech data for mental health research. He is currently investigating the use of multimodal deep learning techniques to detect the emotional content of caregiver speech and what this reveals about family dynamics and the assessment of child mental health outcomes.

André holds a PhD in computational linguistics from Université Paris Diderot.

Research interests

  • Computational linguistics
  • Natural language processing
  • Mental health

    Features

    Words don't come easy: the challenge of identifying perinatal self-harm in healthcare records

    Dr Karyn Ayre discusses a new study investigating the value of a novel way to research self-harm in an extremely vulnerable group – pregnant women and new...

    mother and child hands

      Features

      Words don't come easy: the challenge of identifying perinatal self-harm in healthcare records

      Dr Karyn Ayre discusses a new study investigating the value of a novel way to research self-harm in an extremely vulnerable group – pregnant women and new...

      mother and child hands