Dr Bevan Koopman

Research Scientist

Contact details:

Biography

For more details please see my homepage at http://www.koopman.id.au

Bevan Koopman is a Research Scientist at the [Australian e-Health Research Centre](http://aehrc.com], CSIRO (Commonwealth Scientific and Industrial Research Organisation).

My research interests are in Information Retrieval (i.e., Search). I am specifically focused on Semantic Search. I apply my research within Health Informatics — how can we improve the access, delivery and quality of health information to the people who need it.

About Me

  • I completed a PhD in Information Retrieval from Queensland University of Technology in 2014.
  • I've worked in research as an engineer/scientist for the University of Queensland and the Distributed Systems Technology Centre (DSTC).
  • I've worked in industry as a software engineer for Mincom Limited and Permeance Technologies.

Research Interests and Current Projects

Semantic Search as Inference:

  • My interest here is in models for semantic search: Information Retrieval (IR) models that elicit the meaning behind the words found in documents and queries rather than simply matching keywords.
  • This is achieved by the integration of structured domain knowledge and data-driven information retrieval methods. A key novel components of such models is that retrieval is driven via an inference mechanism, i.e., to what extent does a information need (query) infer a document, rather than match a document.
  • Finally, an important component of these models are graph-based representations and inference realised as as a traversal over the graph.
  • This was the focus of my PhD thesis: "Semantic Search as Inference: Applications in Health Informatics".

Health Informatics and e-Health:

  • Conducted with the Australian e-Health Research Centre, much of my Information Retrieval research is set within health informatics to tackle the unique challenges within this domain; specifically, how to bridge the ‘semantic gap’; that is, how to overcome the mismatch between raw medical data and the way human beings interpret it.
  • This may be a problem for medical professionals searching patients records, for researchers trying to search for eligible patients for clinical trials or even for the public trying to search for accurate, reliable and readable health information online. Overcoming some of these challenges in health informatics has the potential to have significant impact into an issue that is critical to us all: our health.

Current Roles

  • Research Scientist
    I lead a number of projects applying Information Retrieval (IR), Natural Language Processing (NLP) and Machine Learning (ML) to real-world problems, mainly in the health domain. In IR, I contribute core research in ranking and matching methods, evaluation and user behaviour and understanding. In NLP and ML, I contribute the application of relevant technology to domain specific problems.

Academic Qualifications

  • 2014

    PhD
    Queensland University of Technology

  • 2002

    BInfoTech(Hons I) UQ
    University of Queensland

Publications

  • Google Scholar: http

We have publications by Dr Bevan Koopman