I am interested in the implications of data and software quality for scientific research. My background is primarily in search-based software engineering. However, more recently I have worked on geospatial processing and analysis for epidemiological modelling. Scientific software often includes undocumented assumptions; it is unclear why these assumptions were made or how they impact the results. This increases the risk of undetected errors that could compromise inferences made from the models. Scientific software is difficult to test because it is exploratory in nature, so it is not always clear what the correct outputs should be. There are also challenges due to the use of big data, stochastic processes and mathematical assumptions. I am looking for techniques that will help scientists ensure the software they produce is correct.
Complex Systems ; Search-based Optimisation ; Spatial Panel Analysis ; Software Testing