Paul McNicholas

My research group develops computationally intensive statistical approaches for detecting previously unknown groups of similar observations in big, or otherwise complex, data. The process of detecting previously unknown subgroups within data is known as clustering, or cluster analysis, and the approaches we develop are typically based on statistical models. Considering the proliferation of data today, our work is useful for data from many domains. To facilitate the dissemination of our work, we like to release open source software whenever practical.