Issue No.01 - January/February (2010 vol.30)
Joseph MacInnes , Oculus Info
William Wright , Oculus Info
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MCG.2010.18
Humans use intuition and experience to classify everything they perceive, but only if the distinguishing patterns are visible. Machine-learning algorithms can learn class information from data sets, but the created classes' meaning isn't always clear. A proposed mixed-initiative approach combines intuitive visualizations with machine learning to tap into the strengths of human and machine classification. The use of visualizations in an expert-guided clustering technique allows the display of complex data sets in a way that allows human input into machine clustering. Test participants successfully employed this technique to classify analytic activities using behavioral observations of a creative-analysis task. The results demonstrate how visualization of the machine-learned classification can help users create more robust and intuitive categories. [Erratum: The print version contains an error that has been corrected in the online version. The first sentence of the "Method" section (p. 9) contained the URL www.nspace.com. This URL is not related to the nSpace analytic software environment discussed in the article. Instead of the URL, reference 2 should have been cited (P. Proulx et al., "nSpace and GeoTime: A VAST 2006 Case Study," IEEE Computer Graphics and Applications, vol. 27, no. 5, 2007, pp. 46-56). We apologize for these errors.]
visualization, mixed-initiative interfaces, machine learning, classification, workflow modeling, computer graphics, graphics and multimedia.
Joseph MacInnes, William Wright, "Visual Classification: Expert Knowledge Guides Machine Learning", IEEE Computer Graphics and Applications, vol.30, no. 1, pp. 8-14, January/February 2010, doi:10.1109/MCG.2010.18