DDCAL, a novel clustering algorithm developed by Dr. Lux from SWISDATA, addresses the common issue of outliers dominating data visualization in various contexts such as choropleth maps and process models. Traditional methods like k-means or Jenks natural breaks often result in most data points being indistinguishable due to the influence of extreme outliers, which can significantly reduce the analytical value of visualizations.
The following Medium-Article introduces DDCAL in more detail:
DDCAL: A Clustering Algorithm to visualize your Data of Choropleth Maps or Process Models
