View article

To evaluate the quality of a treemap algorithm, two quality metrics should be considered. The first quality metric is the aspect ratio of the treemaps generated by the algorithm. The aspect ratio determines how accurate the areas of the rectangles in the treemap can be interpreted by the user. The second quality metric, which is the main focus of this thesis, is the stability score. The stability score indicates how hard it is to track rectangles when the treemap has changed. The aim of this thesis is twofold. The first aim is to evaluate how the stability score can be objectively determined. The second aim is to develop a treemap algorithm that is able to generate treemap that have both low stability scores and low aspect ratios.

To objectively determine stability score we first evaluated the existing definitions of stability. We noticed that none of the current definitions take the structures in the treemap into account. We then developed a new definition for the stability score, that does take these structure into account. The new definition is based on the change in the relative positions of the rectangles with regard to each other.