In Chapter 8, we visualize our twenty-six datasets and some of our extensive
experimental results by innovative methods from information visualization and
point out some interesting patterns which may aid future studies. We find that
the majority of examples are misclassified because none of the base classifiers
predict correctly. Although Stacking would potentially be able to learn from
such a setting, this is not observed. On the contrary, Stacking even predicts
incorrectly when a majority of base classifiers is right. Thus we found the field
of information visualization to be a valuable addition and inspiration for our
research and are looking forward to applying its methods to real-life problems
in the near future.