Cluster Analysis Experiment

Cluster analysis is another important data-mining technique. It can help you create segmentation models, which are crucial for many e-businesses in developing a solid understanding of customer preferences. We tested the Microsoft Clustering algorithm in a manner similar to the Microsoft Decision Trees (MDT) algorithm, with comparable results.

We don't have the space to describe the details of those experiments in this article. For information about the model-training performance of the clustering algorithm, see the white paper "Performance Study of Microsoft Data Mining Algorithms."

By default, 255 is the maximum number of trees a mining model can have.

Performance for non-nested cases is considerably better than for nested cases, although nested cases bring you valuable functionality.