The Research of Qualitative Data Hierarchical Cluster Based on Dummy Variable

Existing data clustering algorithms measure similarity between objects by using a distance metrics that is defined mainly for quantitative data. According to this, the idea of dummy variables in statistics is used to transform the qualitative data to measure swatch matrix shown by binary scale. Relativities between variables is eliminated in order to reduce complexity in analysis. Then the dissimilarity and distances among the variables are computed. Using the method introduced, an experiment is carried out, and its outcome shows that this method is simple and effective