An Improved Clustering Algorithm for Mixed Attributes Data Based on K-prototypes Algorithm

In many situations, the data are often encountered in mixed attributes. The k-prototypes algorithm is one of the principals for clustering this type of data objects. In view of the shortcomings of this algorithm, an improved algorithm is proposed to determine the initial points based on grouping and averaging method. Then we use the actual data set to test the improved algorithm. Detailed data prove that the improved algorithm has good stability and validity.