Research Article
Novel Two-Dimensional Visualization Approaches for Multivariate Centroids of Clustering Algorithms
Algorithm 6
The weighted K-means++ and mapping by information gain.
| Input: Number of the centroids, k | | Output: Map with C placed, M | | Begin | | C′: Set of the centroids obtained by the traditional K-means++ clustering in Algorithm 1 | | Ω: Set of the clusters of the instances in I, computed by C′ | | I′: Set of the instances, with a new attribute as the target by filling it with Ω | | IG: Set of the weight values obtained by information gain feature selection method | | C: Set of the centroids obtained by the traditional K-means++ clustering in Algorithm 3 | | fc: The highest ranked feature in IG | | FC: Set of the values in the fcth feature in C | | wc: Sum of the scores in the features except the fcth feature | | WC: Set of the average of the values in the other features | | For i = 1 : k | | FCi = Ci,fc | | For i = 1 : k | | For j = 1 : f | | If j is not equal to fc | | WCi = WCi + Ci,j | | For i = 1 : f | | If j is not equal to fc | | wc = wc + IGi | | For i = 1 : k | | WCi = WCi/wc | | minfc: The minimum value is in FC | | maxfc: The maximum value is in FC | | minwc: The minimum value is in WC | | maxwc: The maximum value is in WC | | For i = 1 : k | | For j = 1 : f | | Ei,j = [Ci,j − minj]c/(maxj − minj) | | Return M where the centroids in E are mapped | | End |
|