LGMar 15, 2019
Tackling Initial Centroid of K-Means with Distance Part (DP-KMeans)
arXiv:1903.07977v14 citations
Originality Synthesis-oriented
AI Analysis
This work tackles a specific bottleneck in clustering algorithms for data analysis, but it appears incremental as it builds on existing k-means methods.
The paper addresses the challenge of selecting initial centroids in k-means clustering, which impacts results, particularly as the number of clusters increases, by proposing a method called DP-KMeans.
The initial centroid is a fairly challenging problem in the k-means method because it can affect the clustering results. In addition, choosing the starting centroid of the cluster is not always appropriate, especially, when the number of groups increases.