Estimating Cluster Population
Advances in Intelligent Systems and Computing
First page number:
Last page number:
Partitioning a given set of points into clusters is a well-known problem in pattern recognition, data mining, and knowledge discovery. One of the widely used methods for identifying clusters in Euclidean space is the K-mean algorithm. In using K-mean clustering algorithm it is necessary to know the value of k (the number of clusters) in advance. We present an efficient algorithm for a good estimation of k for points distributed in two dimensions. The techniques we propose is based on bucketing method in which points are examined on the buckets formed by carefully constructed orthogonal grid embedded on input points. We also present experimental results on the performances of bucketing method and K-mean algorithm. © Springer International Publishing AG 2017.
Clustering; Data partitioning; Bucketing method
Sanjeev, K. C.,
Estimating Cluster Population.
Advances in Intelligent Systems and Computing, 539