Grid-based Supervised Clustering Algorithm using Greedy and Gradient Descent Methods to Build Clusters
This paper presents a grid-based supervised clustering algorithm being able to identify clusters of any shapes and sizes without presuming any canonical form for data distribution. The algorithm needs no pre-specified number of clusters and is insensitive to the order of the input data objects. The algorithm gradually partitions data space into equal-size grid cells using one dimension at a time. The greedy method is used to determine the order of dimensions for the gradual partitioning that would give the best quality of clustering, while the gradient descent method is used to find the optimal number of intervals for each partitioning. Finally, any connected grid cells containing data from the same class are merged into a cluster. Using the greedy and gradient descent methods, the algorithm can produce high quality clusters while reduce time to find the best partitioning and avoid the memory confinement problem during the process.
Keywords: Supervised Clustering, Grid-based Clustering, Subspace Clustering, Gradient Descent
Download Full-Text
ABOUT THE AUTHOR
Pornpimol Bungkomkhun
Pornpimol Bungkomkhun is a PhD candidate at the School of Applied Statistics, National Institute of Development Administration, Bangkok, Thailand. She received a Master’s degree in Business Computer Information System from North Texas State University, USA, and completed her Bachelor’s degree in Statistics / Electronic Data Processing from Chulalongkorn University, Bangkok, Thailand.
Pornpimol Bungkomkhun
Pornpimol Bungkomkhun is a PhD candidate at the School of Applied Statistics, National Institute of Development Administration, Bangkok, Thailand. She received a Master’s degree in Business Computer Information System from North Texas State University, USA, and completed her Bachelor’s degree in Statistics / Electronic Data Processing from Chulalongkorn University, Bangkok, Thailand.
