RAD: A Radar-Alike Data-Clustering Algorithm for Large Databases

摘要

The popularity of data analysis for business has created a heavy load on demand. Consequently, data mining and data clustering have become significant topics for revealing the implied information recently. This investigation hence presents a novel data clustering algorithm that can be applied to a large database efficiently. The proposed algorithm, called RAD because it is a Radar-alike data-clustering algorithm for large databases, attains this aim because its computation time rises linearly as the data size increases. Experimental results indicate that RAD performs clustering quickly and with fairly good clustering quality and outperforms K-means, DBSCAN and IDBSCAN.