KIDBSCAN: A New Efficient Data Clustering Algorithm

摘要: Spatial data clustering plays an important role in numerous fields. Data algorithms have been developed recent years. K-means is fast, easily implemented and finds most local optima. IDBSCAN more efficient than DBSCAN. can also find arbitrary shapes detect noisy points for clustering. This investigation presents a new technique based on the concept of IDBSCAN, which used to high-density center then expand clusters from these points. has lower execution time because it reduces by selecting representative seeds. The simulation indicates that proposed KIDBSCAN yields accurate results. Additionally, this approach I/O cost. outperforms DBSCAN IDBSCAN.