An Efficient DBSCAN with Enhanced Agglomerative Clustering Algorithm

One of the most important problems in a large-scale distributed database is the detection of point outliers. Many studies have used clustering to address these issues, and DBSCAN appears to be the most effective method. The paper concentrated on reducing the high temporal complexity of the problem by reducing the data dimensionality of the data points. Due to these issues, previous techniques are not adequate for minimizing inter-and intra-cluster variation on real-world datasets that call for quick responses. In order to tackle this issue without requiring more time complexity or memory, the paper introduces a DBSCAN with Enhanced Agglomerative Clustering (DBSCAN-EAC) technique. This algorithm uses a concurrent execution of Reverse Nearest Neighbor graph traversals using density-based clustering.