A new efficient density-based data clustering technique using cross expansion for data mining

Abstract

This investigation develops a new data clustering technique. It is a new density-based clustering scheme by diagonal sampling and a new method of fold and rotation for enhancing data clustering performance. The proposed algorithm's expansion without selecting data points to increase computation cost and it may considerably lower time cost The experimental results confirm that the presented approach has fairly high clustering accuracy and noise filtering rate, and is faster than numerous well-known existing density-based data clustering algorithms such as DBSCAN, IDBSCAN, KIDBSCAN and FDBSCAN approaches.