A Survey of Distance Metrics in Clustering Data Mining Techniques

Abstract

Lately, due to the increasing size of databases, several aspects have been studied in detail, such as grouping, searching for the closest neighbor and other identification methods. It has been found that in the multidimensional space, the concept of distance does not offer high performance. In this paper, we study the effect of different types of distances on the group to see the similarities between objects. Among these distances we mention two distances: the Euclidean distance and Manhattan distance, implemented in a system developed to identify the architectural styles of the buildings. The aim of this paper is using cluster analysis to identify distance metrics impact in detection of architectural styles using Data Mining techniques.