A fast partitioning algorithm using adaptive Mahalanobis clustering with application to seismic zoning

Research project grouping and ranking by using adaptive Mahalanobis clustering

“…. , β n −α n , are mutually significantly different, they should first be normalized [13]. This can be achieved by transforming…”

“…An efficient algorithm for searching for a locally optimal partition with ellipsoidal clusters is the Adaptive Mahalanobis k-means (see [13]), which can be carried out as a special case of the well-known Expectation Maximization algorithm (see [24]), but its efficiency is significantly greater than the standard Expectation Maximization algorithm. The adaptive Mahalanobis k-means algorithm can be described by two steps which are iteratively repeated:…”

“…An efficient incremental partitioning algorithm is proposed in the paper written by [13], which is able to find either a globally optimal partition or a locally optimal partition of the set A ⊂ R n close to the global one. By knowing an optimal r-partition (r ≥ 1), the algorithm searches for the following additional cluster by using the wellknown DIRECT algorithm for global optimization [6,5,7], and after that by using the adaptive Mahalanobis k-means algorithm it determines the optimal (r +1)-partition.…”

“…Most approaches use different multi-criteria decision making methods, most frequently the well-known Analytic Hierarchy Process (AHP) [15]. In our paper, we have combined the AHP method and the adaptive Mahalanobis clustering (AMC) algorithm proposed in [13]. First, the set of projects that have passed the administrative verification was grouped into several clusters depending upon the features used.…”

“…This can be done by using some of the well-known validity indexes (see e.g. [13,23]). In our paper, we will use the Calinski-Harabasz (CH) index and the Davies-Bouldin (DB) index.…”