k-means has recently been recognized as one of the best algorithms for clustering unsupervised data. Since k-means depends mainly on distance calculation between all data points and the centers, the time cost will be high when the size of the dataset is large (for example more than 500millions of points). We propose a two stage algorithm to reduce the time cost of distance calculation for huge ...

In this work, we study the k-means cost function. The (Euclidean) k-means problem can be described as follows: given a dataset X ⊆ R and a positive integer k, find a set of k centers C ⊆ R such that Φ(C,X) def = ∑ x∈X minc∈C ||x− c|| 2 is minimized. Let ∆k(X) def = minC⊆Rd Φ(C,X) denote the cost of the optimal k-means solution. It is simple to observe that for any dataset X, ∆k(X) decreases as ...

This paper introduces a straightforward generalization of the well-known LVQ1 algorithm for nearest neighbour classifiers that includes the standard LVQ1 and the k-means algorithms as special cases. It is based on a regularizing parameter that monotonically decreases the upper bound of the training classification error towards a minimum. Experiments using 10 real data sets show the utility of t...

In this study, the general ideas surrounding the k-medians problem are discussed. This involves a look into what k-medians attempts to solve and how it goes about doing so. We take a look at why k-medians is used as opposed to its k-means counterpart, specifically how its robustness enables it to be far more resistant to outliers. We then discuss the areas of study that are prevalent in the rea...

Cluster analysis is a useful technique in multivariate statistical analysis. Different types of hierarchical cluster analysis and K-means have been used for data analysis in previous studies. However, the K-means algorithm can be improved using some metaheuristics algorithms. In this study, we propose simulated annealing based algorithm for K-means in the clustering analysis which we refer it a...

We consider the problem of data clustering with unidentified feature quality and when a small amount labelled is provided. An unsupervised sparse method can be employed in order to detect subgroup features necessary for semi-supervised use create constraints enhance solution. In this paper we propose K-Means variant that employs these techniques. show algorithm maintains high performance other ...