Efficiency issues of evolutionary k-means

Naldi, Murilo Coelho; Campello, Ricardo J. G. B.; Hruschka, Eduardo R.; Carvalho, André C. P. L. F. de

doi:10.1016/j.asoc.2010.06.010

Cited by 104 publications

(60 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A common practice consists of assuming k max = √ N [22], [23], which will be used here. Previous experiences and studies also suggest that the usage of n p = 10 or n p = 20 to be enough to find partitions of reasonable quality for a wide variety of data sets [15], [24] and, therefore, these values will be adopted here. Algorithm 4 presents an example of this method, as implemented here, in which π a is the current partition and k max is the maximum number of clusters permitted for the resulting partitions.…”

Section: Systematic (Repetitive) Methodsmentioning

confidence: 99%

“…In [15], a new codification and changes in F-EAC data structures reduced the computational time considerably in relation to the original version of the algorithm, published in [14]. In the present paper, an experimental methodology is used to compare and estimate this algorithm with others also based on k-means and capable of estimating the number of clusters k, known as: ordered and random systematic executions of k-means; variations of Bisecting k-means, variations of X-means; and variations of F-EAC that incur in stochastic local search procedures.…”

Section: Introductionmentioning

confidence: 96%

See 1 more Smart Citation

Comparison Among Methods for k Estimation in k-means

Naldi

Fontana

Campello

2009

2009 Ninth International Conference on Intelligent Systems Design and Applications

View full text Add to dashboard Cite

One of the most influential algorithms in data mining, k-means, is broadly used in practical tasks for its simplicity, computational efficiency and effectiveness in high dimensional problems. However, k-means has two major drawbacks, which are the need to choose the number of clusters, k, and the sensibility to the initial prototypes' position. In this work, systematic, evolutionary and order heuristics used to suppress these drawbacks are compared. 27 variants of 4 algorithmic approaches are used to partition 324 synthetic data sets and the obtained results are compared.

show abstract

Section: Systematic (Repetitive) Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 96%

Comparison Among Methods for k Estimation in k-means

Naldi

Fontana

Campello

2009

2009 Ninth International Conference on Intelligent Systems Design and Applications

View full text Add to dashboard Cite

show abstract

“…Then, transform the result by combining with Hierarchical algorith m in order to determine the initial centroids for K-means. The execution steps of the proposed Hierarchical K-means algorith m to determine initial centroids for K-means are described as follows: Method 5: [18] Efficiency issues of evolutionary Kmeans method suggests that evolutionary techniques conceived to guide the application of K-means can be more co mputationally efficient than systematic (i.e., repetitive) approaches that try to get around the Kmeans drawbacks by repeatedly running the algorithm fro m different configurations for the number of clusters and initial positions of prototypes. To do so, a modified version of a (K-means-based) fast evolutionary algorithm for clustering is employed.…”

Section: Methodsmentioning

confidence: 99%

DIMK-means “Distance-based Initialization Method for K-means Clustering Algorithm”

Aldahdooh¹,

Ashour²

2013

IJISA

View full text Add to dashboard Cite

Abstract-Partition-based clustering technique is one of several clustering techniques that attempt to directly decompose the dataset into a set of disjoint clusters. Kmeans algorith m dependence on partition-based clustering technique is popular and widely used and applied to a variety of domains. K-means clustering results are extremely sensitive to the initial centroid; this is one of the major drawbacks of k-means algorithm. Due to such sensitivity; several d ifferent init ialization approaches were proposed for the K-means algorith m in the last decades. This paper proposes a selection method for init ial cluster centroid in K-means clustering instead of the random selection method. Research provides a detailed performance assessment of the proposed initialization method over many datasets with different dimensions, numbers of observations, groups and clustering complexit ies. Ability to identify the true clusters is the performance evaluation standard in this research. The experimental results show that the proposed initialization method is more effective and converges to mo re accurate clustering results than those of the random initialization method.

show abstract

“…To analyze the differences among EFTs, we examined the functional variation for each type according to independent variables such as the Gross Primary Production (GPP) [88] and the rate of Evapotranspiration (ETP). Seasonal differences among EFTs are produced ( Figure 9).…”

Section: Ecosystem Functional Typesmentioning

confidence: 99%

Identification of Ecosystem Functional Types from Coarse Resolution Imagery Using a Self-Organizing Map Approach: A Case Study for Spain

et al. 2014

View full text Add to dashboard Cite

Ecosystem state can be characterized by a set of attributes that are related to the ecosystem functionality, which is a relevant issue in understanding the quality and quantity of ecosystem services and goods, adaptive capacity and resilience to perturbations. This study proposes a major identification of Ecosystem Functional Types (EFTs) in Spain to characterize the patterns of ecosystem functional diversity and status, from several functional attributes as the Normalized Difference Vegetation Index (NDVI), Land Surface Temperature (LST) and Albedo. For this purpose, several metrics, related to the spatial variability in seasonal and annual patterns (e.g., relative range), have been derived from remote sensing time series of 1 km MODIS over the period 2000-2009. Moreover, precipitation maps from data provided by the AEMet (Agencia Estatal de Meteorología) and the corresponding aridity and humidity indices were also included in the analysis. To create the EFTs, the potential of the joint use of Kohonen's Self-Organizing Map (SOM) and the k-means clustering algorithm was tested. The EFTs were analyzed using different remote sensing (i.e., Gross Primary Production) and climatic variables. The relationship of the EFTs with existing land cover datasets and climatic data were analyzed through a correspondence analysis (CA

show abstract

Efficiency issues of evolutionary k-means

Cited by 104 publications

References 27 publications

Comparison Among Methods for k Estimation in k-means

Comparison Among Methods for k Estimation in k-means

DIMK-means “Distance-based Initialization Method for K-means Clustering Algorithm”

Identification of Ecosystem Functional Types from Coarse Resolution Imagery Using a Self-Organizing Map Approach: A Case Study for Spain

Contact Info

Product

Resources

About