An enhanced K-means genetic algorithms for optimal clustering

Anusha, M.; Sathiaseelan, J. G. R.

doi:10.1109/iccic.2014.7238422

Cited by 10 publications

(4 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…1) C0 question number 5,6,7,10,11,12,13,14,15,16,18,20,22,23,25,26,27,28,29,32,34,35,36, 37 and 39 2) C1 question number 48, 15, 17, 18, 22, 23, 38 and 40 3) C2 question number 1, 2, 3, 4, 9, 19, 21, 24, 30, 31, 33, 41, 42, 43, 44, 45, 46, 47, 48, 49 and 50 Subsequent experiments on Model 2 were also carried out using Fuzzy C-Means which were optimized using Genetic Algorithms. Genetic Algorithm in Fuzzy C-Means clustering is used to optimize the initial determination of the value of the degree of cluster membership to have better results than using the determination of the degree of cluster membership at random.…”

Section: Resultsmentioning

confidence: 99%

“…Meanwhile, fuzzy C-means have advantages in clustering with more than one variable and depend on the degree of membership. So that a very good genetic algorithm is applied to help find the initial cluster center on K-means and determine the degree of initial membership [7]. The K-Means genetic algorithm can produce groupings with a better level of clustering variation than the simple K-Means algorithm as well as the fuzzy C-means algorithm [8].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

K-Means and Fuzzy C-Means Optimization using Genetic Algorithm for Clustering Questions

Sendari

Utama

Putri

et al. 2021

ijasca

View full text Add to dashboard Cite

The grouping of data can be used in the development strategy of an educational game application. The process of grouping data that initially behaved differently into several groups that now behaved more uniformly. As well as grouping the data on the difficulty level of the questions on the educational game question board. This grouping of questions is needed to get the dominant values that will be the characteristics of each group of questions that exist. The clustering method is quite widely used to overcome problems related to data grouping. This clustering is a method of grouping based on the size of the proximity, the more accurate the cluster formed, the clearer the similarity of the difficulty level of the questions. Thus, educational game developers can determine the strategy for placing the existing questions more precisely. Many clustering methods can be used to group the data on this question, including K-Means and Fuzzy C-Means (FCM) which are then optimized using the Algorithm Genetics. From the results of the research conducted, optimization gives better results for clustering questions.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

K-Means and Fuzzy C-Means Optimization using Genetic Algorithm for Clustering Questions

Sendari

Utama

Putri

et al. 2021

ijasca

View full text Add to dashboard Cite

show abstract

“…Anusha et al [18,19] focused on centroid based multi-objective clustering that lacks to reduce the number of clusters. Various feature selection and neighbourhood learning techniques are explained in [20,21] Zihayat et al [22] and a proposed an efficient algorithm THUDS for mining top-k high utility patterns over data streams. Ryang et al [23] proposed the REPT algorithm with four strategies for efficient top-k high utility pattern mining.…”

Section: Literature Surveymentioning

confidence: 99%

Multi-Objective Optimization to Identify High Quality Clusters with Close Referential Point using Evolutionary Clustering Techniques

Anusha

2018

AJCST

Self Cite

View full text Add to dashboard Cite

Most of the real-world optimization problems have multiple objectives to deal with. Satisfying one objective at a time may lead to the huge deviation in other. This paper uses criterion knowledge ranking algorithm solving multi-objective optimization problems. The aim of this research paper is to solve a multi-objective optimization algorithm with close reference point learning method to identify high quality data clusters. A Simple crossover measure is used to quantify the diversity of the whole set, by considering all patterns as a complete entity. In this paper, the task of identifying high quality data clusters using close reference points is proposed to solve multi-objective optimization problem using evolutionary clustering techniques. The proposed algorithm finds the closest feature from the selected features of the data sets that also minimizes the cost while maintains the quality of the solution by producing better convergence. The resultant clusters were analysed and validated using cluster validity indexes. The proposed algorithm is tested with several UCI real-life data sets. The experimental results substantiates that the algorithm is efficient and robust.

show abstract

“…It is noted that the algorithm could produce minimum index value for the maximum datasets. However, there is a need for proper feature selection for better, more optimal solution [14,15]. Ruby et al [16] suggested two methods for ranking of MOPs.…”

Section: Literature Reviewmentioning

confidence: 99%

Multi-Objective Optimization Algorithm to the Analyses of Diabetes Disease Diagnosis

Anusha¹,

Sathiaseelan²

2016

ijacsa

Self Cite

View full text Add to dashboard Cite

Abstract-There is huge amount of data available in health industry which is found difficult in handing, hence mining of data is necessary to innovate the hidden patterns and their relevant features. Recently, many researchers have devoted to the study of using data mining on disease diagnosis. Mining biomedical data is one of the predominant research area where evolutionary algorithms and clustering techniques are emphasized in diabetes disease diagnosis. Therefore, this research focuses on application of evolution clustering multiobjective optimization algorithm (ECMO) to analyze the data of patients suffering from diabetes disease. The main objective of this work is to maximize the prediction accuracy of cluster and computation efficiency along with minimum cost for data clustering. The experimental results prove that this application has attained maximum accuracy for dataset of Pima Indians Diabetes from UCI repository. In this way, by analyzing the three objectives, ECMO could achieve best Pareto fronts.

show abstract

An enhanced K-means genetic algorithms for optimal clustering

Cited by 10 publications

References 15 publications

K-Means and Fuzzy C-Means Optimization using Genetic Algorithm for Clustering Questions

K-Means and Fuzzy C-Means Optimization using Genetic Algorithm for Clustering Questions

Multi-Objective Optimization to Identify High Quality Clusters with Close Referential Point using Evolutionary Clustering Techniques

Multi-Objective Optimization Algorithm to the Analyses of Diabetes Disease Diagnosis

Contact Info

Product

Resources

About