An efficient approach towards clustering using K-means algorithm

Title

An efficient approach towards clustering using K-means algorithm

Creator

Aiyappa S.N.; Ramamurthy B.

Description

Cluster analysis is one of the major knowledge mining methods in the field of data analytics; the approach used for clustering will influence the accuracy of the results and quality of the obtained clusters. A good clustering process or algorithm is one which increases the fit of the data points in each cluster and which satisfies the clustering criteria, if these measures are not met adequately the desired pattern will not be seen and the patterns obtained for analysis may turn out to be inaccurate or insufficient. This paper discusses the standard k-means clustering algorithm and provides an efficient approach towards clustering using the standard global K-means algorithm; the process eliminates the need for initializing random number of clusters multiple times which is followed as the standard process in the field. The effectiveness of the proposed approach was analyzed using the benchmark dataset and the implementation was performed using the well-known analytic tool R Studio and supporting packages. IAEME Publication.

Source

International Journal of Civil Engineering and Technology, Vol-9, No. 2, pp. 705-714.

Date

2018-01-01

Publisher

IAEME Publication

Subject

Distance measure; Elbow method; Gap-statistic method; Hopkins index; Partition clustering; Standard K-means process

Coverage

Aiyappa S.N., Department of Computer Science, CHRIST (Deemed to be University), Bangalore, India; Ramamurthy B., Department of Computer Science, CHRIST (Deemed to be University), Bangalore, India

Rights

Restricted Access

Relation

ISSN: 9766308

Format

Online

Language

English

Type

Article

Identifier

https://www.scopus.com/inward/record.uri?eid=2-s2.0-85042923571&partnerID=40&md5=f7d89e778c57439631dd5457aa75fade