Efficiency of Random Sampling Based Data Size Reduction on Computing Time and Validity of Clustering in Data Mining

Cebeci, Zeynel and Yildiz, Figen (2016) Efficiency of Random Sampling Based Data Size Reduction on Computing Time and Validity of Clustering in Data Mining. AGRÁRINFORMATIKA / JOURNAL OF AGRICULTURAL INFORMATICS, 7 (1). pp. 53-64. ISSN 2061-862X

Preview

Text 266_1161_1_PB_u.pdf - Published Version Available under License Creative Commons Attribution. Download (1MB) | Preview

Abstract

In data mining, cluster analysis is one of the widely used analytics to discover existing groups in datasets. However, the traditional clustering algorithmsbecome insufficient for the analysis ofbig data which havebeen formed with the enormous increase in the amount of collected data in recent years. Therefore, the scalability has been one of the most intensively studied research topics for clustering big data. The parallel clustering algorithms and the Map-Reduce framework based techniques on multiple machines are getting popular in scalability for big data analysis. However, applying the sampling techniques onbigdatasets could be still alternative or complementary taskin order to run the traditional algorithms on single machines. The results obtained in this studyshowed that the data size reduction by the simple random sampling could be successfully usedin cluster analysisfor large datasets. The clustering validitiesby running K-means algorithmon the sample datasetswerefound as highas those of the complete datasets. Additionally the required execution time for cluster analysis onthe sample datasets wassignificantly shorter thanthose obtained for thecomplete datasets.

Actions (login required)

Edit Item