OBJECT ORIENTED AGGLOMERATIVE HIERARCHICAL CLUSTERING MODEL IN DATA MINING


YEŞİLBUDAK M., KAHRAMAN H. T., KARACAN H.

JOURNAL OF THE FACULTY OF ENGINEERING AND ARCHITECTURE OF GAZI UNIVERSITY, cilt.26, sa.1, ss.27-39, 2011 (SCI-Expanded) identifier identifier

Özet

Many techniques that have different intended uses are available in data mining. One of the techniques used frequently in recent years is the agglomerative hierarchical clustering analysis. In this study, an object oriented agglomerative hierarchical clustering model has been developed. Users have used the system interface easily through the interactivity of the model developed and the user requests have been accomplished quickly through the object oriented programming. Thus, the model developed has had optimum performance. In the test process of the model, monthly average wind speed data belonging to 44 cities in Turkey are clustered hierarchically. The clustering results have not only been presented in textual format in the interface created using Microsoft Visual Studio. NET platform but also have been visualized using dendrograms in Matlab. Many inferences which remained hidden previously but uncovered by the model are achieved and the model analysis has been done efficiently.