Evaluating hierarchical and non-hierarchical grouping for develop a smart system
Marina Adriana Mercioni, Ştefan Holban · 2018
This paper represents a comparative study between Data Mining Techniques with the purpose to determine the architectural styles of buildings using a smart developed system and also the WEKA(Waikato Environment for Knowledge Analysis) tool. This study aims to use the clustering hierarchical algorithms (the agglomerative algorithm and the divisive algorithm) and non-hierarchical (DBSCAN - Density-based spatial clustering of applications with noise), in order to evaluate as precise as possible the system. For this evaluation, we used a system which is based on two hierarchical clustering methods.