TY  - JOUR
T1  - Presentation and Evaluation of an Effective Algorithm for Clustering
AU - Alesheykh, R. 
JO  - Journal of Engineering and Applied Sciences
VL  - 12
IS  - 7
SP  - 1890
EP  - 1895
PY  - 2017
DA  - 2001/08/19
SN  - 1816-949x
DO  - jeasci.2017.1890.1895
UR  - https://makhillpublications.co/view-article.php?doi=jeasci.2017.1890.1895
KW  - Machine learning
KW  -C4.5 decision tree
KW  -RIPPER rule learning
KW  -bayesian networks
KW  -Winger-Ville distribution
AB  - Clustering algorithms partition a set of data into numbers of groups according to their similarity. A clustering algorithm is a common technique for statistical data analysis and used in many fields including information retrieval and machine learning. In the current research some machine learning algorithms have been presented to identify three timber species and group them into the correct clusters. Machine learning algorithms such as C4.5 decision tree, RIPPER rule learning method and bayesian network have been experimented across Winger-Ville distribution method to do the identification task. The employment of the most suitable timber for each specific purpose demands for the development of an effective computerized method for the identification of timber species. Since, each species creates different properties in timber, a reliable and powerful evaluation approach of identification plays an important role in suitably use of the timber. The final analysis shows the clustering performance of 91% when the output of Winger-Ville distribution method is employed by RIPPER rule learning algorithm.
ER  -