Title: Unsupervised clustering of materials properties using hierarchical techniques

Authors: Arafa S. Sobh; Sameh A. Salem; Rania Darwish; Mohammed Hussein; Omar Karam

Addresses: Mechanical Engineering Department, Helwan University, Al Sikka Al Hadid Al Gharbeya, Qism Helwan, Cairo Governorate, Egypt ' Electronics, Communications and Computers Department, Helwan University, Al Sikka Al Hadid Al Gharbeya, Qism Helwan, Cairo Governorate, Egypt ' Mechanical Engineering Department, Helwan University (On leave to BUE), Al Sikka Al Hadid Al Gharbeya, Qism Helwan, Cairo Governorate, Egypt ' Mechanical Engineering Department, Helwan University (On leave to BUE), Al Sikka Al Hadid Al Gharbeya, Qism Helwan, Cairo Governorate, Egypt ' Faculty of Informatics and Computer Science, Computer Science Department, British University in Egypt (BUE), Egypt

Abstract: Data mining (DM) algorithms arose as a promising and flourishing discipline at manufacturing and industrial engineering. This paper proposes an efficient decision support approach for manufacturing engineering. The proposed approach tackles clustering challenges for engineering materials properties. It adopts the hierarchal clustering for mining engineering materials properties. Extensive experiments and comparisons are conducted on three different real-world datasets for engineering materials properties. In addition, a study of different similarity measures is carried out to choose the best fit similarity measure to engineering material datasets. A comparison of the results with other competitors clearly shows the robustness of the proposed approach. Therefore, it is highly recommended to use the proposed approach as a scalable engineering material properties tool.

Keywords: decision support systems; DSS; manufacturing systems; linkage hierarchical algorithm; materials properties; unsupervised clustering; data mining; engineering materials.

DOI: 10.1504/IJCENT.2015.073182

International Journal of Collaborative Enterprise, 2015 Vol.5 No.1/2, pp.74 - 88

Received: 04 Jun 2015
Accepted: 11 Jul 2015
Published online: 26 Nov 2015 *