Application of clustering algorithms and classification methods for identifying stench data of automobiles



Chinuk Lee Byeongmin Mun Junseop Lee Hyoseok Kim Hansin Lee Sukjoo Bae


There has been steady increase in consumer’s complaint about the affective quality for automobiles such as the stench inside an automobile. In order to improve the affective quality of smell for consumer, it is crucial to cluster or classify the types of smell at first. We apply several clustering algorithms, such as hierarchical algorithm and K-means algorithm, and classification models, such as decision tree and artificial neural network, and support vector machine, to 26 indices provided from ‘A’ automobile company. After determining optimal number of clusters, we apply three classification methods and compare their performance in terms of computational time and accuracy.

Abstract 11 | PDF Downloads 18




Glenn W. Milligan & Martha C. Cooper. (1985), An examination of procedures for determining the number of clusters in data set. Psychometrika, 50(2), pp159-179, 10.1007/BF02294245
Malay K. Pakhira, Sanghamitra Bandyopadhyay & ujjwal Maulik, (2004), Validity index for crisp and fuzzy clusters. Pattern Recognition, 37(2), pp 487-501, 10.1016/j.patcog.2003.06.005
M. Halkidi, M. Vazirgiannis, & Y. Batistakis, (2002), Quality scheme assessment in the clustering process, Principles of Data Mining and Knowledge Discovery, pp 265-276, 10.1007/3-540-45372-5_26
Pang Ning Tang, Michael Stenbach, Vipin Kumar. (2006)., Introduction To Data Mining, Addison-Wesley Longman Publishing Co., Inc
Regular Session Papers