A Comparative Study of Bootstrap Application to Subjective Clustering
In order to develop successful products in today's competitive market, it is essential to analyze customer needs accurately. Since the number of customer needs collected using marketing survey tends to be large, it is necessary to group similar needs and identify a small number of representative needs. Subjective Clustering is a statistical method that uses Hierarchical Clustering algorithm to group similar customer needs into clusters. This paper compares Subjective Clustering (SC) and Bootstrap applied to Subjective Clustering (BS-SC) for determining how accurately these methods estimate population primary clusters. In order to investigate inherent characteristics in the two methods, this paper proposes a simulation-based comparative study. The result suggests that BS-SC estimates population primary cluster more accurately than SC when the sample size is small relative to population size; however, as sample size increases, SC estimate is more accurate than BS-SC.
S. Takai and N. Bhardwaj, "A Comparative Study of Bootstrap Application to Subjective Clustering," Proceedings of DETC/CIE 2006 ASME 2006 Intl Design Engineering & Computers and Information in Engineering Conferences September 2006, Philadelphia, Pennsylvania USA (DETC2006), American Society of Mechanical Engineers (ASME), Jan 2006.
Mechanical and Aerospace Engineering
Keywords and Phrases
Bootstrap; Comparative Study; Customer Needs; Hierarchical Clustering; Subjective Clustering
Article - Conference proceedings
© 2006 American Society of Mechanical Engineers (ASME), All rights reserved.