Issues on the application of classification methods in Engineering

Abstract

Classification methods have received remarkable demand from various engineering fields for solving pattern recognition problems. Various statistical classification methods ranging from parametric, semi-parametric and non-parametric methods have been proposed to meet such demand. Often, researchers are influenced by choosing a popular method rather than fully understand the appropriateness of using a method. Such action will open them towards of having biased results. This paper carries out some overviews on the existing discussion of classification methods used in solving some problems in engineering. The strengths and weaknesses of each method will be discussed in different context of problems that concern with the size of sample, number of measured variables and types of variables. Some suggestions on identifying suitable classification method will be given.