Comfort Factor and Cluster Analysis with Practice Application

Abstract

When dealing with high-dimensional multivariate data, we often use principal component analysis (PCA) to reduce dimensionality.

The aim of this research is to Comfort (PCA) to identify the most effect variables on the studied ,The observations of the compounds represent the ratio of the effect of each variable in the factor, and then we apply the method K-means Clustering of averages to the new variables (factors) to find out what observation (the ratio of effect for each variable in the factor) closest to each The method of Comforting used to obtain more accurate results , Group similar variables, eliminate classification errors in the case of large data by the number of variables and observations and reduce the time of calculation of the results.