摘要：Analysis of the object and the characteristics will be much easier, efficient, and informative when based on a perceptual map, which can display objects and characteristics. Indicator matrix is a matrix where the rows represent objects and the columns is a dummy variable representing characteristics. This article writes about techniques to make perceptual map from indicator matrix, where that can provide information about the similarity between objects, the diversity of each characteristic, correlations between the characteristics, and characteristic values for each object, the techniques we call Hybrid Latent Class Cluster with PCA Biplot, where Latent Class Cluster Analysis is used to transform the indicator matrix to cross section matrix, where rows represent the objects and columns represent the characteristics, the observation cells is the probability of characteristic for each object, next the cross section matrix mapped using Principal Component Analysis Biplot (PCA Biplot). Key Words : Hybrid Latent Class Cluster with PCA Biplot, Latent Class Cluster Analysis, Biplot Principal Component Analysis, Indicator Matrix.
