An Evaluative Technique for Categorical Maps

A method is developed that can be used to evaluate categorical maps quantitatively. The method is based on the conditional probability of the occurrence of a cell state, given the states of its neighbors, i.e., to determine the most likely state of a given cell. A zero-one loss function can then be used to calculate the total number of improbable cell states as well as category-by-category counts. Experiments with the technique show its ability to predict accurately the amount of pseudo random noise that is added to a test array of categorical data.