The abstract class for all outlier detection clusterers. Cannot be instantiated. Must be inherited. An implementation is available in package streamMOA.
clean_outliers(x, ...)A method that requires removal of all outliers from the addressed outlier detection clusterer.
get_outlier_positions(x, ...)Returns spatial positions of all current outliers.
recheck_outlier(x, outlier_correlated_id, ...)Re-checks the outlier having outlier_correlated_id. If this object is still an outlier, the method returns TRUE.
noutliers(x, ...)Returns the current number of outliers.
print(x, ...)Prints out the general info and statistics about the outlier detection clusterer.
get_assignment(x, points, type=c("auto", "micro", "macro"),
method=c("auto", "nn", "model"), outlier_threshold=0.05, ...)Calculates assignments and outlier marks for the points data frame. Points are assigned to an outlier only if they are inside outlier_threshold distance from its position. This implementation is heavily dependant on the Euclidean distance measure and should be overwritten by concrete outlier detection clusterer implementations.