This website uses cookies to track visitor behaviour anonymously - this helps us to improve the website.
Read our Privacy Policy for more information.    Accept | Do Not Track
Home » Knowledge » Techniques » Soft Independent Modelling of Class Analogies

Soft Independent Modelling of Class Analogies

Soft Independent Modelling of Class Analogies (SIMCA) is a classification method that uses PCA on a number of calibration data sets and calculates a residual parameter (such as Mahalanobis Distance) for each PCA model in the case of a new measurement. This determines which model the new data is closest to and classifies the measurement accordingly, to a preset threshold. If all the distances calculated are outside the threshold, the new sample is unclassified.