Sample Correction
If a given class and feature value never occur together in the training set then the frequency-based probability estimate will be zero. This is problematic since it will wipe out all information in the other probabilities when they are multiplied. It is therefore often desirable to incorporate a small-sample correction in all probability estimates such that no probability is ever set to be exactly zero.
Read more about this topic: Naive Bayes Classifier
Famous quotes containing the words sample and/or correction:
“The present war having so long cut off all communication with Great-Britain, we are not able to make a fair estimate of the state of science in that country. The spirit in which she wages war is the only sample before our eyes, and that does not seem the legitimate offspring either of science or of civilization.”
—Thomas Jefferson (1743–1826)
“Shakespeare, with an improved education and in a more enlightened age, might easily have attained the purity and correction of Racine; but nothing leads one to suppose that Racine in a barbarous age would have attained the grandeur, force and nature of Shakespeare.”
—Horace Walpole (1717–1797)