Validity

The notion of validity of a prediction has an ill-defined status in NLP, and it is not associated with a widely accepted evaluation measure such as precision as a measure of prediction quality, or recall as a measure of prediction quantity, in classification. The goal of this chapter is to give a clear definition of the concept of validity in NLP and data science, which then can be operationalized into methods that allow measuring validity, and applied to general NLP and data science tasks.