VOCABULARY SIZE ESTIMATION APPARATUS, VOCABULARY SIZE ESTIMATION METHOD, AND PROGRAM

An apparatus uses a rank-by-rank set, extracted from a test word sequence having, as elements, a plurality of test words selected from a plurality of words ranked and a potential vocabulary sequence having, as elements, a plurality of potential vocabulary sizes, of the plurality of test words and the plurality of potential vocabulary sizes, and an answer regarding knowledge of the test words of a user to obtain a model representing a relationship between a value based on a probability that the user answers that the user knows the words and a value based on a vocabulary size of the user when the user answers that the user knows the words. Here, the plurality of test words are ranked to have order based on familiarity within subjects to the test words of the subjects belonging to a subject set. The plurality of potential vocabulary sizes correspond to the plurality of test words, are estimated based on familiarities predetermined for the words, and are ranked to have order based on the familiarities.