METHOD FOR SORTING OUT KEYWORD SHOWING DIFFERENCE BETWEEN TWO GROUPS AND SYSTEM FOR EXECUTING THE METHOD

摘要

A method and a system for selecting keywords showing a difference between two groups are provided to easily estimate the difference or interest of two groups by generating an output set sorted by a keyword score of two groups, as the keywords are selected by assigning the keyword score to each keyword showing the difference. A data set definer(510) defines a tuple of the keywords and a data set including a tuple set. A keyword score generator(520) generates the keyword score of the keyword by using the data set. An output set generator(530) generates the output set including a keyword set by generating the keyword set related to the keyword score. The data set definer includes a tuple definer(511) defining the tuple including the keyword set and a repeater(512) repeating the data set including the tuple as an element by repeatedly performing the tuple definer for the keywords. The keyword score generator includes a probability set estimator estimating a probability set based on the data set and a generator generating the keyword score of the keyword through the probability set.