Conventional document mining systems mainly use the presence or absence of keywords to mine texts. However, simple word counting and frequency distributions of term appearances do not capture the meaning behind the words, ...