other,8-1-A97-1042,ak This paper addresses the problem of identifying likely <term> topics </term> of <term> texts </term> by their position in the <term> text </term> .
hide detail