It is crucial for computers to detect metonymic expressions because sentences including them may have different meanings from literal ones. In previous studies, detecting metonymies has been done mainly by rule-based and statistical approaches. The problem of current metonymy detection is that using syntactic and semantic information may be not enough to detect metonymic expressions. In this study, we propose an approach for detecting them with associative information between words. We evaluated our method by comparing it with a baseline that uses syntactic and semantic information. As a result, our method showed significantly better accuracy (0.84) of judging words as metonymic or literal expressions than that of the baseline.