動的に変化するコールセンター情報からの重要情報の特徴発見(<特集>「アクティブマイニング」及び一般)

概要

論文の詳細を見る
This paper reports the results of our experimental study on a new method of applying an association rule miner to discover useful information from inquiry database. It has been claimed that association rule mining is not suited for text mining. To overcome this problem, we propose (1) to generate sequential data set of words with dependency structure from the text database, and (2) to employ a new method for extracting meaningful association rules by applying a new rule selection criterion based on difference between prior and posterior confidences, instead of minimum confidence. This criterion comes from the fact that we put heavier weights to those phenomena with co-occurrence of plural items more than those with single occurrence. Using this method, we succeeded in extracting useful information from the text database, which were not acquired by only simple keywords retrieval.
社団法人電子情報通信学会の論文
2003-03-06