2015 Volume 71 Issue 6 Pages II_13-II_21
In this study, the authors tried to examine the utility and to clarify the problems of text mining for identifying substantive issues and visualizing relationship among the issues in large-scale public comments. As the result of case study with submitted comments of the “public comment for choices on energy and the environment” implemented by National Policy Unit in 2012, it was possible to illustrate the utility of text mining by identifying 17 issues such as “shift of energy policy”, “disposal of radioactive waste”, and so forth, as well as to visualize relationship among the issues by drawing network graphs. It was also possible to clarify the problems such as the existence of a large amount of comments with similar contents, and necessity of developing an objective word selection method for identifying the issues being mentioned less frequently.