作成者 |
|
|
|
本文言語 |
|
発行日 |
|
収録物名 |
|
出版タイプ |
|
アクセス権 |
|
権利関係 |
|
|
|
関連DOI |
|
関連URI |
|
関連情報 |
|
概要 |
This paper describes a method to analyze needs from the remarks of elderly and disabled people by using two text mining systems, which are the concept graph system and matrix search system. First, we ...collected 2,409 remarks about products and services from 19 people. Next, we transformed remarks with meaningful content into an evaluation record with ten aspects, and made 680 evaluation records. The two text mining systems analyze evaluation records in which an input keyword is used in a manner similar to search engines. The concept graph system visualizes a hierarchy of feature words in the records. We can understand important words and their relations in the records before we read the records. The matrix search system displays a matrix of the distribution of clustering from two aspects of the records. We can understand the relation of two aspects in the records by combining various aspects. The two systems achieve multiple and dialogical analysis of evaluation records. Therefore, the systems enable us to examine various problems that we choose to target. In experiments, we made an analysis of the records by combining the results of the two text mining systems.続きを見る
|