Visual Data Mining Using a Constellation Graph

スポンサーリンク

概要

• 論文の詳細を見る

• Software was developed to enable visual data mining of multivariate data using a constellation graph. Two applications of the new software are presented. A constellation graph is an effective way to display multivariate data on to a two dimensional plane. The value of the classification variable used just as it is or it classifies it into some levels. Then, the cluster on each level is made to be able to be identified using the color or the symbol. The operator can change the weight of the variable interactively and do the visual mining by seeing the change of the star on the graph. Moreover, our program computes the weight of the variable to separate each cluster. It is possible to presume the data, which cluster it belongs to. In addition, using a computer mouse, the operator can select an area of the graph to examine data in detail.

• 関西学院大学の論文
• 2001-12-22

著者

• Fujikawa Kenji

  Web System Support Center Hitachi Systems And Services Ltd.

• Tanaka Kazuyoshi

  Web System Support Center Hitachi Systems And Services Ltd.

• NIWA Tokihiko

  Kwansei Gakuin High School

• OYAMA Mayumi

  Kwansei Gakuin University, Center for Information & Media Studies

• Oyama Mayumi

  Kwansei Gakuin University Center For Information & Media Studies

関連論文

• Visual Data Mining Using a Constellation Graph

スポンサーリンク