Concept Testing of Some Visualization Methods for Geographic Metadata

Concepts of metadata visualization have been developed in order to help users in understanding of metadata about geographic datasets. These concepts apply graphic samples and four multivariate visualization methods: scatter plot matrix, parallel coordinates plot, star plots and Chernoff faces. The concepts were tested with users of geographic metadata in the Finnish Defence Forces. The test results indicate that users benefit from graphic samples especially when comparing alternative datasets. When the differences of the datasets are visible users start to question their presumptions about the data. The parallel coordinates plot and star plots support users in comparison of several metadata elements of several datasets simultaneously whereas the scatter plot matrix is difficult method to adopt and does not support a holistic view of datasets. Chernoff faces arouse emotions that may disturb the interpretation of metadata; interpretation of extreme values from the faces is reliable but i...

• Type: Conference
• Year: 2003
• Where: SCANGIS
• Authors: Paula Ahonen-Rainio
• Keywords: Information Technology, Metadata, Parallel Coordinates Plot, SCANGIS 2003, Scatter Plot Matrix