Using Multiple Views for Data Analysis
When a data analyst meets a complex dataset, graphics displays giving overall summaries are examined first, then more specific displays that highlight observed features are studied. Frequently, this involves selection of subsets, and point-and click-methods are intuitive and effective. Sometimes the...
Main Authors: | , , , |
---|---|
Other Authors: | |
Format: | Text |
Language: | English |
Published: |
1992
|
Subjects: | |
Online Access: | http://www.dtic.mil/docs/citations/ADP007184 http://oai.dtic.mil/oai/oai?&verb=getRecord&metadataPrefix=html&identifier=ADP007184 |
Summary: | When a data analyst meets a complex dataset, graphics displays giving overall summaries are examined first, then more specific displays that highlight observed features are studied. Frequently, this involves selection of subsets, and point-and click-methods are intuitive and effective. Sometimes the observed features are investigated by altering details of the analysis, and then an interactive command interface (like S) can be more useful. A rainfall dataset with geographic and time components is used as an example. Graphics displays are done in a modified version of S that permits multiple graphics windows, and this is compared with xlispstat, xgobi, and datadesk. This article is from 'Computing Science and Statistics: Proceedings of the Symposium on Interface Critical Applications of Scientific Computing (23rd): Biology, Engineering, Medicine, Speech Held in Seattle, Washington on 21-24 April 1991,' AD-A252 938, p435-440. |
---|