spenke-beilken - Contribution to the ,Discovery Challenge"...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
1 Visual, Interactive Data Mining with InfoZoom – the Financial Data Set Michael Spenke, Christian Beilken GMD – German National Research Center for Information Technology FIT – Institute for Applied Information Technology, http://www.gmd.de/fit Schloss Birlinghoven, D-53754 Sankt Augustin [email protected], http://fit.gmd.de/hci/pages/michael.spenke.html Abstract This paper describes the application of the data analysis tool InfoZoom to the Financial Data Set for the PKKD’99 Discovery Challenge. No automatic method for data mining is used. Instead, InfoZoom enables the user to interactively explore different visualizations of the data. In this way, the user gets a feeling of the data, detects interesting knowledge, and gains a deep understanding of the data set. Introduction InfoZoom has been developed at GMD – the German National Re- search Center for Information Tech- nology . It is now marketed by the GMD spin-off company humanIT . InfoZoom displays database relations as tables with attributes as rows and objects as columns. In Figure 1 each column corresponds to a bank account with a granted loan. The attributes are hierarchically ordered like files in a directory. The menu left of each attribute name shows the possible values and their frequency. Selecting a value from the menu restricts the table to the objects with this value. Clicking on the arrow outline right of an attribute sorts the table by this attribute. A table can be compressed by click- ing the second of the three mode but- tons in the upper left corner of the table (above „682 of 682 objects“). In Compressed Mode , the column width is reduced until all the objects Figure 1: Wide Table Mode Figure 2: Compressed Table Mode Contribution to the „Discovery Challenge” at the 3rd European Conference on Principles and Practice of Knowledge Discovery in Databases, PKDD ‘99, September 15-18, 1999, Prague, Czech Republic
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
2 fit on the screen. In Figure 2 the column width is about one pixel. In large tables the column width will be even smaller. Some techniques make the table readable in spite of this compression. The most important is that neigh- bouring cells with identical values are combined into one larger cell. Because the table is sorted by the attribute duration , all its values are readable. The width of each cell indicates the number of subsequent objects with this value. If a cell is to small to display a numeric value, a short horizontal line still indicates its relative height. In this way, it can be seen that loans with a short duration typically also have a small amount . Instead of selecting a value from the menu, an attribute can also be restricted by selecting and double-clicking a value or value-range direct- ly in the table. In a short animation, the clicked cells grow while the others shrink. This looks like
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 02/13/2012 for the course CS 91.510 taught by Professor Staff during the Fall '09 term at UMass Lowell.

Page1 / 6

spenke-beilken - Contribution to the ,Discovery Challenge"...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online