Quartile

Cyrano233Cyrano233 Member Posts: 3 Contributor I
edited June 4 in Help
Good morning
I d like to know about what is visualized with Quartile diagram?
How aMedian and Average and Quartile for BOX PLOT re calculated?
because i get different value if i try to calculate with excel for example...
More over for Graphs is it possible to change asses (and value assis and scale)
TKS
Paqualino

Answers

  • landland RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 2,525   Unicorn
    Dear Mr Ialungo,
    this statistical properties of the data is calculated as usual, but since the plotter use only a sample of the data for performance reasons, the results might differ slightly from the results calculated on the complete data. The default for the sample size is 1000, but you might change in the preferences. Therefore select Preferences from the Tools menu and switch to the gui tab. The property called rapidminer.gui.plotter.rows.maximum determines how many rows of the dataset are used for plotting. You might change this value, but please keep in mind, that a higher value might slow down the rendering process.

    To select, which attribute is shown on one of the axes, use the drop down boxes on the left side of the plotter panel. To change the scale of an axis, right click on the plotter and select properties. Most of the plotters support this feature.

    Greetings,
      Sebastian
  • Cyrano233Cyrano233 Member Posts: 3 Contributor I
    Dear Sebastian,
    I tried to change property called rapidminer.gui.plotter.rows.maximum but i didn't obtain the result as estimated (with excel) in terms of median, Q1 and Q3.
    Also i dont understand why in the Quartile Plotter there are whiskers, and also a vertical line with a point.
    What it represents?
    Also it's impossible to change, and set axsis
    Thanks
    Greetings,
    Pasqualino
  • landland RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 2,525   Unicorn
    Hi,
    the plotter can be interpreted as follows:
    The mean value of the data is shown as the black point, while the median (the lower median on a pair number of examples) is shown as a horizontal line in the box. The vertical line crossing the mean value's point is the standard deviation. The box marks the two center quartiles from 25% to 75%. The whiskers show the 5% and 95% percentiles. Circles beyond the whiskers mark outliers.

    Greetings,
      Sebastian
  • Cyrano233Cyrano233 Member Posts: 3 Contributor I
    Dear Sebastian
    your explanation is clear.
    But netherless I changed rapidminer.gui.plotter.rows.maximum property  i don't obtain the result as estimated (with excel) in terms of median, Q1 and Q3 compared with graphs of rapid miner quartile plotter
    Sorry.
    Maybe can i sent your data in order to verify?
    Let me know
    Greetings
    Pasqualino
  • landland RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 2,525   Unicorn
    Yes,
    please send me your data and your process to load the data. I will verify that.

    Greetings,
      Sebastian
Sign In or Register to comment.