7

I have several hundreds of values, each corresponding a tuple of features from different dimensions. For example, the value is price, dimensions are:

  • Fruit: apples, oranges, bananas (enum dimension)
  • Origin: US, Argentina, Marocco, ...
  • Month: Feb 2014, Jan 2014, Dec 2013, ...
  • Minimum buying volume: 1 kg, 10 kg, 1000 kg (pseudo-measure)

I'm looking for a cross-platform software for storing such data and easy visualization of data slices: drawing different charts (3D, stacked, area, percentage, bar), selecting several features for axes.

Examples:

  • Line chart with prices of apples for 1 kg from Argentina (month by X axis).
  • Bar chart with prices of oranges, origin by X axis, and grouped by minimum buying volume.
  • 3D chart with prices of bananas, month by X axis, minimum being volume by Y axis, prices averaged by origin.
  • Time would be nice too, with a this month/last month compression, or April for every year of data – Mawg Aug 7 '18 at 11:54
4

I would recommend Pentaho for this.
It is not exactly easy, but not too hard either once you get the concept.

You have many options to insert your data (for instance CSV).
Then you can generate drill into the dimensions you want and generate nice graphs.

You can also define automatically-updated dashboards and reports.

The community edition (CE) is open source (Apache 2 license) and can be found here.
I used it on Linux and Windows, but it also works on Mac OS X.

enter image description here

  • 1
    The free community edition is not easy to find on the web site - you need to look in community.pentaho.com to find it. Also I don't think that the "participation agreement" would meet the requirements of ANY open source licence. – Steve Barnes Feb 19 '14 at 12:41
  • 1
    Wikipedia states that the Community Edition is under Apache 2 License, which means open source. I haven't seen any "participation agreement" when downloading from sourceforge.net/projects/pentaho/files – Nicolas Raoul Feb 20 '14 at 3:27
1

I would strongly recommend taking a look at python + either if your data is really big pyTables & MatPlotLib or for moderate sizes of data Pandas - either way you get a free, cross platform, totally customisable view of your data.

  • There are only hundreds, maximum serveral thousands of values. Pandas looks interesting. I don't need to analyze the data (means, distributions, etc.), only visualize. I have a little experience with Matplotlib and that's the thing I'd like to avoid. Matplotlib is powerful, but not simple. – leventov Feb 16 '14 at 16:26
  • Take a look at the examples at: pandas.pydata.org/pandas-docs/stable/rplot.html – Steve Barnes Feb 16 '14 at 20:59
  • 1
    This all sounds nice and I personally use them for everything, but it basically requires him to write his own software. While the OP might be able to do that, it won't help someone who for example wants to ditch Excel – Ivo Flipse Feb 19 '14 at 10:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.