So I want to visualize how each characteristic (let's say we have 100) relates to result, for example drawing Bars and/or Pies aggregated by result and sub-aggregated by the characteristic, etc.
Do I need to reproduce each graph 100 times one for each characteristic or there is a better approach?
I don't believe you can create a graph like this in an automated way, but you could certainly plot each characteristic on the same graph by adding multiple "metrics" or "buckets" to a single graph, one for each characteristic.
Could you give more information on exactly the result you're looking for? For example, for a bar graph, what are you trying to use for the X and Y axis?
We have an algorithm that assigns a rating (A, B, C, ...) to a lead (in marketing terms). We want to visualize the distribution of ratings for each characteristic (Age, Gender, Town, O.S., Browser, Phone model, channel, etc).
For a bar, the Y axis would be the count of leads, and the X axis could be the rating aggregated by characteristic (or the characteristic aggregated by rating).
A pie could also be useful, for example
In this case, the inner data is the device (the characteristic) and the outer circle is the rating (result).
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.