Real-time recommender performance evaluation

[GoogleCodeExporter]

It would be of great benefit to display some measure of the recommender's 
effectiveness in real-time. As each new datum is ingested, it's possible to 
evaluate how good a recommendation the engine thought it was before it was 
ingested. Real data ought to have been viewed as good recommendations 
previously. This could be as simple as average estimated strength.

Original issue reported on code.google.com by srowen@myrrix.com on 28 Aug 2012 at 9:33

[GoogleCodeExporter]

Committed a simple version of this. The console now has an "average estimation 
error" statistic. Before each datum is processed, its strength is estimated. 
This statistic is the average difference from 1.0, weighted by the strength 
score. Lower is better.

This is just a first crude measure.

Original comment by srowen@myrrix.com on 1 Sep 2012 at 2:52

• Changed state: Fixed