What are your criteria? What are you trying to measure? Separate Jess evaluation rules could be written to collect statistics while the Jess engine is running. Although the right-hand-side of rules might need some additions to assist statistics collection, such as incrementing global counters, the evaluation rules could live in a separate file that you could optionally load. Evaluation rules could mirror the primary operations (at a higher salience) and summarize results (at a lower salience).
Outside of Jess, if separate Jess output is appropriately formatted, such as creating tab or comma delimited files, then spreadsheets or databases can analyze Jess output. Working with arbitrary Jess output may involve difficult parsing. Bob Kirby At 05:36 AM 8/26/2007, Noël Huebers wrote: >Hi, > >does anyone have a suggestion how to evaluate a Jess based application. >I have directed the Jess output to an external file. But going through the >output file manually is not convenient. >Any existing solutions for automated evaluation known? > >I appreciate your comments, > >Noël Huebers
