Hello,
I tried to write a simple answer to some questions I recieved. I expose
it to this community to get new viewpoints. I don't know how familiar are
you with AI concepts and tools, therefore assume no prior knowledge.
YES! you are right: "... not all learning can be measured easily ..." &
better is to say "Some types of learning can't be measured at all"! But it
is an intrinsic essence of leaning, no matter a manager tries to measure
learning or a software tool. The differences, arise in processing details
and aids used - not in the main aspects of the problem.
There are many software tools that train people & also examine &
evaluate them pointing out their learning problems. You can use search
engines to find some of them (use the keywords "Compute Aided
Instruction", "Intelligent tutoring systems", or "Computer Based
Training").
AI enabled software can do reasoning based on the facts & rules
provided to it. It is now evident that computers can learn; as popular
examples, there are computers learnt to play chess & compose music. Why
can not they measure performance as you do so!?
In my design I've considered three types of information usable to
organizational problem analysis. The first is environmental & operational
info (logged by software applications doing routine operations, or
documented in reports, forms, questionnaires, interviews). for example: an
employee's total late times, total expenses of a project, or frequency of
employees answered NO to the question 7. (Related software systems: MIS,
other traditional IS).
The secod is special events info that is gathered or recorded relative
to our analysis objectives. This info is at higher level & usually derived
by computations, rules, or criteria acting on environmental/operational
info. For example, high rate of personnel tardiness, a contract falling
low on profit margin, low income to face to forecasted inflation rate.
(Related software systems: spreadsheets, DSS, ES, ...).
The last is analytical info can be used to analyze & evaluate
performance vs. objectives. This type is at highest level of cognitive
abstraction. This is my area of interest. (Related software systems:
Machine learning tools adapted or designed to support organizational
problem solving).
Best Wishes,
Shaya Zeighami <zeighami@aalam.net>
--Shaya Zeighami <zeighami@aalam.net>
Learning-org -- Hosted by Rick Karash <Richard@Karash.com> Public Dialog on Learning Organizations -- <http://www.learning-org.com>
"Learning-org" and the format of our message identifiers (LO1234, etc.) are trademarks of Richard Karash.