 An Expert System is a system that uses expertise or
know-how in the form of rules to automate the solution of some problems in some category.1
Examples of expertise include:
- What faults can cause an automobile brake
failure accident?2
- What events can cause a fire or flooding in
a house?3
- What actions can cause a bank or trust
company to fail?
- What bank abuse of commercial powers can
ruin small firms?
- What events can cause a market to crash?
- What medical tests should a physician order?4
The source of the expertise is typically one or more
persons. The expertise is organized as simple if-then rules that connect inputs
with outputs, or causes with their effects. Sophisticated Expert Systems incorporate uncertainty,
allowing for intuitive decision making.5
For example, in the automobile
brake failure case, one of the if-then rules can be:
IF the driver fails to
control the car OR the front breaks fail OR the rear breaks fail THEN an automobile
accident occurs.
In the bank abuse of commercial power case, one of
the if-then rules can be:
IF the banker
suddenly freezes, reduces, or calls the firm's loan AND other
creditors tighten the firm's credit AND the firm is unable to
secure an alternative source of funding on short notice THEN the
firm is destabilized.6
Small business owners often have to assess the relationship
between the likelihood that their banker will freeze, reduce, or call their firm's
short-term loan - especially after the bank has indulged in speculative overinvestments -
and the likelihood that their firm is destabilized. |
The expertise can be generally available,
privileged, or a proprietary trade secret. The expertise can be objective or subjective,
certain or uncertain.
The output of an Expert System can be information, an
instruction, a prediction, or a risk judgment or likelihood. For example, in the
automobile brake failure case, an Expert System can be used to calculate the likelihood of
an accident from the probabilities of the three input premises.
| Macroknow Expert System
Services translates your objective or subjective knowledge of the problem at hand into
probabilistic predictions or likelihoods of output events occurring. Rigorous probability
methods dealing with uncertainties are used.7
What do you get from Macroknow? You get access to a private
web site featuring:
- The input and output events of your knowledge base,
- The results of Monte Carlo-based simulations, and,
- If requested, the associated sensitivity
graphs for one or two types of uncertainties.
Type I uncertainties are uncertainties in the likelihoods
of the input events. Type II uncertainties are uncertainties in the rules themselves.
Only you or those you authorize can access the private web
site on the Internet. The web site is stored for one month. Other arrangements are
possible.
You can use the web site to communicate the
results of the simulations to your customers, your suppliers, your board of directors,
your public relations department, your legal department, etc.
The power of technology must be
tempered with prudence. Prudence requires that you note the
following:
- You must always exercise care in the
interpretation and use of calculated results.
- Calculations can only be one
component, out of many, in the decision making process.
- All calculations and methods have
limited domains of adequacy.
- There is no known method to certify
the absolute correctness of any complex computer program.
- Most important, there is no
substitute for a circumspect judgement.8
|
Sources:
1 For an excellent and rigorous treatment of expert systems, see James N.
Siddall, Expert Systems For Engineers. New York, NY: Marcel Dekker, Inc., 1990.
2 Alfredo H.S. Ang and Wilson H. Tang. Probability Concepts in Engineering
Planning and Design. Vol. II-Decision, Risk, and Reliability. New York, NY: John
Wiley & Sons, Inc., 1984, at 488-489 [Automotive Brake System Failure; Excerpted from
Ang et al., 1979].
3 Ibid., at 495-497 [Basement Flooding].
4 Milton C. Weinstein, Harvey V. Fineberg, et al. Clinical Decision
Analysis. Philadelphia, PA: W.B. Saunders Company, 1980.
5 See James N. Siddall, Expert Systems For Engineers. New York,
NY: Marcel Dekker, Inc., 1990, at 143-186 [Expert Systems Incorporating Uncertainty].
6 See Edward E. Ayoub, Bank-Induced
Risks. Toronto, ON: Macroknow, Inc., 1998.
7 For the application of Monte Carlo simulations to expert systems, see James
N. Siddall, Expert Systems For Engineers. New York, NY: Marcel Dekker, Inc.,
1990, at 165-168.
8 For a discussion of the pros and cons of decision analysis, see Howard
Raiffa, Decision Analysis. New York, NY: Random House, 1968, at 268-272 [Pros and
Cons of Decision Analysis].
|
