MYCIN was one of the earliest and most influential expert systems developed in the 1970s. It helped physicians diagnose blood infections and recommend antibiotic treatments. The physician would enter patient data and MYCIN would analyze the information and provide diagnosis and treatment recommendations to assist the doctor. While very effective, MYCIN was not intended to replace physicians and still required final approval from medical experts.
2. MYCIN
• It is a medical expert system used for
diagnosing blood disorders.
2
Famous Expert Systems
3. MYCIN
• It is one of the most popular expert systems of all time.
• It is a medical expert system that diagnoses bacterial
infections and recommends antibiotic therapy.
• How MYCIN helps a doctor identify patients which
bacteremia and cystitis.
• The job of the physician is just to enter the patient's
age and medical history, results of laboratory tests and
any additional information.
• The use of this program is optimal, BUT the final
decision is depending upon the physician.
3
4. It was designed by Edward shortiffle of
Stanford university in mid 1970’s.
4
5. Conti…
MYCIN has proved to be an extremely reliable
medical advisor, but keep in mind that it was
not intended to replace a physician it can be
assist physician who have a little practical with
a specific experience or diseases involved.
5
6. DENDRAL
• It is used to identify the structure of chemical
compounds.
6
7. DENDRAL
• It helps the chemists to identify the molecular
structure of an unknown compound using its
knowledge.
• The identification of unknown compounds is a
long, difficult and expensive process.
• Such system can save scientists a considerable
amount of time and effort.
7
8. • It is a product of researcher Stanford
university using knowledge of molecular
structure and chemistry ,Denderal attempts to
identify a molecular structure of unknown
compounds.
• A sample of the unknown material is analyzed
a mass spectrometer and nuclear magnetic
resonance equipment. When the data from
the analysis is entered to a denderal.
8
9. • A special procedure goes about generating
molecular structure that satisfy the
constraints imposed by the data then the
molecular structure thus generated are then
checked against the chemical expertise stored
in the knowledge base. This generating
,testing ,and sifting process eventually
produces the inferred molecular structure
9
10. PROSPECTOR
• It is used by geologists to identify sites for
mining or drilling.
10
11. • Prospector is another Expert system that
helps geologists locate ore deposit. It is a
knowledge base contains rules and heuristic
data derived from empirical data. It also
contains taxonomy of various kinds of
minerals and rocks. Many geologists were
interviewed and there expertise with
particular types ore deposits was captured.
11
12. SOPHIE
• It was designed to help students learn to
troubleshoot electronic circuits. The system
presents simulated electronic circuits and a
problems then permits the students
troubleshoot it.Test equipments also
simulated allowing the students to make
measurements in the circuits and thus help
predict the faults.
12
13. PUFF
• It is a medical expert system used for
diagnosis of respiratory conditions.
13
16. Why use Expert Systems ?
• Human experts are not always available.
• An expert system can be used anywhere, any
time.
• Human experts are not 100% reliable or
consistent.
• Experts may not be good at explaining
decisions.
16
17. Problems with Expert System
• Limited domain.
• Systems are not always up to date.
• Expert systems don’t learn.
• No “common sense”.
• Experts need to setup and maintain
expert systems.
17
18. Legal & Ethical Issues
• Who is responsible if the advice is wrong?
The user?
The domain expert?
The knowledge engineer?
The programmer of the expert system?
The company selling the software?
18
19. Features/Characteristics of Expert System
• The most obvious feature of an expert system is that
it operates as an interactive system that responds to
questions, asks for clarifications, makes
recommendations and generally aids the decision-
making process. To a user, this interactive interface
is what would distinguish an expert system from any
ordinary computer tool.
19
20. Features/Characteristics of Expert System
• An expert system must have the capability to
make logical inferences based on the
knowledge stored.
• Expert systems are very domain-specific. A
medical expert system cannot be used to find
faults in the design of an electrical circuit. This
focus on small domains is more a result of
technological limitations than anything else.
20
21. Features/Characteristics of Expert System
•The small size of the domain makes encoding and storing the
domain-specific knowledge an economic process.
•Human experts require educational cost in millions of rupees
whereas expert systems (developed once) can be copied (or
reproduced) on negligible cost.
21
22. Usefulness of Expert Systems
• Salaries of human experts are increasing continuously.
Alternatively, cost for developing and maintaining expert
systems continue to drop. This will indirectly greatly reduce
the cost of producing expertise per user.
• Human experts require educational cost in millions of rupees
whereas expert systems (developed once) can be copied (or
reproduced) on negligible cost.
22
23. Usefulness of Expert Systems
• Human experts can use their expertise for very
limited time; whereas expert systems can live for
ever.
No Retirement.
No Resignation.
No Transfer of Job.
No Death.
Always Operate at Peak Efficiency.
23
24. Dealing with Uncertainty
• The important attribute of an expert system is its ability to
deal with incomplete or uncertain information.
• In some case you simply say that you don’t know.
• Expert systems are designed to deal with inputs like this
because you may not have a particular fact, the search
process will take a different path.
• It may take longer time to come up with an answer, BUT the
expert system will give you an answer.
24
25. The Down Side of Expert System
• Development of an expert system is extremely
difficult, more difficult than developing other
conventional software.
• Good experts are hard to find.
• Extracting their knowledge takes very long time and
is very difficult.
• Coding that knowledge into a computer software is
also too much difficult.
25
26. The Down Side of Expert
Systems
• To implement an expert system practically, it
requires high hardware specification.
• It can be run on a big mainframe computer.
• Personal computer will limit its usefulness.
• A human being should always provide the final
judgment.
26