Knowledge
In order for intelligent machine could solve the problem so that the engine must have knowledge. Knowledge can be obtained from:
Past experience
From expert (Expert)
From the data stored in the database (using data mining techniques)
Knowledge is:
understanding someone either in theory or practice of a domain (specific bidang2)
Everything we know it today or apa2 who we remember today
Knowledge is power someone (people who have called expert)
Expert:
Any person who has extensive knowledge of a domain
Questions People :
Manager
= What can I use?
Technology
= How can I implement with good?
Researcher
= How can I develop?
Customer / User
= How can this help? = Is good enough find solution problems and save money? = How constraints?
Expert System is:
One branch of AI that makes widespread use of specific knowledge to solve human problems that expert level
Human Expert is:
People who have expertise in certain fields, experts who have knowledge or special abilities that other people do not know or capable in the field has.
How to repsentasikan knowledge to make an intelligent machine or a specialist? Knowledge is transformed into a form of rule / production rule
ex:
We want to help aliens to cross the street where we are expert in crossing street And the knowledge we have is:
- If traffic lights are green street action (go)
- If a red traffic light the action stop (stop)
Rule no 2 (two) parts:
part IF (condition)
THEN part (action)
Thus, knowledge on converted / represented in the form of rule, the rule:
IF the traffic light is green THEN the action is go
IF the traffic light is red THEN the action is stop
General syntax of a rule:
IF <condition>
THEN <action>
Other names:
Condition / antecedent
Action / consequent
Antecedent consists of 2 (two) parts:
Object
Value
General syntax of a rule:
IF <antecedent>
THEN <consequent>
Combination rule: AND IF <antecedent 1> <antecedent <antecedent n> 2> ... THEN <consequent>
OR IF <antecedent 1> <antecedent <antecedent n> 2> ... THEN <consequent>
According to Durkin, 1994, that rule can be represented in the form:
Relation
Recommendations
Directive
Strategy
Heuristic
Example:
Rule-Based Expert System
18.26 |
ž Relation
IF the fuel tank is empty
THEN the car is dead
ž Recommendations
IF the season is autumn (musim gugur)
AND the sky is cloudy (berawan)
AND the forecash is drizzle (gerimis)
THEN the advice is take an umbrella
ž Directive
IF the car is dead
AND the fuel tank is empty
THEN the action is refuel the car
ž Strategy
IF the car is dead
THEN the action is check the duel tank;
step1 is complete
IF step1 is complete
AND the fuel tank is full
THEN the action is check the battery;
Step2 is complete
ž Heuristic
IF the spill is liquid
AND the spill pH < 6
AND the spill smell is vinegar (cuka)
THEN the spill material is acetic acid (asam cuka)
Langganan:
Posting Komentar (Atom)
0 komentar:
Posting Komentar