Expert System Shells.
Wednesday, April 28, 2010
2:50 PM
Knowledge Base: Stores all the data, roles, cases and relationships. It can contain textbook knowledge gained by a human expert from many years of experience. Consists of problem-solving rules, procedures, and intrinsic data relevant to the problem domain.
Certainty factor: used by some experts to indicate the probability that a rule is correct
Forward Chaining: an inference engine process which starts with the known facts and works forward through the rules towards a conclusion. Checking the condition part of a rule to determine whether it is true or false.
Backward Chaining: Reverse of forward chaining, which starts with the conclusion and works backwards through the rules to reveal the facts needed.
Fuzzy Logic: A form of multi-valued logic derived from fuzzy set theory to deal with reasoning that is approximate rather than precise. Allows for a range of variable values
Inference engine: uses the knowledge base to find answers, make predictions and suggestions as would a human.
[[ ]]
Artificial Neural Network.
Tuesday, April 27, 2010
12:29 PM
Artifical Neural Network
- is an AI system modelled on the human brain. Used in artifical vision, voice recognition and decision support system. Mathematical or computational model that stimulates the structure and/or function of Biological Neural networks.
The development of Artificial Neural Networks extends from the desire to develop a system that could perform "intelligent" tasks similar to those performed by the human brain.
How is parallel processing different to an artificial neural network:
[[ ]]
Expert System.
Friday, April 23, 2010
9:02 AM
a computer program designed to simulate the problem-solving behavior of a human who is an expert in a narrow domain or discipline
Expert system may be classified into various groups depending on the
types of problem they can solve:
- prediction-systems-weather forecasting, financial forecasting, agricultural estimates.
- diagnostic systems-medical, software and elctronic problem diagnosis.
- design systems-assissting in the design of buildings,electronic circuits and manufacturing processes.
- monitoring systems- security, maufacturing and air traffic control.
- maintenance systems- detecting and repairing faults in electronic systems
Advantage:
The advantage of an expert system over a book written system is that the expert system will actively solve problems itself, rather than simply stroing the knowledge.
Disadvantages:
Cannot adapt to changing environments, unless knowledge base is changed, Errors may occur in the knowledge base, and lead to wrong decisions
[[ ]]
HSC IPT DSS
Tuesday, April 20, 2010
2:18 PM
i) Define the term Data Mining and give an example of how data mining is used to support decision making.
Analyzing through large amounts of of data which converts data into the information necessary for decision making. An example is a super market which may use data mining to to search for patterns in buying trends.
ii) Describe the roles of the knowledge engineer in the creation of an expert system.
The role of the knowledge engineer is they create rules for the knowledge base, stores all the data rules cases and relationships, which is used by the expert system.
b)i]
E4
B4/C4
G4
= if e4 >= L3 and =
Expert System
-Artificial intelligence for finding out diseases through symptoms.
-Weather forecast - characteristics of the current weather.
-Dendral System - scientific hypothesis formation
-
[[ ]]
