Expert system chapter 7

Question 1

Expert system computer–based systems used to solve problems
Components of an expert system shell, knowlage base
shell shell which can be used for solving different problems just by changing the knowledge base to one that matches the type of problem.
contains:
1. user interface
2. explanation system
3. inference engine
4. knowledge base editor
knowledge base contains:
1. database of facts
2. rule base
user interface it allows the user to interact with the computer by using a keyboard, mouse and a monitor to answer yes/no questions or choose the answer from a dropdown list and used to display the output of the explanation system.
infrence engine it receives user input and performs reasoning (chaining) by comparing the user input with the facts contained in the knowledge base using the rule base.
explanation system causes the expert system to explain to a user how a particular conclusion was reached
knowlage base editor software that allows the knowledge engineer to edit the rules and facts within the knowledge base to bring them up to date.
database of facts The knowledge base contains a collection of facts which are collected from human experts in the field by a knowledge engineer
rule base conditional statements (IF…THEN rules) that define relationships between different facts or conditions (also by knowledge engineer)
steps to create an expert system (10 steps) 1. knowlage engineer and human expert work together in order to define the problem the expert system will solve.
2. determine the domain of the system.
3.Outline the goals and objectives of the expert system and what problems or tasks the system is intended to address and solve
4. knowledge engineer collects relevant knowledge from domain experts
5. knowledge engineer creates the knowledge base with the gathered information.
6. knowledge engineer chooses the type of inference enginethat best suits the nature of the problems in the domain
7.Develop the Inference Engine to provide the reasoning aspects.
8.Design the User Interface, Explanation system and the Knowledge base editor
9. test the expert system to ensure that it performs accurately and reliably, validate by comparing its output to expected results and refine the knowledge base and rules as needed.
10. Establish a plan for maintaining the expert system
how does the knowledge engineer colloect information from interviews documents and reviews from human experts
types of inference engines/ chaining backwards chaining and forward chaining
what is chaining inference engine chains together what is known as IF–THEN rules to form a line of reasoning
what is forward chaining (data driven)
inference engine takes the user’s answer and searches the rule base for a rule where the IF part (condition) matches the user’s answer and shoots the THEN part (action) whenever there is a match https://images.cram.com/images/upload-flashcard/92/10/98/38921098_m.png
what is backward chaining (goal driven)
inference engine takes the user’s answer and searches the rule base for a rule where the THEN part (action) matches the user’s answer and shoots the IF part (condition) whenever there is a match https://images.cram.com/images/upload-flashcard/92/11/16/38921116_m.png
applications of forward chaining Planning, monitoring, controlling, and interpreting applications
applications of backward chaining Classification systems or diagnosis applications
advantages of expert systems 1. provide answers to questions outside the knowledge you have
2. aid professional people to remember/ observe new things
3. provides consistent and logical responses
4. can be used at anytime
5. provides solutions faster than humans
disadvantages of expert systems 1. have no intuition that humans have
2. depend on the quality of the data
3. expensive to create
Applications of forward chaining gaming(both)
problem solving(both)
dosing drugs
fraud control
robotics
Applications of backward chaining diagnosis
gaming (both)
problem solving(both)
chess games
self drive
scenarios where expert systems are used mineral prospecting
investment analysis
financial planning
insurance planning
car engine fault diagnosis
medical diagnosis
route scheduling for delivery vehicles
plant and animal identifications