Search This Blog

Tuesday 22 February 2011

ARTIFICIAL INTELLIGENCE AND EXPERT SYSTEMS


                                     
ON
ARTIFICIAL INTELLIGENCE AND EXPERT SYSTEMS




Abstract:
                                          Artificial intelligence is a term that encompasses many definitions and most experts agree that Ai  is concerned  with two basic ideas. First, it involves studying the thought processes of humans, secondly it deals with representing this processes via machines.
                                  Ai definition also follows the behavior by machine that ,if performed by a human being, would be called intelligence. A thought provoking definition is provided by “Rich and knight” that is artificial intelligence is a study of how to make computers do things at which, at the movement people are better.
The term intelligence behavior involves
Ø      Learning or understanding from experience.
Ø      Making sense out of  ambiguous contradictory messages
Ø      Responding quickly and successfully to new situation.
Ø      Using reasoning in solving problems and directing conduct effectively.
Ø      Applying knowledge to manipulate the environment
Thinking and reasoning.
AI method involve some kind of search mechanism that focus on most promising areas.
Ø      Inferencing: artificial intelligence involves an attempt by machines to exhibit reasoning capability. This reasoning consists of inferencing from facts and rules using heuristics or other search approaches. Artificial intelligence is unique in that it makes inferences by using a pattern matching approaches.
Pattern matching: the following of AI on pattern matching techniques artificial intelligence works with pattern matching methods such as, methods that attempt to describe objects events, or processes in terms of their qualitative features and logical computational relationships. Expert systems,robotics ,neural networks,Natural language processing, Speech understanding, Fuzzy logic,Genetic algorithms ,Intelligent tutors.
Also involve fields such as game playing, computer vision, automatic programming, machine learning. Expert system can be viewed as two environments consultation environment and development. That development is used by an expert system builders to build the components and put knowledge into the knowledge base. The consultation environment is used by an non expert to take advise and expert knowledge.
                               The three major parts that appear in expert  systems are    knowledgebase, inference engine and user interface. It also contains additional parts such as
·        Knowledge acquisition subsystem
·        Black board (work space)
·        Explanation sub system
·        Knowledge refining system
·        Enter users




Ø      Introduction:
Artificial intelligence v/s Natural intelligence:
Ø      AI is more permanent: natural intelligence is perishable from a commercial stand point in that workers can change their place of employment or forget information however, AI is permanent as long as computer systems and programs remain unchanged.
Ø      AI offers involves ease of  duplication and dissemination:
Ø      AI can be less expensive than natural intelligence:
Ø      AI can documented: Decisions made by computer can easily documented by tracing the activities of the system. Natural intelligence is difficult to document.
Ø      AI, being a computer technology, is consistent and through. Natural intelligence is erratic because people are erratic as they don’t perform consistently.
There are wide range of application of AI. They are
Expert systems,robotics ,neural networks,Natural language processing, Speech understanding, Fuzzy logic,Genetic algorithms ,Intelligent tutors.Also involve fields such as game playing, computer vision, automatic programming, machine learning.
Basic concept of expert system:
                                                     The basic concept of expert system involve expertise, experts, transferring expertise, inferencing rule and explanation capability this concepts play a major role in the aggregate development of expert systems.
v     Expertise: Expertise is the extensive, task specifying knowledge acquired from training reading and experience it includes the following
·      Theories about the problem area.
·      Metaknowledge
·      Facts about the problem area
·      Rules and procedure regarding the general problem and its situation.
These enable experts to make better and faster decision than non experts in solving complex problems. Expertise is usually associated with high degree of intelligence but it is not always associated smartest person expert knowledge is well stored, organized and quickly retrievable from an expert, expertise is usually with  vast  domain of knowledge. Expertise makes people to learn from their past success and failures.
Experts: It is difficult to define who an expert is because we can actually talk about degree or levels of expertise.                       Nevertheless, it has been said that non experts out number established experts in many fields by a ratio of 100:1. expert in the top tenth in any given area are believed to perform three times as well as a average expert and thirty times as well as those in lowest tenth.
Human expertise includes a consolation of a behavior that involves
·        Recognizing and formulating the problem
·        Solving the problem quickly and correctly
·        Learning from experience
·        Reconstructing knowledge
·        Break rules if necessary
Transferring expertise: The objective of an expert system is to transfer expertise from an expert to a computer system and then to other humans. This involves four activities they are knowledge acquisition, knowledge representation, knowledge inferencing, knowledge transferring to the users.


 










STRUCTURE OF EXPERT SYSTEM
Expert system can be viewed as two environments consultation environment and development. That development is used by an expert system builders to build the components and put knowledge into the knowledge base. The consultation environment is used by an non expert to take advise and expert knowledge.
                               The three major parts that appear in expert  systems are                        knowledgebase, inference engine and user interface. It also contains additional parts such as
·        Knowledge acquisition subsystem
·        Black board (work space)
·        Explanation sub system
·        Knowledge refining system
·        Enter users

Knowledge acquisition sub system: This is the accumulation, transfer and transformation of problem solving expertise from experts or documented knowledge sources to a computer programs for constructing and expanding the knowledge base. Acquiring knowledge from a expert is a complex task that often creates a bottle neck in expert system construction. typically the knowledge engineer helps the expert system structure the problem area by interpreting and integrating human answers to questions, drawing analysis, posing counter examples
Inference engine: The brain of the expert system , and also known as the rule interpreter. This is the component that provide a methodology for reasoning about the knowledge base and formulating conclusions, developing the agenda that organizes and controls the steps taken to solve problems when ever required.
User interface: Expert system contains language processor for friendly, problem oriented communication between the user and computer. This communication can be best be carried out in a natural language. Some times it is supplemented by menus, electronic forms and graphics.
Black Board: Is an area of working memory set a side as a data base for the description of the current problem as specified by the input data three decision can be recorded on black board they are a plan, a solution, an agenda.
 Explanation sub system : The explanation sub system interactively answers questions such as 
·        why was certain question asked by the expert How               was the conclusion reached.
·        why was a certain alternative rejected.
·        What is the plant reach solution

Knowledge refining system: This system analyze its own knowledge and its use, learn from it and improve on it for future consultation. This could lead to improvement that result in more accurate knowledge base and more effective reasoning.
The user: The user of an expert system id usually an non expert who needs adivce and training. The user is considered as part of expert system while other people are involved in its construction.
How an expert system work:
             ES construction and use consist of three major activities development, consultation, improvement .
ü      Development : The development of an expert system involves the construction of a problem specific knowledge base by acquiring   knowledge from knowledge base by acquiring knowledge from experts or documented sources the knowledge is then separated into declarative and procedural then it also includes construction of a interface engine, a blackboard, an explanation facility and any other required software, such as interfaces.
ES shell is a tool of an used to expedite development.
Consultation: Once the system as been developed and validated it can be deployed to users the ES conducts a bidirectional dialogue with a user asking for facts about a specific incident. While answer is being received the Es attempts to reach the conclusion this effort is made by inference engine , which chooses search techniques to be used to determine how the rules in the knowledge base are to be applied to each other. The user can ask for explanations. The quality of the interface capability is determined by the quality and completeness of the rules by the knowledge representation used, by the power of  the inference engine.
Improvement: The expert systems are improved several times through  a process called rapid prototyping. The computer keeps asking questions asked earlier and improve the knowledge base frequently.



PROBLEM AREAS ADDRESSED BY EXPERT SYSTEM
Expert system address a wide range of problem areas such as
                                                                                                         i.            Interpretation systems infer situation descriptions from observations. This category
Includes speech understanding image analysis, signal interpretation and many kinds of intelligent analysis. An interpretation system explains observed data by assigning them symbolic meanings describing the situation
                                                                                                       ii.            Prediction system includes weather forecasting  demographic predictions, economic forecasting, crop estimation, military marketing, and financial forecasting.
                                                                                                      iii.            Designing system develop configuration of the objects that satisfies the constrains of the design problems such problem include circuit layout, and plant layout. This constructed descriptions of objects with various relationships with one another and verify that these configurations conform to stated constraints.
                                                                                                     iv.            Debugging systems relay on planning, design, and prediction capabilities for creating specification to correct a diagnosed problem
                                                                                                       v.            Control systems adaptively govern the over all behavior of the system. To do this, the control system must repeatedly interpret the current situation, predict the future, anticipate the cause of problem and formulate a remedial plan.

BENEFITS OF EXPERT SYSTEMS
Thousands of ES are in use today almost in every industry and in every functional area. The major ES benefits are discussed
Increase output and productivity Flexibility
Easier equipment operation
Elimination of the need for expensive equipment
Knowledge transferred to remote location
Availability to solve complex problems
Ability to work with incomplete or uncertain information
Operation in hazardous  environment
Improved decision quality
Enhancement of other information systems
Decrease decision making time
Increase process and product quality
Reduce down time
Capture of scare expertise
EXPERT SYSTEMS SUCCESS FACTOR  
ü      The level of knowledge must be sufficiently high.
ü      The problem domain should be narrow
ü      Knowledge base should be improved frequently
ü      Management support must be cultivated
ü      Es shell must be of high quality and naturally store and manipulate the knowledge.
ü      Expertise must be available from atleast one cooperative expert
ü      The user interface must be friendly for novice user







       
  









     



2 comments:

Digvijay Chaudhary said...

This is really wonderful blog. Contents over here are so informative. For more on these topics, have a look here..Expert Systems and Stages of Expert System Development and Features of an Expert System

radha said...

Nice post. Keep updating Artificial Intelligence Online Course