1 / 40

Genetic Computer School Computer Systems Fundamentals

Genetic Computer School Computer Systems Fundamentals. Lecturer: Dr. Nguyen Nam Hong Tel: (04) 8781 437 Mobile: 0912 312 816 Email: nguyennamhong2003@yahoo.com.au Chapter 12. Expert Systems. Chapter 12. Expert Systems. 12.1. What Is an Expert System ?

baby
Download Presentation

Genetic Computer School Computer Systems Fundamentals

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Genetic Computer SchoolComputer Systems Fundamentals Lecturer: Dr. Nguyen Nam Hong Tel: (04) 8781 437 Mobile: 0912 312 816 Email: nguyennamhong2003@yahoo.com.au Chapter 12. Expert Systems Dr. Nguyen Nam Hong, Le Quy Don Technical University

  2. Chapter 12. Expert Systems 12.1. What Is an Expert System? 12.2. Major Types of Expert System (9 types) 12.3. How Expert Systems Work? 12.4. An Expert System Life Cycle (6 phases) 12.5. Problems with Expert Systems (6 problems) *12.6. Effects of Expert systems *12.7. Comparing ES with Experts Dr. Nguyen Nam Hong, Le Quy Don Technical University

  3. 12.1. What Is an Expert System? • 12.1.1. Traditional land of AI and ES. • 12.1.2. Definition of ES. • 12.1.3. Characteristics of ESs. • 12.1.4. What ESs are not? • 12.1.5. ESs are not a generalized expert or problem solver. • 12.1.6. Expert System Shells. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  4. 12.1.1. Traditional land of AI and ES • Manufacturing companies have traditionally focused on artificial intelligence of robots and vision systems to assist assembly, management and service. • Industries in particular are more interested in expert systems. • Expert systems automate parts, and in some cases, all of the decision-making process. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  5. 12.1.2. Definition of ES • An expert system is a knowledge-intensive program that solves a problem that normally requires human expertise. • An expert system can merely assist decision making by asking relevant questions and explaining the reasons for adopting certain actions. • Typically, such a system contains • a knowledge base containing accumulated experience • a set of rules for applying the knowledge base to each particular situation that is described in the program. • Sophisticated expert systems can be enhanced with additions to the knowledge base or to the set of rules. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  6. 12.1.3. Characteristics of Ess (1/2) 1. They solve problems as well as or better than human experts. 2. They use knowledge in the form of rules or frames. 3. They interact with humans. 4. They can consider multiple hypotheses simultaneously. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  7. 12.1.3. Characteristics of Ess (2/2) Dr. Nguyen Nam Hong, Le Quy Don Technical University

  8. 12.1.4. What ESs are not? • Today expert systems are quite narrow, shallow, and brittle. • They lack the breadth of knowledge and the understanding of fundamental principles of a human expert. • Expert systems today do not "think" as a human being does. • A human being perceives significance, works with abstract models of causality, and can jump to conclusions. • Expert systems do not resort to reasoning from first principles, do not draw analogies, and lack common sense. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  9. 12.1.5. ESs are not a generalized expert or problem solver • They typically perform very limited tasks, such as interpreting, diagnosing, planning, and scheduling, that can be performed by professionals in a few minutes or hours. • Problems that cannot be solved by human experts in the same short period of time are far too difficult for an expert system. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  10. 12.1.6. Expert System Shells (1/2) • Shells are some sort of program to run our expert system. • Shells provide a layer of abstraction between the computer itself and the expert system. • They range from being programming languages (such as LISP and Prolog) to hybrid environments that mix C++, Java, and COBOL modules with expert system code (such as CLIPS, JESS, Arden, and Flex) to full-fledged GUI authoring environments that let you draw nodes and arcs and specify relationships directly. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  11. 12.1.6. Expert System Shells (2/2) • The programming languages like LISP and Prolog provide maximum flexibility and form the basis of many of the other shells. • The more sophisticated shells provide a framework for modeling your problem, making knowledge engineering a matter of creating your rules, conditions, and outcomes, instead of necessitating programming from scratch. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  12. 12.2. Major Types of Expert System • 12.2.1. Tree-based logic. • 12.2.2. State machines. • 12.2.3. Forward Chaining. • 12.2.4. Backward chaining. • 12.2.5. Case-Based Reasoning (CBR). • 12.2.6. Bayesian Belief Networks. • 12.2.7. Neural Networks. • 12.2.8. Fuzzy Logic. • 12.2.9. Object Oriented Design Dr. Nguyen Nam Hong, Le Quy Don Technical University

  13. 12.2.1. Tree-based logic (1/2) • This isn't really an expert system, so much as the easiest way to program and develop a diagnostic system. • No expert system shell is required to develop a tree-based system • A flowchart needs to be created with all the necessary branches and start coding. • The starting point is the symptom of the problem to solve, then the user is questioned which narrows down the realm of possible causes, until the question that definitively identifies the actual cause is reached. • The model is built from symptom to cause. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  14. 12.2.1. Tree-based logic (2/2) • Advantages. • Coding a procedural branching diagnostic program is fast, easy, and can be done by anyone familiar with whatever platform you choose. • It's easy to update, cheap to implement, and requires no new investments. • Disadvantages. • Everything is done manually - updating all of the links when the business model changes. • If the problem gets complex, the model may breaks down. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  15. 12.2.2. State machines • State machines have some expert system-like capabilities. • They can handle data in different ways at different times. • They model existing systems and they execute in a circular, rather than linear fashion. • A state machine is basically an object that can be in a variety of states. • How the object behaves (what it inputs and outputs) depend on what state it is currently in. • Some inputs may change the object to a new state. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  16. 12.2.3. Forward Chaining • Forward chaining is quite similar to tree-based logic in that you start from the symptoms. • Once the symptoms are identified, the expert system calculates forward to find a result. • The difference is mainly in how it was programmed. • Instead of having to program the flow of the diagnosis, identifying step after step, a set of rules can simply be defined that could all run in parallel. • The shell runs all the rules in an unspecified sequence, setting variables as appropriate and finally arriving at a result. • Forward chaining is useful for forecasting trends. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  17. 12.2.4. Backward chaining (1/2) • Backward chaining is the scientific method. • First, a hypothesis is stated. • Then, the evidence is compared. • If the evidence doesn't match, then a new hypothesis needs to be started again. • If the evidence matches, it means the correct hypothesis has been made. • The semantics of creating a backward-chaining expert system can be very similar to the forward-chaining systems. • It's the engine that handles the systems differently. • Some engines do both forward and backward chaining with the same expert system. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  18. 12.2.4. Backward chaining (2/2) • Advantages of a backward-chaining. • It doesn't have to evaluate all of the evidence to arrive at a conclusion. • If the,, expert system contains a huge number of possible hypothesis, it may perform substantially better using backward chaining, than having to evaluate all of the evidence sequentially, essentially having to process all of the cases. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  19. 12.2.5. Case-Based Reasoning (CBR) • This part can be described as defined intelligence: learning from your mistakes. • A CBR system compares the problem with previous cases, and then suggests the most successful course of action. • Instead of requiring a thoughtful domain expert and knowledge engineer to determine rules in advance, CBR systems rely on databases of past results to train itself. • This is the realm of data mining. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  20. 12.2.6. Bayesian Belief Networks • This is probably the most difficult expert system to create, and the least common. • But that depends on how accurately it is defined. • Belief networks are the champions of uncertainty, because they rely on probability rather than definite rules. • The expert system doesn't blink if a user can't answer a question • it just goes to the next most likely recommendation without missing a beat. • The knowledge engineers must have a sense for how often given causes occur, and then separately decide how likely they are to know whether or not they occurred. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  21. 12.2.7. Neural Networks (1/2) • Neural networks have to do with distributed computing. • Instead of having a single processor doing all the work in sequence, you have a bunch of processors doing the work at the same time, and then share the results with the other processors. • Neural networks attempt to mimic the behavior of the brain, processors acting as neurons firing information through synapses. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  22. 12.2.7. Neural Networks (2/2) • Most of the work of neural networks is going into pattern recognition, Japanese toy robots, and simulations. • Like CBR, neural networks are trained rather than programmed, and potentially will have the most adaptability of any expert system. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  23. 12.2.8. Fuzzy Logic • Fuzzy logic isn't an expert system but rather a set of definitions to define tolerances in manufacturing or pattern recognition. • It's more of a component than a system, but it can be integrally related to the other expert systems. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  24. 12.2.9. Object Oriented Design • All expert systems are essentially models. • they represent real-world systems and problems. • Object orientation is a very powerful tool for developing these models, and in many cases, extending them. • Most of the expert system shells described above are being expanded to allow for discrete, interdependent objects. • Object-oriented design may well prove to be essential to making expert systems effective. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  25. 12.3. How Expert Systems Work? • 12.3.1. Rule-based systems. • 12.3.2. Frame-based systems. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  26. 12.3.1. Rule-based systems (1/2) Rule–based systems have three components: • The rule base (knowledge base): A collection of “If x Then y” statements where x is a condition and y is an action. • a blackboard: A collection of facts about the real world, supplied by the user. • The inference engine: A program to search the rule base and facts and come up with a solution. • Rule-based systems stay with predefined boundaries. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  27. 12.3.1. Rule-based systems (2/2) Dr. Nguyen Nam Hong, Le Quy Don Technical University

  28. 12.3.2. Frame-based systems • Frames. • A collections of knowledge that describe related concepts by listing each concept's features and showing the relationships to other concepts. • Slots are specific pieces of information attached to frames. • Frame-based system: • a collection of frames, each of which defines a concept from a different point of view or at a different level of abstraction. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  29. 12.4. An Expert System Life Cycle(1/2) • Phase 1: Select appropriate problem • 1-3 months, resulting in proposal. • Phase 2: Development of prototype • 6-9 months, resulting in prototype and design for complete system. • Phase 3: Development of complete system • 12-18 months, resulting in operational system. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  30. 12.4. An Expert System Life Cycle(2/2) • Phase 4: Evalution/ Calibration • 12 months, resulting in adjustments to fit working envirenment. • Phase 5: Integration of system • 3 months, resulting in field-tested, fully operational system. • Phase 6: Maintaince • Ongoing changes in rule base. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  31. 12.5. Problems with Expert Systems • 12.5.1. Expert systems are limited to certain problems. • 12.5.2. Important Theoretical Problems Exist. • 12.5.3. Expert Systems Are Not Applicable to Managerial Problems. • 12.5.4. Expertise Is Collective. • 12.5.5. Expert Systems Are Expensive to Teach. • 12.5.6. A More Limited Role for Expert Systems. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  32. 12.5.1. Expert systems are limited to certain problems • In comparatively simple situations, expert systems require large, lengthy, and expensive development efforts. • For these kinds of problems, hiring or training more experts may be less expensive than building an expert system. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  33. 12.5.2. Important Theoretical Problems Exist • No expert system can adequately employ deep causal models of interesting phenomena. • No expert system can write a textbook on information systems or engage in other creative activities not explicitly foreseen by system designers. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  34. 12.5.3. Expert Systems Are Not Applicable to Managerial Problems • The applicability of expert systems to managerial problems is currently highly limited. • Managerial problems generally involve drawing facts and interpretations from divergent sources, evaluating the facts, and comparing one interpretation of the facts with another. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  35. 12.5.4. Expertise Is Collective • For many problems there is no single expert. • Expertise may be distributed throughout an organization. • Coordinating this expertise to formulate policies and actions is a key focus of management efforts. • Expert systems cannot help here because they cannot synthesize knowledge based in several different experts Dr. Nguyen Nam Hong, Le Quy Don Technical University

  36. 12.5.5. Expert Systems Are Expensive to Teach • The knowledge base of expert systems is fragile and brittle; it cannot learn or change over time. • In fast moving fields like medicine or the computer sciences, keeping the knowledge base up to date is a critical problem. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  37. 12.5.6. A More Limited Role for Expert Systems • In complex problems, many critics point out that it is a very long step from the classroom to the factory floor, to the battlefield, to the management of a Star Wars defense, or to the corporate boardroom. Dr. Nguyen Nam Hong, Le Quy Don Technical University

  38. 12.6. Applications of Expert systems 1. Controlling (VM of Fagan, 1978). 2. Designing (PEACE of Dincbas, 1980). 3. Diagnotising (NEAT, 1989). 4. Teaching (GUIDON of Clancey, 1979). 5. Intepretting (FXAA at Manhattan, 1988). 6. Supervision (NAVEX of Marsh, 1984). 7. Planning (PLANPOWER of Stansfield and Greenfield, 1987). 8. Predicating (PLAN of Boulanger, 1983). 9. Treating (BLUE BOX of Mulsant and Servan Schreiber, 1984). 10. Choosing (IREX of Gardone and Ragade, 1990). 11. Simulation (STEAMER of William, 1983). Dr. Nguyen Nam Hong, Le Quy Don Technical University

  39. 12.7. Comparing ES with Experts Dr. Nguyen Nam Hong, Le Quy Don Technical University

  40. Vocabularies • 1/ Common applications of expert systems. • 2/ Common characteristics of Expert Systems. • 3/ Different types of expert systems. • 4/ Expert Systems. • 5/ Frame-based expert systems. • 6/ Functions of the expert system shells. • 7/ Life cycle of expert system. • 8/ Problems of expert systems. • 9/ Rule -based expert systems. • 10/ Three branches of Artificial Intelligence (AI) Dr. Nguyen Nam Hong, Le Quy Don Technical University

More Related