Rule-Based/Expert Systems

1. Rule-Based/Expert Systems Catie Welsh CSE 435 – IDSS October 13, 2006

2. Outline • Defining Expert Systems • Describing uses and components of Expert Systems • Showing an example of an Expert System • Describing the underlying programming used to build an expert system.

3. Definition • Expert systems are: • Knowledge based systems • Part of the Artificial Intelligence field • Computer programs that contain some subject-specific knowledge of one or more human experts • Made up of a set of rules that analyze user supplied information about a specific class of problems. • Systems that utilize reasoning capabilities and draw conclusions.

4. Terms Associated with Expert Systems • Knowledge Engineering – building an expert system • Knowledge Engineers – the people who build the system • Knowledge Representation – the symbols used to represent the knowledge • Factual Knowledge – knowledge of a particular task domain that is widely shared • Heuristic Knowledge – more judgmental knowledge of performance in a task domain.

5. Uses of Expert Systems • Very useful to companies with a high-level of experience and expertise that cannot easily be transferred to other members. • Solves problems that would normally be tackled by a medical or other professional. • Currently used in fields such as accounting, medicine, process control, financial service, production, and human resources

6. Prominent Expert Systems • MYCIN – used to diagnose infectious blood diseases and recommend antibiotics. • DENDRAL – embedded a chemist’s knowledge of mass spectrometry rules to use in analysis. • CADUCEUS – used to analyze blood-borne infectious bacteria • CLIPS and Prolog programming languages are both used in expert systems • The Age of Empire game uses CLIPS to control its AI

7. Building an Expert System • Can be built from scratch – using lots of if-then-else statements. • There are many products being sold to make programming these systems easier. • A few that I’m aware of: • Exsys – www.exsys.com – provides an easy to use user interface to develop traditional applications or web-based solutions. • Barisoft – www.barisoft.com – developed by a PSU professor and finely tuned by his students • CLIPS - provides a complete environment for the construction of rules and/or object based expert systems • Jess – the Rule Engine, built on top of CLIPS, for the Java Platform

8. Components of an Expert System • Set of Rules – derived from the knowledge base and used by the interpreter to evaluate the inputted data • Knowledge Engineer – decides how to represent the experts knowledge and how to build the inference engine appropriately for the domain • Interpreter – interprets the inputted data and draws a conclusion based on the users responses.

9. Problem-solving Models • Forward-chaining – starts from a set of conditions and moves towards some conclusion • Backward-chaining – starts with a list of goals and the works backwards to see if there is any data that will allow it to conclude any of these goals. • Both problem-solving methods are built into inference engines or inference procedures

10. Knowledge Objects • Classes – questions the user is asked • Parameters – place holder for a character string which can be variable and used to decide if/when a question is asked • Procedures – definitions of calls to external procedures • Rule Nodes – inferencing is done by a tree structure (decision trees!) and the nodes are called rule nodes.

11. Rule Base as a forest of trees • Top node is called the goal node. • Leaves are rule nodes – several types of rule nodes • Evidence node – functions to obtain information from the operator by asking specific questions • External node – indicates that data will be used which was obtained from a procedure call • Reference node – refers to another tree or subtree

12. Advantages of Expert Systems • Provide consistent answers for repetitive decisions, processes and tasks. • Hold and maintain significant levels of information. • Reduce employee training costs • Centralize the decision making process. • Create efficiencies and reduce the time needed to solve problems.

13. Advantages (cont’d) • Combine multiple human expert intelligences • Reduce the amount of human errors. • Give strategic and comparative advantages creating entry barriers to competitors • Review transactions that human experts may overlook.

14. Disadvantages of Expert Systems • Lack human common sense needed in some decision making. • Will not be able to give the creative responses that human experts can give in unusual circumstances. • Domain experts cannot always clearly explain their logic and reasoning. • Challenges of automating complex processes. • Lack of flexibility and ability to adapt to changing environments. • Not being able to recognize when no answer is available.

15. Example Expert System • This is an example expert system built to help show a user their potential risk of contracting colon cancer based on their lifestyle and family history. • http://www.barisoft.com/mycancercheck/

16. Example System • OBJECT NUMBER 21 • Name: [Age_Related] • Prompt: How OLD were you when you were FIRST diagnosed with any of the other related cancer? • Max. selections: 100 • Legal Values: • -0 • ----------------- • OBJECT NUMBER 22 • Name: {orc_age_rpt} • Prompt: Please enter a value for the ORC_AGE_RPT. • Max. selections: open • Legal Values: • -OPEN STRING • ----------------- • OBJECT NUMBER 23 • Name: Two_Related_Cancers • Prompt: Have you ever been diagnosed with any TWO or more of the following cancers? Colon, Gastric, Small Bowel, Hepatobiliary, Pancreatic, Endometrial, Ovarian, Renal pelvis, or Ureter. • Max. selections: 1 • Legal Values: • -Yes • -No • ----------------- • OBJECT NUMBER 24 • Name: {Non_Related_Cancers} • Prompt: Please enter any additional CANCERS for which YOU have been diagnosed. Example: Breast, Lymphoma. If no other cancers, then type None. • Max. selections: open • Legal Values: • -OPEN STRING

17. Example System (cont’d) • * * * * RULES SUMMARY * * * * • No. of rules defined: 129 • RULE NUMBER 1:disclaimer-no • IF • Disclaimer=No - I do not agree to the terms and conditions in the disclaimer • THEN • {status}=stop program • !Action=!LOOP • RULE NUMBER 2:disclaimer-yes • IF • Disclaimer=Yes - I have read the disclaimer and agree to all of its terms and conditions • THEN • {status}=continue with program • RULE NUMBER 3:Object Initialization • IF • THEN • [LS_Score]=0 • RULE NUMBER 4:BMI Calculation • IF • [height]==!NOTBLANK • [weight]==!NOTBLANK • THEN • [bmi]=([weight]*703)/([height]*[height])

18. More Rules for the Example System • RULE NUMBER 123:Calc Greater Next or Suggested • IF • [age]==!NOTBLANK • [Suggested_Screen_Age]==!NOTBLANK • [Next_Scrn_Age]==!NOTBLANK • THEN • [Dif_Next]=[Next_Scrn_Age]-[Suggested_Screen_Age] • RULE NUMBER 124:If Next later than Suggested • IF • [Dif_Next]=>0 • THEN • [Age_Difference]=[age]-[Next_Scrn_Age] • RULE NUMBER 125:If Next less or equal Suggested • IF • [Dif_Next]=<1 • THEN • [Age_Difference]=[age]-[Suggested_Screen_Age] • RULE NUMBER 126:Get1stScrn Now • IF • [Age_Difference]==0 • THEN • {screening_rec_age_rpt}=Schedule Screening Now at Age ([age]). • [Get_Screened_Age]=[age]

19. Conclusions • Very interesting field of AI. • Expert Systems are extremely useful in the right domain. • They are inflexible and require a lot of collaboration between a knowledge engineer and a domain expert. • When implemented correctly, expert systems remove human error from the equation.

20. References • “Expert Systems and Artificial Intelligence”. Engelmore, R., Feigenbaum, E., Chapter 1. http://www.wtec.org/loyola/kb/c1_s1.htm • “The Origin of Rule-Based Systems in AI”, Davis, R., King, J. • “Expert System – Wikipedia”, http://en.wikipedia.org/wiki/Expert_system