1. 1 Rule Based Systems Introduction to Production System Architecture

2. 2 Rules and ‘Productions’ ‘Production’ - a rule for changing grammar Englishs regular verbs –regular_verb + (past) --> verb + ‘ed’ Pig-latin productions –consontant string  string + consonant + ‘ay’ –vowel + string  string + ay Is this enough? What do we have to add? Can you reverse it?

3. 3 Work through this production system to decode the result z  l u  p p  h lw  llo q  r v  w r  e uqzv  ??

4. 4 Rules Form –IF antecedent THEN conclusion –IF condition THEN action –IF antecedent THEN goal Interpreters –Backward chaining »Trigger on conclusion/goal –Forward chaining »Trigger on antecedent/condition

5. 5 Rules and Logic Rules are really statement in logic –About what we believe to be true –About what should occur Translation 1 - situations –IF may_rain THEN should_take_umbrella  s. Situation s  May_rain s  Take_umbrella s “For all s, if s is a situation & s is a may_rain situation, then s is a take_umbrella situation.

6. 6 Rules and Logic Translation 2 - implied objects –IF may_rain THEN should_take_umbrella  wp. Weather w  Person p  May_rain w  Out_in p w  Take_umbrella p “For all w and p, if w is a weather and w is may_rain weather and p is a person and p is out in w, then p (should) take an umbrella.” Translations: be careful of the words –The words are just symbols »They have no meaning except help human know what the symbols are for »Just like variable names - could all be x,y,z...  wp. W w  P p  M w  O p w  T p

7. 7 Forward and Backward Chaining Rules –r1: IF may_rain THEN should_take_umbrella –r2: IF cloudy THEN may_rain Questions –“Should I take an umbrella?” –“What should I do if it is cloudy?” How did you answer the questions? –Which part of the rule did you look for? (‘match’)

8. 8 Backward Chaining Rules –R1: IF may_rain THEN should_take_umbrella –R2: IF cloudy THEN may_rain “Should I take an umbrella?” –“Do the rules indicate I should take an umbrella? »Is there a rule about “taking umbrellas”? R1: goal: should_take_umbrella »How can I prove that goal? What has to be true for r1 to hold? –may_rain is the antecedent of r1 »Can I prove that it may_rain? R2: goal: may_rain »How can I prove that goal2 What has to be true for r2 to hold –cloudy is the antecedent of r2 »How can I prove ‘cloudy’?

9. 9 Backward Chaining Goal_1  Goal_2 Goal_2  Goal_3 Goal_3  Goal_4 Goal_4  Goal_5 Question rules Direction of reasoning

10. 10 Backward Chaining Rules –R1: IF may_rain THEN should_take_umbrella –R2: IF cloudy THEN may_rain –R3: IF may_be_intense_sun THEN should_take_umbrella –R4: IF summer AND in_tropics THEN may_be_intense_sun “Should I take an umbrella?” –“Do the rules indicate I should take an umbrella? »Is there are rule about “taking umbrellas”? R1: goal: should_take_umbrella »What is antecedent for r1? R1:antecedent may_rain »Can I prove that it may_rain? R2: goal: may_rain »How can I prove may_rain R2:antecedent: cloudy »BUT NOT CLOUDY!

11. 11 Backward Chaining: Backtracking Rules –R1: IF may_rain THEN should_take_umbrella –R2: IF cloudy THEN may_rain –R3: IF may_be_intense_sun THEN should_take_umbrella –R4: IF summer AND in_tropics THEN may_be_intense_sun “Should I take an umbrella?” »Are there any other rules about umbrellas? R3: goal: should_take_umbrella »What is antecedent of R3? R3:antecedent: summer AND in tropics

12. 12 Backwards Chaining with Backtracking Goal_1  Goal_2 Goal_2  Goal_3 Goal_3  Goal_4 Goal_4  Goal_5 Question rules fail Goal_6  Goal_4 Goal_7  Goal_6 Goal_8  Goal 7 Direction of reasoning

13. 13 Backwards Chaining Systems MYCIN –‘The original expert system’ »Diagnosis of acute infections (Meningitis, blood infections) Still a good example of how it works »Also used uncertain reasoning Explanation –‘How’ did you prove that? –‘Why’ are you asking me that? »Never used ‘for real’ PROLOG –One of the two standard AI languages »A simple backwards chaining engine with backtracking

14. 14 Backwards Chaining Engines usually written ‘backwards’ Goal  Antecedent –Umbrella  may_rain –may_rain  cloudy Prolog ‘Edinburgh’ notation –umbrella :- may_rain. –may_rain :- cloudy. NB upper and lower case very important in Prolog In KnowledgeWorks –(defrule r1 :backward ) »Or special functions ‘any’, ‘test’, and ‘not’ Awkward - don’t try it first time.

15. 15 Forward Chaining Rules –R1: IF may_rain THEN should_take_umbrella –R2: IF cloudy THEN may_rain “What should I do if it is cloudy?” –“What do the rules indicate I should do if it is cloudy?” »Is there a rule that applies when it is cloudy? R2: antecedent: cloudy »What do I conclude from that antecedent, ‘cloudy’ R2: conclusion: may_rain »Is there a rule that applies when it may_rain? R1: antecedent: may_rain »What do I conclude from that antecedent: ‘may_rain’ R1: conclusion: should_take_umbrella

16. 16 Forward chaining ‘Production Systems’ –Vocabulary used differently on west and east coast of US for many years »On east coast, ‘production systems’ means forward chaining »On west coast, ‘production systems’ just means rule based systems –Usually, and in this course, ‘Production System’ means ‘forward chaining’ Classic system is OPS5 Basic strategy of KnowledgeWorks

17. 17 Forward Chaining Fact_1  Fact_2 Fact_2  Fact_3 Fact_3  Fact_4 Fact_4  Fact_5 Action=Fact_5 rules Direction of reasoning Fact_1

18. 18 Consider R1: IF sky=cloudy THEN expect=rain R2: IF expect=?X THEN weather=?X R3: IF sky=cloudy AND temperature=freezing THEN expect=snow R4: IF weather=rain THEN termperature=above_freezing What happens if ‘sky=cloudy’? What happens if ‘sky=cloudy and ‘temperature=freezing’? Conflict Resolution

19. 19 Production system interpreter Objectives: –Fire rules as the facts come in to the knowledge base –Never fire a rule unless its conditions are satisfied –Fire every rule whose conditions are satisfied Are these objectives consistent. –Forward chaining rules sometimes called ‘demons’ »From a system called “Pandemonium” –How can they be made consistent?

20. 20 Production System Strategy All rules tested at each cycle Only one rule fires at a time

21. 21 Production System Cycle 1Test all rules 2Put all rules satisfied into the ‘conflict set’ 3Choose one rule from the conflict set 4Fire the rule 5Update the dynamic database 6Repeat until goal reached or no more rules satisfied

22. 22 Conflict Resolution R1: IF sky=cloudy THEN expect=rain R2: IF expect=?X THEN weather=?X R3: IF sky=cloudy AND temperature=freezing THEN expect=snow R4: IF weather=rain THEN termperature=above_freezing What happens if ‘sky=cloudy’? What happens if ‘sky=cloudy and ‘temperature=freezing’?

23. 23 Possible Conflict Resolution Strategies Specificity Priority Lexical Ordering Source file ordering Explicit rules for conflict resolution –a rule based system within a rule based system

24. 24 Basic Production System Architecture Dynamic Memory Rule Store Conflict Set Rule Execution tickle check satisfactionSelect (resolve conflicts) execute

25. 25 Modules (Contexts) & Agendas How to modularise RBSs Agenda = Sequence of modules (Stack) Module = Named Set of rules –(Rule MODULE::rule_name IF...THEN...) Contexts popped off stack when finished.

26. 26 Agenda/Modules Example focussearch stuck defrule search::r1 IFhero is at ?Place-1 & ?Place-1 gives_access_to ?Place-2 THENmove hero to ?Place-2 defrule stuck::r1 IF context stuck & hero is at ?Place not ?Place = goal THENask-user “What do I do now?”

27. 27 Expert Systems Configuring systems defruleBACKPLANE_STRUCTURE::R1 IFthe_processor IS in place ANDthe_devices_to_mount INCLUDE ?Device THENmount ?Device defruleBACKPLANE_STRUCTURE::R1 IFmount ?Device AND?Slot_1 IS_A backplane_slot ANDNOT lower_number_available_than ?Slot_1 ?Slot_2 THEN?Slot_1 is_occupied_by ?Device AND?Device is_mounted.

28. 28 The Structure of Heuristics Trigger Constraints Antecedents Actions Conclusions Control

29. 29 Formalising Heuristics Triggers –quick & sensitive Constraints –quick & specific Antecedents –complete and sufficient Actions Conclusions Control

30. 30 Formalising Heuristics Triggers Constraints Antecedents Actions –Things effecting the outside world Conclusions –New facts for the dynamic data base Control –changes to agenda