Presentation is loading. Please wait.

Presentation is loading. Please wait.

1 Computer Science Department California Polytechnic State University San Luis Obispo, CA, U.S.A. CPE/CSC 481: Knowledge-Based Systems Franz J. Kurfess.

Similar presentations


Presentation on theme: "1 Computer Science Department California Polytechnic State University San Luis Obispo, CA, U.S.A. CPE/CSC 481: Knowledge-Based Systems Franz J. Kurfess."— Presentation transcript:

1 1 Computer Science Department California Polytechnic State University San Luis Obispo, CA, U.S.A. CPE/CSC 481: Knowledge-Based Systems Franz J. Kurfess

2 2 Usage of the Slides  these slides are intended for the students of my CPE/CSC 481 “Knowledge-Based Systems” class at Cal Poly SLO  if you want to use them outside of my class, please let me know  I usually put together a subset for each quarter as a “Custom Show”  to view these, go to “Slide Show => Custom Shows”, select the respective quarter, and click on “Show”  in Apple Keynote (.key files), I use the “Skip” feature to achieve similar results  To print them, I suggest to use the “Handout” option  4, 6, or 9 per page works fine  Black & White should be fine; there are few diagrams where color is important © Franz J. Kurfess 2

3 3 Rule-Based Reasoning  Motivation & Objectives  Reasoning in Knowledge-Based Systems  Shallow and Deep Reasoning  Forward and Backward chaining  Rule-based Systems  CLIPS/JESS  Facts  Rules  Variables  Pattern Matching  Other Rule-based Systems  Important Concepts and Terms  Chapter Summary © Franz J. Kurfess 3

4 4 Reasoning in Knowledge- Based Systems  shallow and deep reasoning  forward and backward chaining  alternative inference methods  metaknowledge 4 © Franz J. Kurfess

5 5 Motivation  CLIPS is a decent example of an expert system shell  rule-based, forward-chaining system  it illustrates many of the concepts and methods used in other ES shells  it allows the representation of knowledge, and its use for solving suitable problems © Franz J. Kurfess

6 6 Objectives  be familiar with the important concepts and methods used in rule-based ES shells  facts, rules, pattern matching, agenda, working memory, forward chaining  understand the fundamental workings of an ES shell  knowledge representation  reasoning  apply rule-based techniques to simple examples  evaluate the suitability of rule-based systems for specific tasks dealing with knowledge © Franz J. Kurfess

7 7 Shallow and Deep Reasoning  shallow reasoning  also called experiential reasoning  aims at describing aspects of the world heuristically  short inference chains  possibly complex rules  deep reasoning  also called causal reasoning  aims at building a model of the world that behaves like the “real thing”  long inference chains  often simple rules that describe cause and effect relationships 7 © Franz J. Kurfess

8 8 Examples Shallow and Deep Reasoning  shallow reasoning ❖ deep reasoning IF a car has a good battery good spark plugs gas good tires THEN the car can move IF the battery is good THEN there is electricity IF there is electricity ANDgood spark plugs THEN the spark plugs will fire IF the spark plugs fire AND there is gas THEN the engine will run IF the engine runs AND there are good tires THEN the car can move 8 © Franz J. Kurfess

9 9 Forward Chaining  given a set of basic facts, we try to derive a conclusion from these facts  example: What can we conjecture about Clyde? IF elephant(x) THEN mammal(x) IF mammal(x) THEN animal(x) elephant (Clyde) modus ponens: IF p THEN q p q unification: find compatible values for variables 9 © Franz J. Kurfess

10 10 Forward Chaining Example IF elephant(x) THEN mammal(x) IF mammal(x) THEN animal(x) elephant(Clyde) modus ponens: IF p THEN q p q elephant (Clyde) IF elephant( x ) THEN mammal( x ) unification: find compatible values for variables 10 © Franz J. Kurfess

11 11 Forward Chaining Example IF elephant(x) THEN mammal(x) IF mammal(x) THEN animal(x) elephant(Clyde) modus ponens: IF p THEN q p q elephant (Clyde) IF elephant(Clyde) THEN mammal(Clyde) unification: find compatible values for variables 11 © Franz J. Kurfess

12 12 Forward Chaining Example IF elephant(x) THEN mammal(x) IF mammal(x) THEN animal(x) elephant(Clyde) modus ponens: IF p THEN q p q elephant (Clyde) IF elephant(Clyde) THEN mammal(Clyde) IF mammal( x ) THEN animal( x ) unification: find compatible values for variables 12 © Franz J. Kurfess

13 13 Forward Chaining Example IF elephant(x) THEN mammal(x) IF mammal(x) THEN animal(x) elephant(Clyde) modus ponens: IF p THEN q p q elephant (Clyde) IF elephant(Clyde) THEN mammal(Clyde) IF mammal(Clyde) THEN animal(Clyde) unification: find compatible values for variables 13 © Franz J. Kurfess

14 14 Forward Chaining Example IF elephant(x) THEN mammal(x) IF mammal(x) THEN animal(x) elephant(Clyde) modus ponens: IF p THEN q p q elephant (Clyde) IF elephant(Clyde) THEN mammal(Clyde) IF mammal(Clyde) THEN animal(Clyde) animal( x ) unification: find compatible values for variables 14 © Franz J. Kurfess

15 15 Forward Chaining Example IF elephant(x) THEN mammal(x) IF mammal(x) THEN animal(x) elephant(Clyde) modus ponens: IF p THEN q p q elephant (Clyde) IF elephant(Clyde) THEN mammal(Clyde) IF mammal(Clyde) THEN animal(Clyde) animal(Clyde) unification: find compatible values for variables 15 © Franz J. Kurfess

16 16 Backward Chaining  try to find supportive evidence (i.e. facts) for a hypothesis  example: Is there evidence that Clyde is an animal? IF elephant(x) THEN mammal(x) IF mammal(x) THEN animal(x) elephant (Clyde) modus ponens: IF p THEN q p q unification: find compatible values for variables 16 © Franz J. Kurfess

17 17 Backward Chaining Example IF elephant(x) THEN mammal(x) IF mammal(x) THEN animal(x) elephant(Clyde) modus ponens: IF p THEN q p q IF mammal( x ) THEN animal( x ) animal(Clyde) unification: find compatible values for variables ? 17 © Franz J. Kurfess

18 18 Backward Chaining Example IF elephant(x) THEN mammal(x) IF mammal(x) THEN animal(x) elephant(Clyde) modus ponens: IF p THEN q p q IF mammal(Clyde) THEN animal(Clyde) animal(Clyde) unification: find compatible values for variables ? 18 © Franz J. Kurfess

19 19 Backward Chaining Example IF elephant(x) THEN mammal(x) IF mammal(x) THEN animal(x) elephant(Clyde) modus ponens: IF p THEN q p q IF elephant( x ) THEN mammal( x ) IF mammal(Clyde) THEN animal(Clyde) animal(Clyde) unification: find compatible values for variables ? ? 19 © Franz J. Kurfess

20 20 Backward Chaining Example IF elephant(x) THEN mammal(x) IF mammal(x) THEN animal(x) elephant(Clyde) modus ponens: IF p THEN q p q IF elephant(Clyde) THEN mammal(Clyde) IF mammal(Clyde) THEN animal(Clyde) animal(Clyde) unification: find compatible values for variables ? ? 20 © Franz J. Kurfess

21 21 Backward Chaining Example IF elephant(x) THEN mammal(x) IF mammal(x) THEN animal(x) elephant(Clyde) modus ponens: IF p THEN q p q elephant ( x ) IF elephant(Clyde) THEN mammal(Clyde) IF mammal(Clyde) THEN animal(Clyde) animal(Clyde) unification: find compatible values for variables ? ? ? 21 © Franz J. Kurfess

22 22 Backward Chaining Example IF elephant(x) THEN mammal(x) IF mammal(x) THEN animal(x) elephant(Clyde) modus ponens: IF p THEN q p q elephant (Clyde) IF elephant(Clyde) THEN mammal(Clyde) IF mammal(Clyde) THEN animal(Clyde) animal(Clyde) unification: find compatible values for variables 22 © Franz J. Kurfess

23 23 Forward vs. Backward Chaining Forward ChainingBackward Chaining planning, controldiagnosis data-drivengoal-driven (hypothesis) bottom-up reasoningtop-down reasoning find possible conclusions supported by given facts find facts that support a given hypothesis similar to breadth-first searchsimilar to depth-first search antecedents (LHS) control evaluation consequents (RHS) control evaluation 23 © Franz J. Kurfess

24 24 Reasoning in Rule-Based Systems 24 © Franz J. Kurfess

25 25 ES Elements  knowledge base  inference engine  working memory  agenda  explanation facility  knowledge acquisition facility  user interface 25 © Franz J. Kurfess

26 26 ES Structure Knowledge Base Inference EngineWorking Memory User Interface Knowledge Acquisition Facility Explanation Facility Agenda 26 © Franz J. Kurfess

27 27 Rule-Based ES  knowledge is encoded as IF … THEN rules  these rules can also be written as production rules  the inference engine determines which rule antecedents are satisfied  the left-hand side must “match” a fact in the working memory  satisfied rules are placed on the agenda  rules on the agenda can be activated (“fired”)  an activated rule may generate new facts through its right-hand side  the activation of one rule may subsequently cause the activation of other rules 27 © Franz J. Kurfess

28 28 Example Rules Production Rules the light is red ==> stop the light is green ==> go antecedent (left-hand-side) consequent (right-hand-side) IF … THEN Rules Rule: Red_Light IFthe light is red THENstop Rule: Green_Light IFthe light is green THENgo antecedent (left-hand-side) consequent (right-hand-side) 28 © Franz J. Kurfess

29 29 MYCIN Sample Rule Human-Readable Format IF the stain of the organism is gram negative AND the morphology of the organism is rod AND the aerobiocity of the organism is gram anaerobic THEN the there is strongly suggestive evidence (0.8) that the class of the organism is enterobacteriaceae MYCIN Format IF(AND (SAME CNTEXT GRAM GRAMNEG) (SAME CNTEXT MORPH ROD) (SAME CNTEXT AIR AEROBIC) THEN (CONCLUDE CNTEXT CLASS ENTEROBACTERIACEAE TALLY.8) [Durkin 94, p. 133] 29 © Franz J. Kurfess

30 30 Inference Engine Cycle  describes the execution of rules by the inference engine  conflict resolution  select the rule with the highest priority from the agenda  execution  perform the actions on the consequent of the selected rule  remove the rule from the agenda  match  update the agenda  add rules whose antecedents are satisfied to the agenda  remove rules with non-satisfied agendas  the cycle ends when no more rules are on the agenda, or when an explicit stop command is encountered 30 © Franz J. Kurfess

31 31 Forward and Backward Chaining  different methods of rule activation  forward chaining (data-driven)  reasoning from facts to the conclusion  as soon as facts are available, they are used to match antecedents of rules  a rule can be activated if all parts of the antecedent are satisfied  often used for real-time expert systems in monitoring and control  examples: CLIPS, OPS5  backward chaining (query-driven)  starting from a hypothesis (query), supporting rules and facts are sought until all parts of the antecedent of the hypothesis are satisfied  often used in diagnostic and consultation systems  examples: EMYCIN 31 © Franz J. Kurfess

32 32 Foundations of Expert Systems Rule-Based Expert Systems Knowledge BaseInference EngineRules Pattern Matching Facts Rete AlgorithmMarkov Algorithm Post Production Rules Conflict Resolution Action Execution 32 © Franz J. Kurfess

33 33 Post Production Systems  production rules were used by the logician Emil L. Post in the early 40s in symbolic logic  Post’s theoretical result  any system in mathematics or logic can be written as a production system  basic principle of production rules  a set of rules governs the conversion of a set of strings into another set of strings  these rules are also known as rewrite rules  simple syntactic string manipulation  no understanding or interpretation is required  also used to define grammars of languages  e.g. BNF grammars of programming languages 33 © Franz J. Kurfess

34 34 Emil Post  20th century mathematician  worked in logic, formal languages  truth tables  completeness proof of the propositional calculus as presented in Principia Mathematica  recursion theory  mathematical model of computation similar to the Turing machine  not related to Emily Post ;-) 34 © Franz J. Kurfess

35 35 Markov Algorithms  in the 1950s, A. A. Markov introduced priorities as a control structure for production systems  rules with higher priorities are applied first  allows more efficient execution of production systems  but still not efficient enough for expert systems with large sets of rules  he is the son of Andrey Markov, who developed Markov chains 35 © Franz J. Kurfess

36 36 Rete Algorithm  developed by Charles L. Forgy in the late 70s for CMU’s OPS (Official Production System) shell  stores information about the antecedents in a network  in every cycle, it only checks for changes in the networks  this greatly improves efficiency 36 © Franz J. Kurfess

37 37 Rete Network © Franz Kurfess: Reasoning 37 © Franz J. Kurfess

38 38 CLIPS Introduction  CLIPS stands for  C Language Implementation Production System  forward-chaining  starting from the facts, a solution is developed  pattern-matching  Rete matching algorithm: find ``fitting'' rules and facts  knowledge-based system shell  empty tool, to be filled with knowledge  multi-paradigm programming language  rule-based, object-oriented (Cool) and procedural © Franz J. Kurfess

39 39 The CLIPS Programming Tool  history of CLIPS  influenced by OPS5 and ART  implemented in C for efficiency and portability  developed by NASA, distributed & supported by COSMIC  runs on PC, Mac, UNIX, VAX VMS  CLIPS provides mechanisms for expert systems  a top-level interpreter  production rule interpreter  object oriented programming language  LISP-like procedural language [Jackson 1999] © Franz J. Kurfess

40 40 Components of CLIPS  rule-based language  can create a fact list  can create a rule set  an inference engine matches facts against rules  object-oriented language (COOL)  can define classes  can create different sets of instances  special forms allow you to interface rules and objects [Jackson 1999] © Franz J. Kurfess

41 41 Notation  symbols, characters, keywords  entered exactly as shown:  (example)  square brackets [...]  contents are optional:  (example [test])  pointed brackets (less than / greater than signs)  replace contents by an instance of that type  (example )  star *  replace with zero or more instances of the type  *  plus +  replace with one or more instances of the type  + (is equivalent to * )  vertical bar |  choice among a set of items:  true | false © Franz J. Kurfess

42 42 Tokens and Fields  tokens  groups of characters with special meaning for CLIPS,  e.g. ( ) \ separated by delimiters  ( space, tab, Carriage Return,...)  fields  particularly important group of tokens  CLIPS primitive data types  float, integer, symbol, string, external address, instance name, instance address © Franz J. Kurfess

43 43 CLIPS Primitive Data Types  float: decimal point ( 1.5 ) or exponential notation ( 3.7e10 )  integer: [sign] +  symbol: +  e.g. this-is-a-symbol, wrzlbrmft,  string: delimited by double quotes  e.g. "This is a string"  external address  address of external data structure returned by user-defined functions  instance name (used with Cool)  delimited by square brackets  instance address (used with Cool)  return values from functions © Franz J. Kurfess

44 44 Invoke / Exit CLIPS  entering CLIPS double-click on icon, or type program name(CLIPS) system prompt appears: CLIPS>  exiting CLIPS at the system prompt CLIPS> type (exit)  Note: enclosing parentheses are important; they indicate a command to be executed, not just a symbol © Franz J. Kurfess

45 45 Facts  elementary information items (“chunks”)  relation name  symbolic field used to access the information  often serves as identifier for the fact  slots (zero or more)  symbolic fields with associated values  deftemplate construct  used to define the structure of a fact  names and number of slots  deffacts  used to define initial groups of facts © Franz J. Kurfess

46 46 Examples of Facts  ordered fact (person-name Franz J. Kurfess)  deftemplate fact (deftemplate person "deftemplate example” (slot name) (slot age) (slot eye-color) (slot hair-color)) © Franz J. Kurfess

47 47 Defining Facts  Facts can be asserted CLIPS> (assert (today is sunday))  Facts can be listed CLIPS> (facts) f-0 (today is sunday)  Facts can be retracted CLIPS> (retract 0) CLIPS> (facts) [Jackson 1999] © Franz J. Kurfess

48 48 Instances  an instance of a fact is created by (assert (person (name "Franz J. Kurfess") (age 46) (eye-color brown) (hair-color brown)) ) © Franz J. Kurfess

49 49 Initial Facts (deffacts kurfesses "some members of the Kurfess family" (person (name "Franz J. Kurfess") (age 46) (eye-color brown)(hair-color brown)) (person (name "Hubert Kurfess") (age 44) (eye-color blue)(hair-color blond)) (person (name "Bernhard Kurfess") (age 41) (eye-color blue)(hair-color blond)) (person (name "Heinrich Kurfess") (age 38) (eye-color brown)(hair-color blond)) (person (name "Irmgard Kurfess") (age 37) (eye-color green)(hair-color blond)) ) © Franz J. Kurfess

50 50 Usage of Facts  adding facts  (assert +)  deleting facts  (retract +)  modifying facts  (modify ( )+ )  retracts the original fact and asserts a new, modified fact  duplicating facts  (duplicate ( )+ )  adds a new, possibly modified fact  inspection of facts  (facts)  prints the list of facts  (watch facts)  automatically displays changes to the fact list © Franz J. Kurfess

51 51 Rules  general format (defrule ["comment"] * ; left-hand side (LHS) ; or antecedent of the rule => *) ; right-hand side (RHS) ; or consequent of the rule © Franz J. Kurfess

52 52 Rule Components  rule header  defrule keyword, name of the rule, optional comment string  rule antecedent (LHS)  patterns to be matched against facts  rule arrow  separates antecedent and consequent  rule consequent (RHS)  actions to be performed when the rule fires © Franz J. Kurfess

53 53 Examples of Rules  simple rule (defrule birthday-FJK (person (name "Franz J. Kurfess") (age 46) (eye-color brown) (hair-color brown)) (date-today April-13-02) => (printout t "Happy birthday, Franz!") (modify 1 (age 47)) ) © Franz J. Kurfess

54 54 Properties of Simple Rules  very limited:  LHS must match facts exactly  facts must be accessed through their index number  changes must be stated explicitly  can be enhanced through the use of variables © Franz J. Kurfess

55 55 Variables, Operators, Functions  variables  symbolic name beginning with a question mark " ? "  variable bindings  variables in a rule pattern (LHS) are bound to the corresponding values in the fact, and then can be used on the RHS  all occurrences of a variable in a rule have the same value  the left-most occurrence in the LHS determines the value  bindings are valid only within one rule  access to facts  variables can be used to make access to facts more convenient: ?age <- (age harry 17) © Franz J. Kurfess

56 56 Wildcards  question mark ?  matches any single field within a fact  multi-field wildcard $?  matches zero or more fields in a fact © Franz J. Kurfess

57 57 Field Constraints  not constraint ~  the field can take any value except the one specified  or constraint |  specifies alternative values, one of which must match  and constraint &  the value of the field must match all specified values  mostly used to place constraints on the binding of a variable © Franz J. Kurfess

58 58 Mathematical Operators  basic operators ( +,-,*,/) and many functions (trigonometric, logarithmic, exponential) are supported  prefix notation  no built-in precedence, only left-to-right and parentheses  test feature  evaluates an expression in the LHS instead of matching a pattern against a fact  pattern connectives  multiple patterns in the LHS are implicitly AND -connected  patterns can also be explicitly connected via AND, OR, NOT  user-defined functions  external functions written in C or other languages can be integrated  Jess is tightly integrated with Java © Franz J. Kurfess

59 59 Examples of Rules  more complex rule (defrule find-blue-eyes (person (name ?name) (eye-color blue)) => (printout t ?name " has blue eyes.” crlf)) © Franz J. Kurfess

60 60 Example Rule with Field Constraints (defrule silly-eye-hair-match (person (name ?name1) (eye-color ?eyes1&blue|green) (hair-color ?hair1&~black)) (person (name ?name2&~?name1) (eye-color ?eyes2&~?eyes1) (hair-color ?hair2&red|?hair1)) => (printout t ?name1 " has "?eyes1 " eyes and " ?hair1 " hair."crlf) (printout t ?name2 " has "?eyes2 " eyes and " ?hair2 " hair."crlf)) © Franz J. Kurfess

61 61 Using Templates (deftemplate student “a student record” (slot name (type STRING)) (slot age (type NUMBER) (default 18))) CLIPS> (assert (student (name fred))) (defrule print-a-student (student (name ?name) (age ?age)) => (printout t ?name “ is “ ?age) ) [Jackson 1999] © Franz J. Kurfess

62 62 An Example CLIPS Rule (defrule sunday “Things to do on Sunday” (salience 0); salience in the interval [-10000, 10000] (today is Sunday) (weather is sunny) => (assert (chore wash car)) (assert (chore chop wood)) ) [Jackson 1999] © Franz J. Kurfess

63 63 [Jackson 1999] Getting the Rules Started  The reset command creates a special fact CLIPS> (load “today.clp”) CLIPS> (facts) CLIPS> (reset) CLIPS> (facts) f-0 (initial-fact)... (defrule start (initial-fact) => (printout t “hello”)) © Franz J. Kurfess

64 64 Variables & Pattern Matching  Variables make rules more applicable (defrule pick-a-chore (today is ?day) (chore is ?job) => (assert (do ?job on ?day)) )  if conditions are matched, then bindings are used [Jackson 1999] © Franz J. Kurfess

65 65 Retracting Facts from a Rule (defrule do-a-chore (today is ?day); ?day must have a consistent binding ?chore <- (do ?job on ?day) => (printout t ?job “ done”) (retract ?chore) )  a variable must be assigned to the item for retraction [Jackson 1999] © Franz J. Kurfess

66 66 Pattern Matching Details  one-to-one matching (do ?job on ?day) (do washing on monday)  use of wild cards (do ? ? monday) (do ? on ?) (do ? ? ?day) (do $?) (do $? monday) (do ?chore $?when) [Jackson 1999] © Franz J. Kurfess

67 67 Defining Functions in CLIPS  Uses a LISP or Scheme-like syntax (deffunction function-name (arg... arg) action... action) (deffunction hypotenuse (?a ?b) (sqrt (+ (* ?a ?a) (* ?b ?b)))) (deffunction initialize () (clear) (assert (today is sunday))) [Jackson 1999] © Franz J. Kurfess

68 68 Defining Classes & Instances  defining the class CAR (defclass car (is-a user) (name) (made-by))  defining an instance of CAR (make-instance corvette of car (made-by chevrolet)) [Jackson 1999] © Franz J. Kurfess

69 69 Concrete & Abstract Classes  some classes only exist for inheritance purposes Person ManWoman JackJill [Jackson 1999] © Franz J. Kurfess

70 70 Managing Instances  Commands to display instances CLIPS> (instances) [corvette] of car CLIPS> (send [corvette] print) [corvette] of car (made-by chevrolet)  Command to group instances (in a file) (definstances (corvette of car (made-by chevrolet)) (thunderbird of car (made-by ford))) [Jackson 1999] © Franz J. Kurfess

71 71 Clearing & Resetting Instances  deleting an instance CLIPS> (send [corvette] delete)  deleting all instances CLIPS> (unmake-instance *)  resetting creates an initial object CLIPS> (reset) CLIPS> (instances) [initial-object] of INITIAL-OBJECT [Jackson 1999] © Franz J. Kurfess

72 72 Manipulation of Constructs  show list of constructs (list-defrules), (list-deftemplates), (list-deffacts)  prints a list of the respective constructs  show text of constructs (ppdefrule ), (ppdeftemplate ), (ppdeffacts )  displays the text of the construct (``pretty print'')  deleting constructs (undefrule ), (undeftemplate ), (undeffacts )  deletes the construct (if it is not in use)  clearing the CLIPS environment (clear)  removes all constructs and adds the initial facts to the CLIPS environment © Franz J. Kurfess

73 73 Input / Output  print information (printout *)  logical device frequently is the standard output device t (terminal)  terminal input (read [ ]), (readline [ ])  read an atom or string from a logical device  the logical device can be a file which must be open  open / close file (open [ ]), (close [ ])  open /close file with as internal name  load / save constructs from / to file (load ), (save )  backslash \ is a special character and must be ``quoted'' (preceded by a backslash \)  e.g. (load "B:\\clips\\example.clp") © Franz J. Kurfess

74 74 Program Execution  agenda  if all patterns of a rule match with facts, it is put on the agenda  ( agenda) displays all activated rules  salience  indicates priority of rules  refraction  rules fire only once for a specific set of facts  prevents infinite loops  (refresh )  reactivates rules © Franz J. Kurfess

75 75 Execution of a Program  (reset) prepares (re)start of a program:  all previous facts are deleted  initial facts are asserted  rules matching these facts are put on the agenda  (run [ ]) starts the execution  breakpoints  (set-break [ ])  stops the execution before the rule fires,  continue with ( run )  (remove-break [ ])  (show-breaks) © Franz J. Kurfess

76 76 Watching  watching the execution  (watch ) prints messages about activities concerning a  (facts, rules, activations, statistics, compilation, focus, all)  (unwatch )  turns the messages off © Franz J. Kurfess

77 77 Watching Facts, Rules and Activations  facts  assertions (add) and retractions (delete)  of facts  rules  message for each rule that is fired  activations  activated rules: matching antecedents  these rules are on the agenda © Franz J. Kurfess

78 78 More Watching...  statistics  information about the program execution  (number of rules fired, run time,... )  compilation (default)  shows information for constructs loaded by (load)  Defining deftemplate:...  Defining defrule:... +j=j  +j, =j indicates the internal structure of the compiled rules  +jjoin added  =j join shared  important for the efficiency of the Rete pattern matching network  focus  used with modules  indicates which module is currently active © Franz J. Kurfess

79 79 User Interface  menu-based version  most relevant commands are available through windows and menus  command-line interface  all commands must be entered at the prompt  (don’t forget enclosing parentheses) © Franz J. Kurfess

80 80 Limitations of CLIPS  single level rule sets  in LOOPS, you could arrange rule sets in a hierarchy, embedding one rule set inside another, etc  loose coupling of rules and objects  rules can communicate with objects via message passing  rules cannot easily be embedded in objects, as in Centaur  CLIPS has no explicit agenda mechanism  the basic control flow is forward chaining  to implement other kinds of reasoning you have to manipulate tokens in working memory [Jackson 1999] © Franz J. Kurfess

81 81 Alternatives to CLIPS  JESS  see below  Eclipse  enhanced, commercial variant of CLIPS  has same syntax as CLIPS (both are based on ART)  supports goal-driven (i.e., backwards) reasoning  has a truth maintenance facility for checking consistency  can be integrated with C++ and dBase  new extension RETE++ can generate C++ header files  not related to the (newer) IBM Eclipse environment  NEXPERT OBJECT  another rule- and object-based system  has facilities for designing graphical interfaces  has a ‘script language’ for designing user front-end  written in C, runs on many platforms, highly portable [Jackson 1999] © Franz J. Kurfess

82 82 JESS  JESS stands for Java Expert System Shell  it uses the same syntax and a large majority of the features of CLIPS  tight integration with Java  can be invoked easily from Java programs  can utilize object-oriented aspects of Java  some incompatibilities with CLIPS  COOL replaced by Java classes  a few missing constructs  more and more added as new versions of JESS are released © Franz J. Kurfess

83 83 Post-Test © Franz J. Kurfess

84 84 Evaluation  Criteria © Franz J. Kurfess

85 85 CLIPS Summary  notation  similar to Lisp, regular expressions  facts  (deftemplate), (deffacts), assert / retract  rules  (defrule...), agenda  variables, operators, functions  advanced pattern matching  input/output  (printout...), (read...), (load...)  program execution  (reset), (run), breakpoints  user interface  command line or GUI © Franz J. Kurfess

86 86 Important Concepts and Terms  agenda  antecedent  assert  backward chaining  consequent  CLIPS  expert system shell  fact  field  forward chaining  function  inference  inference mechanism  instance  If-Then rules  JESS  knowledge base  knowledge representation  pattern matching  refraction  retract  rule  rule header  salience  template  variable  wild card © Franz J. Kurfess


Download ppt "1 Computer Science Department California Polytechnic State University San Luis Obispo, CA, U.S.A. CPE/CSC 481: Knowledge-Based Systems Franz J. Kurfess."

Similar presentations


Ads by Google