intelligent rule system alcirs
play

Intelligent Rule System ALCIRS Research Skills The Presentation - PowerPoint PPT Presentation

Adaptive Loose Coupling Intelligent Rule System ALCIRS Research Skills The Presentation Presented by: Irfan Subakti 1054257 Supervisor: Prof. John A. Barnden School of Computer Science University of Birmingham United Kingdom 25 January


  1. Adaptive Loose Coupling Intelligent Rule System ALCIRS Research Skills – The Presentation Presented by: Irfan Subakti – 1054257 Supervisor: Prof. John A. Barnden School of Computer Science University of Birmingham United Kingdom 25 January 2012 Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 1

  2. Ideas  Adaptive  Able to adjust to another type of situation  Loose coupling  Overcome rule dependency changing problem  Intelligent  Learn for improving its rule semantic understanding, rule learning & generating Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 2

  3. Motivation  Variable-Centered Intelligent Rule System (VCIRS) - Subakti (2005, 2006, 2007)  Monotonically increasing Rule Based (RB) as time goes by  Tight coupling  inflexibility in rule changing  Too simple rule generating  need more creativity  Blackboard Systems – Erman et al. (1980), Corkill (1991)  Dealing with complex applications which are roughly defined  flexible in representation & in contributing problem solving  Disadvantage: formal specification  Contextual Ontologies – Benslimane et al. (2006)  A concept’s set of properties is vary depend on a context  Semantic Understanding – Shih et al. (2011)  Capturing the interpretation of the behaviours and situations Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 3

  4. Motivation (cont’d)  Robust Growing Neural Gas (RGNG) – Kin and Suganthan (2004)  Robust properties in clustering  Outlier resistance  Adaptive modulation of learning rates  Cluster repulsion  Insensitivity  Initialization  Input sequence ordering  Outlier presence  Particle Swarm Optimization (PSO) - Kennedy and Eberhart (2001)  Simple idea with outstanding result in optimisation  Creativity in Reasoning – Indurkhya (1997)  New categories & interpretation can be created in legal reasoning Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 4

  5. Basic Component  Blackboard model Knowledge Sources Blackboard (Corkill, 1991) Control Component Rule Based  ALCIRS Rules Interface Implement Ontology Inference Engine Contextual Ontology Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 5

  6. Basic Framework  PSO (Kennedy and Eberhart, 2001 )  RGNG (Kin and Suganthan, 2004)  Rule generating  creativity (Indurkhya, 1997) Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 6

  7. Basic Framework (cont’ed)  Basic BS Blackboard Executing Library of KS Activation KSs (Corkill, 1991) Events Control Pending Components KS Activations Rule Based Inference Engine Rules PSORGNG - clustering Interface  Basic ALCIRS Adaptive Implement Ontology Rule dependency Loose Coupling Rule learning, reasoning, generating Intelligence Contextual Ontology Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 7

  8. Adaptive - Methodology  Adaptive  Raw data  Rules will be clustered in proper place, using PSORGNG  Existing rules in Rule Base (RB)  Interface & Implement parts will be classified, supported by Contextual Ontology  Generating new rules  Supported by Contextual Ontology Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 8

  9. Loose Coupling - Methodology  Loose coupling  Rule dependency changing  Each rule has  Interface o A part that can be shared to other rules  global o Other rules may use a little or none of this part o Flexibility concept applied, since all rules loosely can be connected with this part o As a bridge for contextual ontologies layer  Implement o a specific part which dedicated to its rule  local  Ontology o Linked to contextual ontology  further rule learning, reasoning & generating  Core ontology  the lowest level of contextual ontology can be used as the last resort if higher contextual ontologies failed to do so Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 9

  10. Intelligence - Methodology  Intelligence  Semantic understanding  Understand the meaning of rule given a context  supported by contextual ontology  Rule learning, reasoning & generating  Contextual ontology  continually learning to optimise the usefulness of the rules  Contextual reasoning  supported by contextual ontology gives an inference based on the context  Core ontology  performing creativity in producing a new rule from the existing rules in RB given a new case Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 10

  11. Loose Coupling  Looseness definition  When a rule only uses none or little part of other rules  loose coupling mechanism  Part usages on rules  None  Explicit: direct assignment. E.g., weight = input_weight  Implicit: by using the contextual ontology  Little part  Using exactly the same term. E.g., Rule #1 uses input_weight in its Interface, while Rule #2 also uses input_weight in its Implement Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 11

  12. Case Study (1)  Supermarket goods purchasing Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 12

  13. Case Study (2)  Owning a car and a house  An example of a rule, which has  Interface  Implement  Ontology Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 13

  14. Case Study (3)  Owning a car and a house (continued)  Loose coupling rules example #1  A user starts creating a new rule #Car-house owning#  defining a  relation between owning a car and a house #Vehicle type#  defining the types  of vehicles  Loose coupling No part from #Car-house owning# is  used in #Vehicle type# Direct assignment: weight =  input_weight at #Vehicle type# Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 14

  15. Case Study (4)  Owning a car and a house (continued)  Loose coupling rules example #2  Another day, the user willing to add up a rule  #House type#  defining the types of houses  Loose coupling  Little part from #Car-house owning# is used in #House type# house_type is used in both  rules  Little part from #Vehicle type# is used in #House type# vehicle_type is used in both  rules Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 15

  16. Case Study (5)  Owning a car and a house (continued)  Loose coupling rules example #3  Then it turned another day and the user willing to add up a rule  #Garage type#  Loose coupling Little part from #Vehicle type# is  used in #Garage type# wheels is used in both rules  No part from #Car-house owning# is  used in #Garage type# Direct assignment: wheels =  input_wheels at #Vehicle type# Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 16

  17. Case Study (6)  Owning a car and a house (continued)  A whole rules in RB  Loose coupling  No part from #Car-house owning# is used in #Vehicle type# Direct assignment: weight = input_weight at  #Vehicle type#  Little part from #Car-house owning# is used in #House type# house_type is used in both rules   Little part from #Vehicle type# is used in #House type# vehicle_type is used in both rules   Little part from #Vehicle type# is used in #Garage type# wheels is used in both rules   No part from #Car-house owning# is used in #Garage type# Direct assignment: wheels = input_wheels at  #Vehicle type# Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 17

  18. Rule Generating Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 18

  19. Anticipated Result  Adaptive  Rule is clustered in certain group  establishing a rule for each formed cluster  Interface & Implement is classified in certain category  New rule can be generated for a new case  creativity  Loose coupling  Rule changing/updating in a given contextual meaning is easily performed without worry about rule dependency  Intelligent  Comprehend the meaning of rule given a context  Optimise the usefulness of the rules  Contextual reasoning  Able to perform creativity in a new rule creation Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 19

  20. Conclusion  Adaptive Loose Coupling Intelligent Rule System (ALCIRS)  A Rule-Based system  Use a specific framework which works adaptively & intelligently comparing to Blackboard Systems  Treating rule in the loose coupling manner  Rule dependency in a given context is automatically preserved  Rule updating is easily perform without worry about this dependency  Rules are clustered and classified automatically  Understanding the contextual meaning of given rule  Optimising usefulness of the rules  Contextual reasoning  Perform creativity in the rule generation Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 20

  21. Future Work  Continue reading the literature and comprehend it deeper to suit the proposed system  Implementing the framework  Using some examples Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 21

  22. Adaptive Loose Coupling Intelligent Rule System ALCIRS Thank you for your attention!  Research Skills | ALCIRS | Irfan Subakti (1054257) 25 Jan 2012 22

Recommend


More recommend