### Fuzzy Logic theory -- Jos De Roo, Gijs Muys @prefix log: <http://www.w3.org/2000/10/swap/log#>. @prefix math: <http://www.w3.org/2000/10/swap/math#>. @prefix list: <http://www.w3.org/2000/10/swap/list#>. @prefix owl: <http://www.w3.org/2002/07/owl#>. @prefix xsd: <http://www.w3.org/2001/XMLSchema#>. @prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#>. @prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>. @prefix func: <http://www.w3.org/2007/rif-builtin-function#>. @prefix e: <http://eulersharp.sourceforge.net/2003/03swap/log-rules#>. @prefix fl: <http://eulersharp.sourceforge.net/2003/03swap/fl-rules#>. ### classes fl:Lower a rdfs:Class; rdfs:comment "class of things that are the first element of e:valuation object". fl:Upper a rdfs:Class; rdfs:comment "class of things that are the last element of e:valuation object". ### properties fl:mu a rdf:Property; rdfs:domain rdf:List; rdfs:range e:Number; rdfs:comment "to express fuzzy set membership e.g. (:x :C) fl:mu 0.8 says that :x a :C to a degree of 0.8". fl:sigma a rdf:Property; rdfs:domain rdf:List; rdfs:range e:Number; rdfs:comment "to express fuzzy subsethood e.g. (:C :D) fl:sigma 0.9 says that :C rdfs:subClassOf :D to a degree of 0.9". fl:pi a rdf:Property, e:Builtin; rdfs:domain rdf:List; rdfs:range e:Number; rdfs:comment "builtin to calculate fuzzy set membership according to Elpiniki Papageorgiou method". # FGCM properties fl:gmu a rdf:Property; rdfs:domain rdf:List; rdfs:range rdf:List; rdfs:comment "to express grey membership e.g. (:x :C) fl:gmu (0.1 0.4) means that :x belongs to :C with a degree between 0.1 and 0.4. It can be any value in-between but we do not know the exact value for sure.". fl:gsigma a rdf:Property; rdfs:domain rdf:List; rdfs:range rdf:List; rdfs:comment "to express grey subsethood e.g. (:C :D) fl:gsigma (0.1 0.4) means that :C rdfs:subClassOf :D with a degree between 0.1 and 0.4. It can be any value in-between but we do not know the exact value for sure."; rdfs:comment "Sigma can also be considered as one way correlation where 0 is negative correlation and 1 is positive correlation.". fl:gpi a rdf:Property, e:Builtin; rdfs:domain rdf:List; rdfs:range rdf:List; rdfs:comment "built-in to calculate fuzzy grey inference according to Elpiniki Papageorgiou method". ### conjunction according to Lukasiewicz { ?C owl:intersectionOf ?E. ?E func:distinct-values ?U. ?U list:first ?A. ?U list:rest ?B. (?X ?A) fl:pi ?I. ?D owl:intersectionOf ?B. (?X ?D) fl:pi ?J. (?I ?J -1.0) math:sum ?K. (0.0 ?K) e:max ?M } => { (?X ?C) fl:pi ?M. }. ### disjunction according to Lukasiewicz { ?C owl:unionOf ?E. ?E func:distinct-values ?U. ?U list:first ?A. ?U list:rest ?B. (?X ?A) fl:pi ?I. ?D owl:unionOf ?B. (?X ?D) fl:pi ?J. (?I ?J) math:sum ?K. (1.0 ?K) e:min ?M. } => { (?X ?C) fl:pi ?M. }. ### negation according to Lukasiewicz { ?C owl:complementOf ?A. (?X ?A) fl:pi ?I. (1.0 ?I) math:difference ?M. } => { (?X ?C) fl:pi ?M. }. ### implication { ?A rdfs:subClassOf ?B. } => { (?A ?B) fl:sigma 1.0 . }. # Creates edge to itself (necessary calculation in algorithm) { (?B ?A) fl:sigma ?I. } => { (?A ?A) fl:sigma 1.0 . }. { ?C owl:complementOf ?A. (?B ?A) fl:sigma ?I. (1.0 ?I) math:difference ?M. } => { (?B ?C) fl:sigma ?M. }. ### Allows you to run FGCM algorithm on FCM model & ability to express non-grey nodes in FGCM model { (?B ?A) fl:sigma ?I. } => { (?B ?A) fl:gsigma (?I ?I). }. { (?B ?A) fl:mu ?I. } => { (?B ?A) fl:gmu (?I ?I). }. { (?B ?A) fl:pi ?I. } => { (?B ?A) fl:gpi (?I ?I). }. ### Fuzzy Description Logic to Fuzzy Cognitive Map translation theory { {?P a ?Q} e:valuation (?L ?U). } => { (?P (?Q fl:Lower)) fl:mu ?L. }. { {?P a ?Q} e:valuation (?L ?U). } => { (?P (?Q fl:Upper)) fl:mu ?U. }. { {?X rdfs:subClassOf ?Y} e:valuation (?L ?U). } => { ((?X fl:Lower) (?Y fl:Lower)) fl:sigma ?L. }. { {?X rdfs:subClassOf ?Y} e:valuation (?L ?U). } => { ((?X fl:Upper) (?Y fl:Upper)) fl:sigma ?U. }. { {?X owl:complementOf ?Q. ?Q rdfs:subClassOf ?Y} e:valuation (?L ?U). (1.0 ?L) math:difference ?M. } => { ((?Y fl:Lower) (?X fl:Lower)) fl:sigma ?M. }. { {?X owl:complementOf ?Q. ?Q rdfs:subClassOf ?Y} e:valuation (?L ?U). (1.0 ?U) math:difference ?M. } => { ((?Y fl:Upper) (?X fl:Upper)) fl:sigma ?M. }. { (?P (?C fl:Lower)) fl:pi ?X. (?P (?C fl:Upper)) fl:pi ?Y. ?Z e:optional { ?X math:notGreaterThan ?Y. ?S log:equalTo (?X ?Y). }. ?Z e:optional { ?X math:greaterThan ?Y. ?S log:equalTo (?Y ?X). }. } => { {?P a ?C} e:valuation ?S. }. ### support owl:Thing owl:intersectionOf (). owl:Nothing owl:unionOf (). { ?C owl:intersectionOf ?U. ?U list:rest ?B. } => { _:D owl:intersectionOf ?B. }. { ?C owl:unionOf ?U. ?U list:rest ?B. } => { _:D owl:unionOf ?B. }. { (?X ?A) fl:mu ?I. } => { (?X owl:Thing) fl:mu 1.0 . }. { (?X ?A) fl:mu ?I. } => { (?X owl:Nothing) fl:mu 0.0 . }.