/
Generalized Modus Ponens This rule allows us to derive an implication Generalized Modus Ponens This rule allows us to derive an implication

Generalized Modus Ponens This rule allows us to derive an implication - PDF document

cheryl-pisano
cheryl-pisano . @cheryl-pisano
Follow
478 views
Uploaded On 2014-12-27

Generalized Modus Ponens This rule allows us to derive an implication - PPT Presentation

True implies p and p and p implies q p i1 and p i1 p implies q allows and a and a implies p and p and p implies qp a p i1 and p i1 p and and a and a implies q A benefit of this approach is that the reasoner now can have a sin ID: 29998

True implies and

Share:

Link:

Embed:

Download Presentation from below link

Download Pdf The PPT/PDF document "Generalized Modus Ponens This rule allow..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.


Presentation Transcript

An Example:Converting Sentences into Clause Form 1.isHoosier( Eugene ) likes( x, Basketball ) likes( y, Basketball ) likes( x, Basketball ) likes( x, March )5.daughter( Ellen, Eugene ) childOf( x, y ) 1.isHoosier( Eugene ) isHoosier(x) likes( x, Basketball )3.[ [ likes( y, Basketball ) ] likes( x, Basketball )4.[ likes( x, Basketball ) ] likes( x, March )5.daughter( Ellen, Eugene )6.[ logic, using only the connectives and Any sentence that uses can be convertedusing the rule:(p q) == ( p) Any sentence that uses can be rewritten asform. If only we had an inference rule that worked on connective lead us to a new inference rule: lead us to a new inference rule: disjunction or disjunction disjunction because it the case analysis that is required whenever one and another asserts 1.isHoosier( Eugene ) isHoosier(x) likes( x, Basketball )3.[ childOf( x, y ) ] [ likes( y, Basketball ) ] likes( x, Basketball )4.[ likes( x, Basketball ) ] likes( x, March )5.daughter( Ellen, Eugene )6.[ as our inference rule, we conclude:1.Assume that our goal sentence is false. likes( Ellen, March )2.Try to show that the goal sentence being false causes false. Everything else disjunction disjunction1.isHoosier( Eugene )isHoosier( Eugene )not isHoosier(x) ] likes( x, Basketball )3.[ childOf( x, y ) ] childOf( x, y ) ] not likes( y, Basketball ) ] likes( x, Basketball )4.[ likes( x, Basketball ) ] likes( x, March )5.daughter( Ellen, Eugene )6.[ 7.[ likes( Ellen, March ) ]8.[ likes( Ellen, Basketball ) ]9.[ childOf( Ellen, y ) ] childOf( Ellen, y ) ] not likes( y, Basketball ) ]A.[ daughter( Ellen, y ) ] daughter( Ellen, y ) ] not likes( y, Basketball ) ]B.[ likes( Eugene, Basketball ) ]C.[ isHoosier( Eugene ) ] like March cause a like March. with modus ponens. TueThu( AI, 2:30 PM ) TueThu( AI, 2:30 PM ) and MonWedFri( AI, 2:30 PM ) and 1.TueThu( AI, 2:30 PM ) MonWedFri( AI, 2:30 PM ) MonWedFri( AI, 2:30 PM ) TueThu( AI, 2:30 PM ) TueThu( AI, 2:30 PM ) busy( Thursday, 2:30 PM ) busy( Thursday, 2:30 PM ) MonWedFri( AI, 2:30 PM ) MonWedFri( AI, 2:30 PM ) busy( Friday, 2:30 PM ) 4.busy( Thursday, 2:30 PM )5.busy( Friday, 2:30 PM ) conflict( AI ) TueThu( AI, 2:30 PM ) TueThu( AI, 2:30 PM ) busy( Thursday, 2:30 PM )8.MonWedFri( AI, 2:30 PM ) or or busy( Thursday, 2:30 PM ) 9.MonWedFri( AI, 2:30 PM )MonWedFri( AI, 2:30 PM ) busy( Friday, 2:30 PM ) B.conflict( AI ) childOf( x, y ) likes( y, Basketball ) likes( y, Basketball ) not childOf( x, y ) ]or[ not likes( y, Basketball ) More generally,p1 ... and p ) or (not p) ... or (not p) or q So, Horn clause form is incomplete. But, if we allowand negations. If we find one, then we will be able toWhy? Because we can use the semantics of implication likes( y, Basketball ) likes( y, Basketball ) childOf( x, y ) likes( y, Basketball ) to [2, 4] and [3, 5] to derive cannot take us any farther, despite cannot figure that out! See:implies( So, Horn clause form is incomplete. That is, we cannot breaks down ifMon/Wed/Fri or Tue/Thu. The AI course meets at 2:30 isCrazy( Eugene ) hoosier( Eugene ) isCrazy( Eugene ) isCrazy( Eugene ) hoosier( Eugene ) isHoosier( Eugene ) isCrazy( Eugene ) is sound and complete. It derives only works only for knowledge bases that implications of positive literals. as the inference rule. It is simple to program and Generalized Modus PonensThis rule allows us to derive an implication...True p 1 1i+11 p 1 q(phave a single goal. To derive q from (p q), use generalized modus ponens to