The research ofnonï¼classical lope has always been a very attractive dlrecï¼ non of Artificial Intelligence.In recent years,theory and technology of automated theorenlprovinghased on nonï¼classlcalloglcs are developingrapldlyand many automated theorem provers have been tised successfully In matllelnatlcs,engineerï¼ lug and other areas.Used Previous Internal and external outcome ofresearches In these direction for reference。the allthor stlldled the properties oflattlce Inlï¼ l)licatlon algebra and latticeï¼vallled logical systelll wltll trllthï¼valllesfield belllg a lattice lmpllcatlon algebra,andon the base ofwhich,the authordlscussed In deï¼ tails about the latticeï¼valued resolution principle oflattlceï¼valued logical system. The specific contents are s follows: Part One The Study of Lattice Implication Algebra On the basis of previous results of lattice implication algebra.the part(:onï¼ sists of*theili*lliereallwffollilli**ã€ï¼¾,000taltal* fOllowingthree points: l.A lllethod ofconlputingthe colllplelllentary elemellt ofally one In alattlce implication algebra was discussed In details and a practicable approach was given. 2.The theorem was shown that lattice lmpllcatlon algebra can not be conï¼ Stl·lit:tAtl Off CAlt3lll SI>eC181 dlstl'lhlltlV618tt1Cesï¼›Sllf:h s《5.心6,心7,令8 3lidO8+k. 3.The structural feature ofnonï¼chainï¼type and nonï¼Booleanï¼algebratype lattice lmpllcatlon algebra was discussed and the result that ifa lattice implication algebrahas alld only has onedual atom then It Is a chain was proved. Part Two The Study of ReResolution Principle in L8tticeï¼vå¤ined Logic based on Lattice Implication Algebra Ill this P8ftï¼›th6 3llthof dï¼SCilSS6d th6 f6SOIï¼»ltlOllï¼b8S6d Ih6thod Of slltoinst6d f6Anolllllg.ThiS psft COliSIStS Ofth6 follOWlllg poilltS: 1.The semantical andsyntactlcalfeatures ofresolutlonï¼basedautomated reasoningln classical logic was analyzed.The difference on resolutionï¼basediv Jgmethod between the classical logic and lattice-valued logic was discussed in details.2. A algebraical method about resolution-based automated reasoning in classical logic was given. That method translated the decidability of satisfiability of logical formula into the decidability of solution of congruence equation array.3. The resolution-based automated reasoning method of simple and complicate clause sets in LP(X) was discussed in details. The concept of resolution on filter was given and a practicable approach was attained. The soundness and weak completeness theorem of that method was proved.4. The concept and method of resolution on filter in lattice- valued first-order logic SLF(X] was discussed and the concepts and properties of Skolemlzation and Herbrand field in it was discussed too.5. The concept of constrained resolution was give after the analyzing of resolution-based method of classical logic in lattice-valued logic.Part Three The Study of Lattice- Valued Logical SystemIn this part, the author constructed a lattice- valued prepositional logical system Jz?,,p with the truth-field of it is a lattice implication algebra. This part consists of the following points:1 . The concepts and properties of (a, /3)-semantical implication were discussed; the concepts and condition of keeping (a, /3)-rule of lattice- valued logical formula were given too.2. The condition of choosing axioms and inference rule in ^fvp was analyzed.3. The concepts of proof under valuation and its properties were given. The soundness and weak deduction theorems were proved.4. The concepts of consistency of lattice-valued logical formula under valuation was given and the consistent theorem of it was proved also.
|