M. Paramasivam and David A. Plaisted. Automated deduction techniques for classification in description logic systems. Journal of Automated Reasoning, to appear, 1997.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....is currently happening in the case of planning) Yet the logical approach is still not entering the engineering practice because it is believed to lack efficiency in comparison with specialized systems. Eventually, someone in automated deduction disproves this belief 19 experimentally (see e.g. PP97] in the case of KR) Nevertheless the well established tools persist in applications for obvious reasons (resistence against transition costs and such) Logicists deplore the time lost by the multiple development of the same techniques in many disguises and dream of a state of the art where ....

M. Paramasivam and David A. Plaisted. Automated deduction techniques for classification in description logic systems. Journal of Automated Reasoning, to appear, 1997.

....finite satisfiability is achieved by generating models with minimal universes. This notion of model minimality , which can be called domain or universe minimality , is different and complementary to that investigated in the present article. For many applications such as those addressed in [54, 61, 53, 6, 7, 20, 51, 2] both notions of minimality, on the one hand domain minimality, on the other hand minimality of the set of satisfied ground atoms, are needed. In [34, 42] a tableaux method is defined for first order logic formulas which generates models with minimal universes by relying on so called ghost ....

M. Paramasivam and D. Plaisted. Automated Deduction Techniques for Classification in Description Logic Systems. Jour. of Automated Reasoning, 20(3), 1998.

....is currently happening in the case of planning) Yet the logical approach is still not entering the engineering practice because it is believed to lack efficiency in comparison with specialized systems. Eventually, someone in automated deduction disproves this belief 19 experimentally (see e.g. PP97] in the case of KR) Nevertheless the well established tools persist in applications for obvious reasons (resistence against transition costs and such) Logicists deplore the time lost by the multiple development of the same techniques in many disguises and dream of a state of the art were all ....

M. Paramasivam and David A. Plaisted. Automated deduction techniques for classification in description logic systems. Journal of Automated Reasoning, to appear, 1997.

.... of ALC expressions [ Schmidt, 1999 ] This approach was adopted in the experiments of Hustadt, Schmidt, and Weidenbach [ 1997; 1998 ] Experiments using the standard translation and a combination of a first order theorem prover augmented with a finite model finder are described in [ Paramasivam and Plaisted, 1998 ] The problem of empirical investigations based on competitive testing is the difficulty in identifying the major factors having a positive or negative influence on the performance of a theorem prover. As long as the theorem provers which are being compared follow different proof strategies this ....

M. Paramasivam and D. A. Plaisted. Automated deduction techniques for classification in description logic systems. J. Automated Reasoning, 20(3):337--364, 1998.

.... which can be decided by resolution and condensing using any compatible ordering or selection strategy [17] This embedding was used in recent experimental work which provides evidence that resolution theorem provers can serve as reasonable and efficient inference tools for description logics [11,15]. In this paper we consider an expressive description logic, which we believe has not been considered before in the literature on description logics or modal logics. We call the logic ALB which is short for attribute language with boolean algebras on concepts and roles . ALB extends ALC with the ....

M. Paramasivam and D. A. Plaisted. Automated deduction techniques for classification in description logic systems. J. Automated Reasoning, 20:337--364, 1998.

.... The subclass, called basic path logic, can be decided by resolution and condensing using any compatible ordering or selection strategy [13] Recent experimental work provides evidence that resolution theorem provers can serve as reasonable and efficient inference tools for description logics [8, 12]. In this paper we consider an expressive description logic, which we believe has not been considered in the literature on description logics or modal logics. We call the logic ALB which is short for attribute language with boolean algebras on concepts and roles . ALB extends ALC with the top ....

M. Paramasivam and D. A. Plaisted. Automated deduction techniques for classification in description logic systems. Journal of Automated Reasoning, 20(3):337--364, 1998.

....model generator complete for finite satisfiability is needed. This is possible, since finite satisfiability is semidecidable [36] For as well detecting unsatisfiability, one can rely on a refutation prover coupled with a finite model finder such as FINDER [31] and SEM [37] as described e.g. in [30]. The present paper proposes instead a single deduction method complete for both, unsatisfiability and finite satisfiability. Arguably, this method is more convenient for many applications. In particular, a single method is better amenable to user interaction and explanation as provided by the ....

....this method is more convenient for many applications. In particular, a single method is better amenable to user interaction and explanation as provided by the system [8, 9] For some applications e.g. natural language understanding a single method is necessary [14] and the coupling approach of [30] is not applicable. Note that the applications mentioned here in general do not give hints for the size of the finite models seeked for. Note also that most applications require that the model generator constructs only minimal models in the sense of e.g. 11] This related issue is beyond the ....

[Article contains additional citation context not shown here]

M. Paramasivam and D. Plaisted. Automated Deduction Techniques for Classification in Description Logic Systems. J. of Automated Reasoning, 20(3), 1998.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC