Semi-Quantitative Comparative Analysis And ItsApplication

Semi-Quantitative Comparative Analysis And ItsApplication,Ivayla Vatcheva,Hidde de Jong

Semi-Quantitative Comparative Analysis And ItsApplication   (Citations: 1)
BibTex | RIS | RefWorks Download
SQCA is an implemented technique for the semi- quantitativecomparative analysisof dynamical sys- tems . It isboth able to deal with incompletelyspeci- fiedmodels and make precisepredictionsby exploiting semi-quantitativeinformationin the form of numeri- calbounds on the variablesand functionsoccuring in the models . The techniquehas a solidmathematical foundation which facilitates proofsof correctnessand convergenceproperties.SQCA representsthecoreof a method forthe automated predictionof experimental results . Raalte's CEC* is likelyto generate a large number of possiblecomparative behaviors. Besides these ambigu- ities,due to the qualitativenature of the availablein- formation, itonly characterizesdifferencesas higher or lower,without giving an indicationof theirmagnitude . In thispaper we introduce SQCA, a technique which arrivesat more preciseconclusions than qualitativeCA techniques, while retaining their abilityto deal with incomplete information. The technique exploits semi- quantitative information about the systems, in the form of numerical bounds on the variablesand functionsoc- curring in the models . Although SQCA will be pre- sented as a self-containedtechnique, itcan also be inte- grated as a filteron comparative behaviors into a qual- itativeCA algorithm . The implementation of SQCA has been used to answer CA questions involving struc- turaldifferencesin combination with differencesin the initialconditions of the systems . SQCA forms the core of a method forthe automated predictionsof experimental resultsthat iscurrentlybe- ing developed . The approach predicts an estimated value of an unperformed measurement by exploiting availableknowledge about experiments already carried out. To this end it uses an integration of techniques from the fieldof automated modeling and rVasoning about physical systems . The presentation startswith a briefreview of semi- quantitative simulation, since semi-quantitative mod- els and behaviors form the input of SQCA . Semi- quantitative CA is basicallya constraint propagation process. The next section describes how the requisite constraintsare derivablefrom the models and behaviors of the systems to be compared . The SQCA algorithm isthen presented, together with guarantees on itscor- rectnessand convergence . In the following sectionsthe resultsobtained by means of SQCA are given, followed by a short descriptionofthe approach to the automated prediction of experimental results. A briefdiscussion and ideas forfurtherwork conclude this article.
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.
Sort by: