Transformational approach to the development of software architectures on the basis of fuzzy graph models

Abstract

An approach to the creation of software architectures is examined that is based on the model-oriented paradigm that opens up the way to the implementation of transformational evolution of software systems. The emphasis is on the formalization of fuzzy graphs, fuzzy graph models, and rules of their transformation. A theoretical categorical representation of fuzzy software architectures is developed, which allows one to manage the process of their transformations on the basis of making decisions in a fuzzy modeling space on the functionality of the target platform-specific system.

Category-theoretic formalization in designing fuzzy intelligent systems

“…The category-theoretic approach can be used to formalize FGTSs, which are a generalization of sequential GTS and take into account the main forms of fuzziness that occur both in constructing basis categories of fuzzy objects and in describing the transformations of fuzzy graphs generated by fuzzy sets [7][8][9]. To this end, we will use the "pushout" structure, which is a category generalization of the union of two objects with allowance for the common subobject.…”

“…The weights and estimate of each parameter are specified by fuzzy numbers or words; therefore, the fuzzy multiattribute decision-making method [8] is implemented.…”

Transformational Technique of Interactive Systems Development

“…In order to increase the efficiency of a development process and to attain high qua1ity of these systems we need the appropriate technological tools. Among such technological tools transformational technique of software development seems to be very useful [1,2] . This technique provides a possibility of computer-assisted transformation of programs according to given transformation rules to generate new programs or to make reduction (optimizing) of existing ones.…”

A Unifying Theory for Graph Transformation

“…To the best of our knowledge, Mori and Kawahara were the rst to propose this [MK95], using a single pushout construction. There also exists a series of papers by Parasyuk and Yershov, in which fuzzy graph transformations are studied using single pushouts [PY06] and double pushouts [PE07,PY08]. Because PBPO + rules have both a pattern span and a type span (unlike in the single and double pushout approaches), both lower and upper bounds can be speci ed on fuzzy values.…”