Fuzzy-neuro risk assessment of distributed real-time systems through formal and object-oriented systems modelling
نویسنده
چکیده
The purpose of this paper is to advocate and encourage the application of fuzzy-neuro algorithms in the risk assessment of Distributed Real-Time (DRT) systems, where object-oriented and formal techniques by means of UML-SDL models are used as knowledge (data) providers to the fuzzy-neuro analyser. The paper includes an introductory description of DRT systems and the presentation of the UML and SDL techniques as software modelling languages, along with their advantages and limitations. Next, a combination of UML and SDL for requirement and detailed design modelling is suggested to produce the proper knowledge bases to a risk analyser. An example of a fuzzy-neuro system is then utilised to illustrate how it can be trained to assess various kinds of risks by using the combined UML-SDL models as input and through deductive reasoning.
منابع مشابه
Modeling and Evaluation of Stochastic Discrete-Event Systems with RayLang Formalism
In recent years, formal methods have been used as an important tool for performance evaluation and verification of a wide range of systems. In the view points of engineers and practitioners, however, there are still some major difficulties in using formal methods. In this paper, we introduce a new formal modeling language to fill the gaps between object-oriented programming languages (OOPLs) us...
متن کاملModeling and Evaluation of Stochastic Discrete-Event Systems with RayLang Formalism
In recent years, formal methods have been used as an important tool for performance evaluation and verification of a wide range of systems. In the view points of engineers and practitioners, however, there are still some major difficulties in using formal methods. In this paper, we introduce a new formal modeling language to fill the gaps between object-oriented programming languages (OOPLs) us...
متن کاملA NEURO-FUZZY GRAPHIC OBJECT CLASSIFIER WITH MODIFIED DISTANCE MEASURE ESTIMATOR
The paper analyses issues leading to errors in graphic object classifiers. Thedistance measures suggested in literature and used as a basis in traditional, fuzzy, andNeuro-Fuzzy classifiers are found to be not suitable for classification of non-stylized orfuzzy objects in which the features of classes are much more difficult to recognize becauseof significant uncertainties in their location and...
متن کاملساختار دهی آنی دادههای مکانی ورودی GIS با تأکید بر عارضه راه
An important issue in implementation of a GIS system is preparation of data to be entered in GIS. To produce spatial data for GIS using photogrammetric techniques, conventional method is to apply photogrammetric and GIS systems individually (off-line procedure). This approach is costly, time consuming and somehow unreliable due to the fact that 3D photogrammetric model is not available at the ...
متن کاملA Fuzzy Expert System & Neuro-Fuzzy System Using Soft Computing For Gestational Diabetes Mellitus Diagnosis
Gestational diabetes mellitus (GDM) is a kind of diabetes that requires persistent medical care in patient self management education to prevent acute complications. One of the common and main problems in diagnosis of the diabetes is the weakness in its initial stages of the illness. This paper intends to propose an expert system in order to diagnose the risk of GDM by using FIS model. The knowl...
متن کامل