A Practical Approach to Object-oriented State Modeling

A widely accepted approach in object-oriented methods is the early decomposition of the state-dependent behavior of the system between its objects and specifying it at the object or class level. In the flexible state modeling approach presented in this paper, analysis avoids strong mapping of statecharts to objects. The state-dependent behavior of the system is specified by a number of statecharts that can be at different levels of abstraction and detail. A statechart can be associated with a class or an object, but can also be of more ‘floating nature’, capturing aspects of the behavior that cut across objects and classes. Later in design, statecharts are refined by recognizing the two dimensions of how statecharts are integrated with objects, namely type (active or passive) and scope (shared, per class or per object). These are first illustrated by general examples and later by more specific examples. Pieces of source code are also provided to clarify some aspects. The flexible state modeling approach is more practical than the common approach because it enables a comprehensible specification of the state-dependent behavior without requiring object interactions to be specified for the purpose of accessing state information. ©1997 by John Wiley & Sons, Ltd.     

Cooperating/distributed grammar systems with hypothesis languages

Abstract

Motivated by the blackboard model of artificial intelligence we introduce the concept of context-free cooperating/distributed grammar systems with hypothesis languages. We prove that these grammar systems have the same generative power as context-sensitive grammars.     

Detection of Quality Changes in Freeze-Dried Beef by Measurement of the Sorption Isobar Hysteresis

SUMMARY— Experiments with cooked, freeze-dried beef support previous evidence that changes in its physical and chemical properties are reflected in its water sorption characteristics. As a new approach, sorption isobars and their hysteresis were studied. In the reported temperature and pressure range, a decrease in the sorption capacity and in the area of the hysteresis loop was observed after 1 yr of storage.     

A Local Learning Algorithm for Dynamic Feedforward and Recurrent Networks

Most known learning algorithms for dynamic neural networks in non-stationary environments need global computations to perform credit assignment. These algorithms either are not local in time or not local in space. Those algorithms which are local in both time and space usually cannot deal sensibly with ‘hidden units’. In contrast, as far as we can judge, learning rules in biological systems with many ‘hidden units’ are local in both space and time. In this paper we propose a parallel on-line learning algorithms which performs local computations only, yet still is designed to deal with hidden units and with units whose past activations are ‘hidden in time’. The approach is inspired by Holland's idea of the bucket brigade for classifier systems, which is transformed to run on a neural network with fixed topology. The result is a feedforward or recurrent ‘neural’ dissipative system which is consuming ‘weight-substance’ and permanently trying to distribute this substance onto its connections in an appropriate way. Simple experiments demonstrating the feasibility of the algorithm are reported.