================================ VOLUME 20 NUMBER 4 DECEMBER 1996 ================================ 1. Distributed Shared Memory on Loosely Coupled Systems Vicente Cholvi-Juan, Department of Computer Science, University Jaume I, Campus Penyeta Roja, Castello, Spain, vcholvi@inf.uji.es AND Roy Campbell, Department of Computer Science, University of Illinois at Urbana-Champaign, 1304 W. Springfield Av, Urbana, IL 61801, roy@cs.uiuc.edu pp. 419-428 Keywords: Distributed systems, distributed shared memory, concurrency, operating systems Abstract: The distributed shared memory model DSMM is considered a feasible alternative to the traditional communication model CM, especially in loosely coupled distributed systems. While the CM is usually considered a low-level model, the DSMM provides a shared address space that can be used in the same way as local memory. This paper provides a taxonomy of distributed shared memory systems, focusing on different implementations and the factors which affect the behavior of those implementations. ------------------- 2. Human Adaptation to Qualitatively Novel Environment: The role of information and knowledge in developing countries A. Dreimanis, Environmental State Inspectorate, 25 Rupniecibas Str., LV-1877 Riga, Latvia pp. 429-434 Keywords: Developing countries, human adaptation, information, knowledge, novel environment Abstract: A systemic analysis of information and knowledge functions in human adaptation to qualitatively novel environment is proposed. The term "environment", being treated in line of C. Popper's and J. Eccle's concept of human's three worlds - the set of: 1) physical and 2) mental objects and states as well as that of 3) mental products - would include a multitude of various economical-material, social, cultural and psychological conditions. Knowledge and information - the necessary factors, in order humans and society could develop and elevate their internal variety. Acquired and transferred knowledge - efficient source of proper adaptation and harmonization in qualitatively novel environment. Capabilities of a developing system to be in harmony with changing environment will depend on mutual interrelations between information adaptation and self-creativity, and, in particularly, on creative use of available knowledge and information. ------------------- 3. On the Performance of Back-Propagation Networks in Econometric Analysis Montserrat Guillen and Carlos Soldevilla, Dept. Econometria, Estadistica i Economia Espanyola, Universidad de Barcelona, Tte. Cor. Valenzuela, 1-11. 08034 Barcelona, Spain, Phone: (343) 402 14 09, Fax: (343) 402 18 21, E-mail: guillen@riscd2.eco.ub.es, csolde@riscd2.eco.ub.es pp. 435-441 Keywords: Neural networks, discriminant analysis, logistic regression, time series Abstract: Neural networks may be applied in the context of econometric analysis, both when discussing issues that have traditionally been attached to multivariate analysis and in the field of time series. This paper compares the performance of backpropagation networks with classical approaches. Firstly, an example in banking is presented. The network outperforms discriminant analysis and logistic regression when conditional classificafion error is considered. Secondly, the identification of simple stationary time series is analyzed. Some series following simple autoregressive, moving average schemes were simulated, and the network successfully identified them. Conclusions are presented in the closing section. ------------------- 4. The Challenge of Integrating Knowledge Representation and Databases Paolo Bresciani, Knowledge Representation and Reasoning group, Istituto per la Ricerca Scientifica e Tecnologica via Sommarive 18, I-38050 Povo (Trento), Italy, Email: bresciani@irst.itc.it pp. 443-453 Keywords: Artificial intelligence, knowledge representation, description logics, databases Abstract: Two different aspects of data management are addressed by Knowledge Representation (KR) and Databases (DB): the semantic organization of data and powerful reasoning services by KR, and their efficient management and access by DB. It is recently emerging that experiences from both KR and DB should profitably cross-fertilize each other, and a great interest is rising about this topic. In particular, among several ways to approach knowledge representation, Description Logics (DL) are gaining, in the last years, a privileged place. In this paper, after briefly showing the importance of an integrated view of description logics and databases, our approach to this topic is presented. Our technique allows uniform access - by means of a DL-based query language - to information distributed over knowledge bases and databases. The separately existing retrieving functions of description logics management systems and of database management systems are integrated, in our extended paradigm, in order to allow, via a query language grounded on a DL-based schema knowledge, uniformly formulating and answering queries and, thus, uniform retrieval from mixed knowledge/data bases. ------------------- 5. Rough Sets: Facts Versus Misconceptions Jerzy W. Grzymala-Busse, Department of Computer Science and Electrical Engineering, University of Kansas, Lawrence, KS 66045, USA, E-mail: Jerzy@eecs.ukans.edu AND Jerzy Stefanowski, Institute of Computing Science, Poznan University of Technology 60-965 Poznan, Poland, E-mail: Stefanj@pozn1v.put.poznan.pl AND Wojciech Ziarko, Department of Computer Science, University of Regina, Regina, SK, Canada S4S 0A2, E-mail: Ziarko@cs.uregina.ca pp. 455-464 Keywords: Rough set theory, machine learning Abstract: This article is a response to the critical assessment of the role of the methodology of rough sets in Machine Learning as presented by Kononenko and Zorc in their paper Critical Analysis of Rough Sets Approach to Machine Learning, Informatica 18, pp. 305-313. We correct the inaccuracies and respond to unfounded claims made by the authors while trying to present the rough set theory in the broader context of scientific and engineering methodologies and practical applications utilizing the theory of rough sets as their basic theoretical paradigm. ------------------- 6. On Facts Versus Misconceptions about Rough Sets Igor Kononenko, University of Ljubljana, Faculty of Computer and Information Science Trzaska 25, SI-1000 Ljubljana, Slovenia, E-mail: igor.kononenko@fri.uni-lj.si pp. 465-468 Keywords: Rough set theory, critical analysis, machine learning Abstract: This note is a response to the paper Rough Sets: Facts Versus Misconceptions by J.Grzymala-Busse, J.Stefanowski and W. Ziarko, Informatica, this volume, which is in turn the response to the paper (Kononenko and Zorc, 1994). I clarify some points from our original paper that were mistakenly interpreted by Grzymala-Busse et al. and stress points from our original paper that were ignored by Grzymala-Busse et al. I conclude that with additions to the Rough Sets theory one can achieve good performance, which is however not due to Rough Sets but due to the additions, and that the use of Rough Sets is an unnecessary burden for machine learning algorithms. ------------------- 7. Application of Neural Networks to Nuclear Power Plants in Korea Se Woo Cheon, Korea Atomic Energy Research Institute, Yu Song P.O. Box 105, Taejon 305-600, Korea, Phone: +82 42 868 8869, Fax: +82 42 868 8357, E-mail: swcheon@nanum.kaeri.re.kr pp. 469-474 Keywords: Overview, neural networks, nuclear power plants, in Korea Abstract: This paper introduces an overview of neural network applications for nuclear power plants in Korea. Neural networks have been applied to various fields of nuclear power plants since early 1990s. The fields of applications can be categorized into two areas. One is plant diagnostics based on the pattern recognition of input symptoms, which includes transient identification and multiple alarm diagnosis. The other is modeling of systems with interpretation of input-output relationships, which includes the prediction of plant parameters, the prediction of signal trend, and signal validation. Most neural networks were based on the backpropagation network model.