873 publications from this institution
Artificial intelligence (Al) has received an explosion of interest during the last five years in various fields. There is no longer any question that expert systems and neural networks will be of central importance for developing the next generation of more intelligent geographic information systems. Such knowledge based geographic information systems will especially play a key role in spatial decision and policy analysis related to issues such as environmental monitoring and management, land use planning, motor vehicle navigation and distribution logistics. This paper sketches briefly the major characteristics of conventional geographic information systems, and then looks at some of the potentials of Al principles and techniques in a GIS environment where emphasis is laid on expert systems and artificial neural networks technologies and techniques.
Veronica lycica Lehm. is a distinct, diploid member of theVeronica cymbalaria group, endemic in Lycia (S.W. Anatolia). Closely related is the newV. stamati
No abstract is provided for this article.
Nach klinischer und laborchemischer Diagnose einer endokrinen Störung im Bereich der Nebennierenrinde oder ihrer Beteiligung an einer Endokrinopathie steht der behandelnde Arzt vor dem Problem der Aufdeckung der Ätiologie der abnormalen Hormonproduktion und...
Building a feedforward computational neural network model (CNN) involves two distinct tasks: determination of the network topology and weight estimation. The specification of a problem adequate network topology is a key issue and the primary focus of this contribution. Up to now, this issue has been either completely neglected in spatial application domains, or tackled by search heuristics (see Fischer and Gopal 1994). With the view of modelling interactions over geographic space, this paper considers this problem as a global optimization problem and proposes a novel approach that embeds backpropagation learning into the evolutionary paradigm of genetic algorithms. This is accomplished by interweaving a genetic search for finding an optimal CNN topology with gradient-based backpropagation learning for determining the network parameters. Thus, the model builder will be relieved of the burden of identifying appropriate CNN-topologies that will allow a problem to be solved with simple, but powerful learning mechanisms, such as backpropagation of gradient descent errors. The approach has been applied to the family of three inputs, single hidden layer, single output feedforward CNN models using interregional telecommunication traffic data for Austria, to illustrate its performance and to evaluate its robustness.
The paper emphasises the importance of a research programme focused on developing and making widely available GIS relevant spatial analysis technology. It outlines generic criteria able to discriminate between GIS-relevant and GIS-irrelevant spatial analysis tools and outlines a list of six researchable spatial analysis themes. It is argued that presently there is an opportunity to develop a EU based spatial analysis research programme and then install the technology in the World's GIS. (authors' abstract)
This article is the position paper for the ESF-GISDATA Specialist Meeting on GIS & Spatial Analysis, Amsterdam, 1-5 December1993. The focus here is on the two major themes of the meeting: Spatial Data Analysis and Spatial Modelling. Special emphasis is laid on specific problems and possibilities for interfacing spatial analysis tools (i.e. spatial data analysis techniques and spatial models) and GIS. Both GIS application fields, the environmental sciences and the social sciences, are taken into consideration.
In this paper, we distinguish three constrained variants of the gravity model of spatial interaction: doubly constrained, production constrained and attrac
BACKGROUND There is a clinical need for therapeutic alternative in patients with persisting painful arthritis of AC-joint and failure of previous treatments. However, no radiopharmaceutical is currently explicitly approved for radiosynoviorthesis of acromioclavicular joint. The aim of our study was to prospectively assess the efficacy and safety of radiosynoviorthesis of acromioclavicular joint using erbium-169 citrate. MATERIAL AND METHODS Radiosynoviorthesis of acromioclavicular joint was performed in 51 consecutive patients (18 males, 33 females) mean age 64.3 (range 43.8–82.6, median 63.6) years with clinically confirmed arthritis of 85 acromioclavicular joints. The efficacy of RSO was reported by patients according to 10-step visual analogue scale of pain (VAS) (0 = no pain, 10 = most severe pain) at 6 months after radiosynoviorthesis and by ranking the global therapeutic effect of RSO in 4 categories (1 = the best effect, 4 = no change). To assess the variation of blood perfusion in treated joints, the efficacy of RSO was also evaluated by variation of target (acromioclavicular joint)/non-target (soft tissue) uptake ratio (T/NTR) of metylendiphosphonate ( 99m Tc) measured as number of counts over region of interest on blood pool phase of two-phase bone scintigraphy performed before and 6 months after RSO. RESULTS Radiosynoviorthesis was followed by significant decrease in VAS, mean — 3.1 (-47%). Excellent, good, moderate and bad response was observed in 57 (67%), 25 (29%), 1 (1%) and in 2 (2%) of acromioclavicular joints respectively. A significant correlation between decrease of T/NTR and variation of VAS in % (ρ = 0.532, p < 0.0001) and between T/NTR and subjective evaluation of therapeutic effect in scale 1–4 (ρ = 0.388, p = 0.0002) was observed. However, it was not possible to identify the cut-off value of relative decrease in T/NTR showing sufficient sensitivity and specificity to detect the therapeutic response. CONCLUSION Results of this prospective study permit to conclude a good efficacy and safety of radiosynoviorthesis using erbium-169 citrate in a series of patients with arthritis of acromioclavicular joint in whom previous line(s) of treatment did not lead to satisfactory pain relief.
No abstract is provided for this article.
The paper is concerned with the impact of market research prior to integration with European Union (EU) on the structures of noncompetitive industries in integrated economy. The analysis focuses on monopolistic markets with stochastic demand. Firms are considered in dynamic multiperiod model, where intertemporal links are determined by expenditures on market research in a present period and benefits from this activity (i.e., smaller variance of the prediction error) in the future. We show that the optimal market research strategy is stationary and depends on market size. Consequently, after accession firms operating prior to integration in small markets are expected to have much less information about the total market than their competitors from the EU. This informational asymmetry may affect the structure of the industry in integrated economy. In the extreme case, the firm operating before integration in the small market can be ruled out from the integrated market.
Die Verwendung von Radionukliden zur Therapie von benignen oder malignen Erkrankungen ist mit einer Reihe gesetzlicher Vorschriften, Richtlinien oder Leitlinien verbunden, die für den Nuklearmediziner seine Verantwortlichkeit bei der Patientenbehandlung definieren. Diese beziehen sich auf den Schutz der Allgemeinheit, des Patienten sowie die Sicherheit des Arztes und seines Personals. Die rechtliche Wertigkeit der Vorschriften ist unterschiedlich, sollte aber dem Nuklearmediziner bekannt sein und im geforderten Maß beachtet werden.
This paper presents a theoretical growth model that extends the Mankiw–Romer–Weil (MRW) model by accounting for technological interdependence a
In this contributon we evaluate on-line and off-line techniques to train a single hidden layer neural network classifier with logistic hidden and softmax output transfer functions on a multispectral pixel-by-pixel classification problem. In contrast to current practice a multiple class cross-entropy error function has been chosen as the function to be minimized. The non-linear diffierential equations cannot be solved in closed form. To solve for a set of locally minimizing parameters we use the gradient descent technique for parameter updating based upon the backpropagation technique for evaluating the partial derivatives of the error function with respect to the parameter weights. Empirical evidence shows that on-line and epoch-based gradient descent backpropagation fail to converge within 100,000 iterations, due to the fixed step size. Batch gradient descent backpropagation training is superior in terms of learning speed and convergence behaviour. Stochastic epoch-based training tends to be slightly more effective than on-line and batch training in terms of generalization performance, especially when the number of training examples is larger. Moreover, it is less prone to fall into local minima than on-line and batch modes of operation. (authors' abstract)
In this paper a novel modular product unit neural network architecture is presented to model singly constrained spatial interaction flows. The efficacy of the model approach is demonstrated for the origin constrained case of spatial interaction using Austrian interregional telecommunication traffic data. The model requires a global search procedure for parameter estimation, such as the Alopex procedure. A benchmark comparison against the standard origin constrained gravity model and the two-stage neural network approach, suggested by Openshaw (1998), illustrates the superiority of the proposed model in terms of the generalization performance measured by ARV and SRMSE.
