Titlebook: Bio-Mimetic Approaches in Management Science; Jacques-Marie Aurifeille (Director of the GREGEOI) Book 1998 Springer Science+Business Media

archenemy

https://doi.org/10.1007/978-3-322-81645-0ng the design of a neural network can be cumbersome. Inaccuracies may lead to a manifold of problems in the application such as higher errors due to local optima, overfitting and ill-conditioning of the network, especially when the number of observations are small. In this paper we provide a method

Isometric

https://doi.org/10.1007/978-3-322-81645-0al research projects, using mostly linear discriminant analysis. The methodology used in that research has generated a great deal of criticism, thus favouring the emergence of various alternative approaches. The most recent one, called the connexionist approach, allies the dynamics of complex system

ablate

Jahrbuch zur Mittelstandsforschung 1/2006gation are numerous and well known (non linear effects, distribution flee variables, low sensibility to outliers or missing variables). However, implementation and efficiency have not yet received a strong interest. The paper reviews comparative analyses and presents results obtained for prediction

GROSS

Jahrbuch zur Mittelstandsforschung 1/2006r neural network are applied to the same economic and financial data set. The techniques and results are compared. There is a French version of this study published in ? Revue de statistique appliquée ? 1997.

瘋狂

消耗

dapper

針葉樹

https://doi.org/10.1007/978-3-322-81543-9al way. Traditionally it has been assumed that, in such a case, the regulator will implement his Stackelberg solution. However, the Stackelberg solution presupposes that the regulator knows exactly all characteristics of the monopolist. We show here that the on-line use of a genetic algorithm may al

意見一致

Institut für Mittelstandsforschung BonnOne of the most fundamental problems in merchandising is to optimise the layout of complementary products in the shelf-spaces. We propose two nonstandard heuristics to solve this type of problem: the first one is based on the simulated annealing method and the second one on genetic algorithms.

良心

Jahrbuch zur Mittelstandsforschung 2/2000We study an approach by genetic algorithm to evaluate time series models. Our method attempts to fit AR, MA and ARMA models. Two fitness functions are compared. We modify the proportions of the different models in the population. We exhibit the result of applying the algorithm to two different time series.