Most of the neural networks-based intrusion detection systems (IDS) examine all data features to detect intrusion or misuse patterns. Some of the features may be redundant or contribute little (if anything) to the detection process. That is why the purpose of this study is to identify important KDD features which will be used to train a neural network (NN), in order to best classify and detect attacks. Four NNs were studied: Modular, Recurrent, Principal Component Analysis (PCA), and Time-Lag recurrent (TLR) NNs. We investigated the performance of combining the Fisher's filter used as a feature selection technique, with one of the previously cited NNs. Our simulations show that using Fisher's filter improves largely the performance of the four considered NNs in terms of detection rate, attack classification, and computational time.
In the following paper, the use of fuzzy models in qualitative rating systems is analyzed in detail. The author works in an Austrian finance institution. There are at the moment two rating systems in use. The main purpose of such a rating system is to analyze company ratios to calculate a rating score, which is a measure for the financial situation and rigidity of a company. The first one is a solely hard fact rating system based on the Quicktest by Kralicek. The second one uses self-organizing maps and neural networks to calculate a rating classification and also offers the possibility to dispose personal appraisal in the calculation process.
The following work examines the application spectrum of fuzzy logic and fuzzy models in soft-fact rating systems.
We show that the use of fuzzy models in rating systems enables visualization of additional knowledge and offers the possibility to enhance the influence of a company's soft fact rating to the overall rating.
If the Poisson integral of the unit disc is replaced by its square root, it is known that normalized Poisson integrals of $L^{p}$ and weak $L^{p}$ boundary functions converge along approach regions wider than the ordinary nontangential cones, as proved by Rönning and the author, respectively. In this paper we characterize the approach regions for boundary functions in two general classes of Orlicz spaces. The first of these classes contains spaces $L^{\Phi }$ having the property $L^{\infty }\subset L^{\Phi }\subset L^{p}$, $1\le p<\infty $. The second contains spaces $L^{\Phi }$ that resemble $L^{p}$ spaces.
This article introduces an algorithm for implicit High Dimensional Model Representation (HDMR) of the Bellman equation. This approximation technique reduces memory demands of the algorithm considerably. Moreover, we show that HDMR enables fast approximate minimization which is essential for evaluation of the Bellman function. In each time step, the problem of parametrized HDMR minimization is relaxed into trust region problems, all sharing the same matrix. Finding its eigenvalue decomposition, we effectively achieve estimates of all minima. Their full-domain representation is avoided by HDMR and then the same approach is used recursively in the next time step. An illustrative example of N-armed bandit problem is included. We assume that the newly established connection between approximate HDMR minimization and the trust region problem can be beneficial also to many other applications.
This paper considers a distribution inventory system that consists of a single warehouse and several retailers. Customer demand arrives at the retailers according to a continuous-time renewal process. Material flow between echelons is driven by reorder point/order quantity inventory control policies. Our objective in this setting is to calculate the long-run inventory, backorder and customer service levels. The challenge in this system is to characterize the demand arrival process at the warehouse. We present a Markovian methodology to elucidate and approximate this process. We illustrate the use of this methodology in the distribution inventory system under stochastic transportation times with identical and non-identical retailers.