Profitability of Turkish banking sector gained importance after national and international financial crisis happened in the last decade, which revealed the need to make a research on profitability and the factors determining profitability. In recent years, new techniques of soft computing (SC) like genetic algorithms (GAs), fuzzy logic (FL) and especially artificial neural networks (ANNs) have been applied into the financial domain to solve the domain issues because of their successful applications in nonlinear multivariate situations. An adaptive system was needed due to the fact that insufficient use of application software programs for SC and the fact that single software is only applicable for specific model. Furthermore, even though ANNs have been applied to many areas; little attention has been paid to estimation of bank profitability with ANNs. This article is intended to analyze and estimate the profitability of deposit banks in Turkey with an adaptive software model of ANNs which have not been previously applied for this context, comprehensively. The results from the software model, which processes the factors affecting profitability, indicate that all of the variables used have significant impacts in varying proportions on profitability and that obtained estimations achieved the targeted and acceptable performance of success. This software model is expected to provide easiness on estimating bank profitability, since giving successful estimations and not being affected by user differences. Additionally, it is aimed to construct a software model for being used in different fields of study and financial domain.
When dealing with the curse of dimensionality (small sample size with many dimensions), feature selection is an important preprocessing strategy for the analysis of biomedical data. This issue is particularly germane to the classification of high-dimensional class-labeled biomedical spectra as is often acquired from magnetic resonance and infrared spectrometers. A technique is presented that stochastically selects feature subsets with varying cardinality for automated discrimination using two types of neural network classifiers. The results are benchmarked against classifiers using the entire feature set with and without averaging. Stochastic feature subset selection had significantly fewer misclassifications than either of the benchmarks.
The paper deals with a predictive vector quantization of an image
based on a neural network architectures, wliere a vector predictor is iniplernented by three-layer neural network with various hidden nodes and bias units, sigrnoid function as nonlinearity and where vector quantizer is inipleinented by Kohonen self-organizing feature maps, it means the codebook is obtained by neural network clustering algorithm. We have tested aíi influence of a nuinber of hidden nodes, various convergention rates of a learning algorithm and a presence of the sigrnoid function to a rnean square prediction error. Next we háve studied an influence of codebook size to a rnean sciuare quantization error, that means a performance of predictive vector quantization system for various bit rates. The image of Lena of size 512 X 512 pels was coded for various bit rates, where we háve ušed onedirnensional and two-dimensional vector prediction of the blocks of pels.
Aiming to hide the real money gains and to avoid taxes, fictive prices are sometimes recorded in the real estate transactions. This paper is concerned with artificial neural networks based screening of real estate transactions aiming to categorize them into "clear" and "fictitious" classes. The problem is treated as an outlier detection task. Both unsupervised and supervised approaches to outlier detection are studied here. The soft minimal hyper-sphere support vector machine (SVM) based novelty detector is employed to solve the task without the supervision. In the supervised case, the effectiveness of SVM, multilayer perceptron (MLP), and a committee based classification of the real estate transactions are studied. To give the user a deeper insight into the decisions provided by the models, the real estate transactions are not only categorized into "clear" and "fictitious" classes, but also mapped onto the self organizing map (SOM), where the regions of "clear", "doubtful" and "fictitious" transactions are identified. We demonstrate that the stability of the regions evolved in the SOM during training is rather high. The experimental investigations performed on two real data sets have shown that the categorization accuracy obtained from the supervised approaches is considerably higher than that obtained from the unsupervised one. The obtained accuracy is high enough for the technique to be used in practice.
The language EpsiloNN allows a high-level specification of arbitrary
neural network structures. It is especially designed for the automatic generation of simulation code, which can run efficiently on different parallel Computer architectures. In this páper soine applications of EpsiloNN are presented. First, the basic syntactical and semantical aspects of the language are described briefly. Then the EpsiloNN specifications of a popular multilayer perceptron (MLP) and of a more complex hybrid LVQ/FfBF neural network architecture are presented. Further features of the language are explained by example.