Informatica

In this paper we propose modifications of the well-known algorithm of particle swarm optimization (PSO). These changes affect the mapping of the motion of particles from continuous space to binary space for searching in it, which is widely used to solve the problem of feature selection. The modified binary PSO variations were tested on the dataset SVC2004 dedicated to the problem of user authentication based on dynamic features of a handwritten signature. In the example of k-nearest neighbours (kNN), experiments were carried out to find the optimal subset of features. The search for the subset was considered as a multicriteria optimization problem, taking into account the accuracy of the model and the number of features.

Color quantization is the process of reducing the number of colors in a digital image. The main objective of quantization process is that significant information should be preserved while reducing the color of an image. In other words, quantization process shouldn't cause significant information loss in the image. In this paper, a short review of color quantization is presented and a new color quantization method based on artificial bee colony algorithm (ABC) is proposed. The performance of the proposed method is evaluated by comparing it with the performance of the most widely used quantization methods such as K-means, Fuzzy C Means (FCM), minimum variance and particle swarm optimization (PSO). The obtained results indicate that the proposed method is superior to the others.

A simple Stock Exchange Game Model (SEGM) was introduced in Mockus (2003), to simulate the behavior of several stockholders using fixed buying-selling margins at fixed bank yield.

The presented article is about a research using artificial neural network (ANN) methods for compound (technical and fundamental) analysis and prognosis of Lithuania's National Stock Exchange (LNSE) indices LITIN, LITIN-A and LITIN-VVP. We employed initial pre-processing (analysis for entropy and correlation) for filtering out model input variables (LNSE indices, macroeconomic indicators, Stock Exchange indices of other countries such as the USA – Dow Jones and S&P, EU – Eurex, Russia – RTS). Investigations for the best approximation and forecasting capabilities were performed using different backpropagation ANN learning algorithms, configurations, iteration numbers, data form-factors, etc. A wide spectrum of different results has shown a high sensitivity to ANN parameters. ANN autoregressive, autoregressive causative and causative trend model performances were compared in the approximation and forecasting by a linear discriminant analysis.

The paper presents new method for sequential classification of the time series observations. Methods and algorithms of sequential recognition are obtained on the basis of the recursive equations for sufficient statistics. These recursive equations allow to construct algorithms of current classification of observable sequences in the rate of entering its values into the on-line operation. Classification algorithms are realized in the form of computer programs, including personal computers. They allow to build multi-channel conveyer computational structures for the sequential recognizers of time series observations.