Vol 27, No 1 (2017)

One of the central problems of machine learning is considered—the regression restoration problem. A qualitatively new regression decision tree (RDT) is proposed that is based on the concept of a full decision tree (FDT). Earlier, a similar construction of a decision tree (DT) was successfully tested on the problem of classification by precedents, whose statement is close to the problem considered. The results of testing the model of a full RDT (FRDT) on real data are presented.

In poorly formalized problems of recognition and classification, there are plenty of methods to generate feature descriptions of objects, which are subsequently studied by the methods of computer science. Using the factorization approach (reduction of a recognition/classification problem to a binary form), here we obtain combinatorial criteria for the solvability and regularity properties of the problems and criteria of correctness of algorithms and completeness of algorithmic models (the analysis of which is an important component of the algebraic approach to the synthesis of correct algorithms). The study presents a hierarchy of the criteria and cross-validation methods for assessing their feasibility. Based on the hierarchy of criteria, we formulate a general approach to the analysis of poorly formalized problems.

This article considers the problem of median line contouring, which lies at the heart of marine boundary delimitation methods. The specifics of computational geometry tasks in the geodetic system of coordinates are determined by the fact that there is no analytical description for geodetic lines on a spheroid and, therefore, no direct method of carrying out elementary geometrical operations. The approach suggested in this work is based on a preliminary evaluation of geodetic distances by converting coordinates (φ, λ) of original curves to Cartesian coordinates (X, Y, Z). In addition, we suggest a special data structure arranged as a circular probe. Our algorithm for computing the geodetic coordinates of points of a line equidistant from two given lines ensures high precision and efficiency.

The main problem in the handwritten character recognition systems (HCR) is to describe each character by a set of features that can distinguish it from the other characters. Thus, in this paper, we propose a robust set of features extracted from isolated Amazigh characters based on decomposing the character image into zones and calculate the density and the total length of the histogram projection in each zone. In the experimental evaluation, we test the proposed set of features, to show its performance, with different classification algorithms on a large database of handwritten Amazigh characters. The obtained results give recognition rates that reach 99.03% which we presume good and satisfactory compared to other approaches and show that our proposed set of features is useful to describe the Amazigh characters.

Images captured under non-laboratory conditions potentially suffer from various degradations. Especially noise, blur and scale-variations are often prevalent in real world images and are known to potentially affect the classification process of textured images. We show that these degradations not necessarily strongly affect the discriminative powers of computer based classifiers in a scenario with similar degradations in training and evaluation set. We propose a degradation-adaptive classification approach, which exploits this knowledge by dividing one large data set into several smaller ones, each containing images with some kind of degradation-similarity. In a large experimental study, it can be shown that our method continuously enhances the classification accuracies in case of simulated as well as real world image degradations. Surprisingly, by means of a pre-classification, the framework turns out to be beneficial even in case of idealistic images which are free from strong degradations.

The paper considers the use of a method for the identification and verification of significant patterns to find clear and measurable differences between groups of countries by the nature of the relationship of the dynamics of their macroeconomic indicators. The analysis is conducted using panel data that include annual values of a number of economic indicators in a specified time intervals. An approach based on permutation tests is used to take into account the effect of multiple testing. A technology that combines correlation analysis with the detection of significant patterns has made it possible to reveal statistically significant differences between groups of countries with two types of institutional matrices as identified by sociologists.

Two-dimensional fast Fourier transform (FFT) for image processing and filtering is widely used in modern digital image processing systems. This paper concerns the possibility of using a modification of two-dimensional FFT with an analog of the Cooley–Tukey algorithm, which requires a smaller number of complex addition and multiplication operations than the standard method of calculation by rows and columns.

The standard approach for assessing the response of breast tumors to neoadjuvant chemotherapy is to monitor gross changes in tumor size as measured by physical exam and/or conventional imaging such as mammography. However, the deformable nature of the breast and variation in the imaging procedure, make it difficult to match the shape of breasts between serials of temporal mammograms, particularly when the tumor is shrinking due to the treatment. In this paper we propose a method for assessing residual tumor size following neoadjuvant chemotherapy by analyzing changes between pre-treatment and post-treatment mammograms. Our method consists of three steps: (1) pre-treatment and post-treatment mammograms were first registered in order to circumvent the problem of patient repositioning and breast deformation. (2) Tumors at corresponding locations were segmented using region growing segmentation, (3) and based on changes in tumor sizes the response rate is quantified. The proposed method has been tested on 6 breast cancer patients undergoing neoadjuvant chemotherapy, and the experimental results demonstrate that our approach may improve the ability of mammography to evaluate breast tumor response.

The inverse problem with respect to functions proportional to a voice source and volume velocity of the air flow through the glottis is solved as follows: we compute the inverse Fourier transform of the regularized fraction of short-term speech-signal spectra at intervals with an opened (closed) glottis, minimizing the optimality criterion with respect to the regularization parameter and the glottis opening (closing) time. The optimality criterion for solutions includes the values of the volume velocity and its time derivative at the ends of the interval with an opened glottis and the total value of the negative volume velocity. To obtain an empirical error estimate for the solution, experiments using synthesized signals with various parameters, direct measurements of the glottis, and signals synchronously recorded through pairs of microphones of different types are performed. The most probable determination error for the volume velocity is less than 5% for synthetic sources; if the area of the glottis of the source is measured experimentally, then the said error is about 10%. The discrepancy of solutions for the same signal synchronously recorded through a pair of microphones of different types is less than 10%.