In the article, analysis of the work conditions and the metrological characteristics of the air gauges dedicated for the roundness assessment is described. To reach the required accuracy of the acquired data, the phenomena of gas dynamics had to be analyzed in the whole flow through elements of the air gauge. A model based on the second critical parameters was used, because it reflects the true processes of the air flow. As a result, fast and accurate simulations provided series of the characteristics to be considered. Nevertheless, the chosen air gauge configuration underwent the experimental verification of its metrological characteristics. Finally, the entire measurement system Geoform with the gauge head based on the chosen air gauges underwent the accuracy test in order to make sure of its overall measurement quality.

The concept of `diversity' has been one of the main open issues in the field of multiple classifier systems. In this paper we address a facet of diversity related to its effectiveness for ensemble construction, namely, explicitly using diversity measures for ensemble construction techniques based on the kind of overproduce and choose strategy known as ensemble pruning. Such a strategy consists of selecting the (hopefully) more accurate subset of classifiers out of an original, larger ensemble. Whereas several existing pruning methods use some combination of individual classifiers' accuracy and diversity, it is still unclear whether such an evaluation function is better than the bare estimate of ensemble accuracy. We empirically investigate this issue by comparing two evaluation functions in the context of ensemble pruning: the estimate of ensemble accuracy, and its linear combination with several well-known diversity measures. This can also be viewed as using diversity as a regularizer, as suggested by some authors. To this aim we use a pruning method based on forward selection, since it allows a direct comparison between different evaluation functions. Experiments on thirty-seven benchmark data sets, four diversity measures and three base classifiers provide evidence that using diversity measures for ensemble pruning can be advantageous over using only ensemble accuracy, and that diversity measures can act as regularizers in this context.

We all face a wide array of different choices every day of our lives. Asst. Prof. Miłosz Kadziński explains how artificial intelligence could be used to help us make decisions.

In 2015 an important part of the official evaluation of Polish scientific journals was left to experts’ judgement. In this paper we try to establish which observable factors (with available data) are closely related to the outcome of experts’ evaluation of Polish journals in economic sciences. Using the multiple regression statistical model we show that only 5 variables (out of 17) significantly explain almost 50% of the empirical variance of the experts’ evaluation. The determinants of particular interest, not entering the formal criteria and not related to the impact on global science, are: the number of citations mainly in Polish journals and the affiliation with the Polish Academy of Sciences.

The correlation-regression method, as one of the indirect sampling methods, is only sporadically used in geological and mining activities. Theoretically, it should be particularly useful for predicting the content of some chemical components in limestone and marl deposits due to the correlation between them. The results of simple and multiple correlation and regression analysis for 5 selected components (CaO, SiO2, Al2O3, MgO, and SO3), determined in samples from exploratory boreholes and blast holes carried out in the Barcin-Piechcin-Pakość deposit, are presented in the article. The determination coefficients were used as a measure of the correlation power and the quality of the regression models. A very strong linear correlation between CaO and SiO2 content and strong linear correlations between CaO and Al2O3 and SiO2 with Al2O3 have been found. The correlation relationships of the remaining pairs of oxides are weak or very weak and do not provide a basis for prediction of their content based on regression models binding them with the content of other components. The use of nonlinear models for these pairs of oxides results in only a slight improvement in the quality of regression, insignificant from a practical point of view. The application of multiple regression models, linking the content of the mentioned components (with the exception of CaO), leads to similar conclusions. Compared to the determination coefficients of a simple linear correlation, a strong increase in determination coefficients obtained in two cases was found to be artificial and caused by a correlation between the content of the selected components acting as independent variables. From the geological and mining point of view, the results of the analysis indicate the possibility of a fully reliable prediction of SiO2 content and the limited reliability of the Al2O3 content prediction when the CaO content is determined using simple linear regression models.

The structural system of a multiple strip-shaped pillar-roof is common in underground mine exploitation, and research on its mechanics and micro/macroeconomics is meaningful for utilizing strip-shaped pillar resources. A general model of the structural system of a multiple strip-shaped pillar-roof was established, the deformation mechanism of the model was analysed by material mechanics, and the deflection curve equations of the model were obtained. Based on the stress strain constitutive relation of the strip pillar and cusp catastrophe theory, the nonlinear dynamic instability mechanism of the structural system of a multiple strip-shaped pillar-roof was analysed, and the expressions of the pillar width for maintaining the stability of different types of structural systems were derived. The benefits of different structural systems were calculated using micro/macroeconomic theory, the type of the structural system was determined, and different recovery schemes were obtained. Theoretical application research was applied to a large manganese mine, and the results demonstrate that no pillar recovery was needed in 2016, a 9-m wide artificial pillar could be built to replace a pillar in 2017, and the construction of 14-m wide artificial pillars can be conducted in 2018.

Infiltration process plays important role in water balance concept particularly in runoff analysis, groundwater re-charged, and water conservation. Hence, increasing knowledge concerning infiltration process becomes essential for water manager to gain an effective solution to water resources problems. This study employed multiple linear regression for esti-mating infiltration rate where the soil properties used as the predictor variable and measured infiltration rate as the response variable. Field measurement was conducted at sixteen points to obtain infiltration rate using double ring infiltrometer and soil properties namely soil porosity, silt, clay, sand content, degree of saturation, and water content. The result showed that measured infiltration rate had an average initial infiltration rate (f0) of 6.92 mm∙min–1 and final infiltration rate (fc) of 1.49 mm∙min–1. Soil porosity and sand content showed a positive correlation with infiltration rate by 0.842, 0.639, respectively, while silt, clay, water content, and degree of saturation exhibited a negative correlation by –0.631, –0.743, –0.66 and –0.49, respectively. Three types of regression equations were established based on type of soil properties used as predictor varia-bles. The model performance analysis was conducted for each equation and the result shows that the equation with five predictor variables fMLR_3 = – 62.014 + 1.142 soil porosity – 0.205 clay, – 0.063 sand – 0.301, silt + 0.07 soil water content with R2 (0.87) and Nash–Sutcliffe (0.998) gave the best result for estimating infiltration rate. The study found that soil po-rosity contributes mostly to the regression equation that indicates great influence in controlling soil infiltration behavior.

The method described in this work allows to determine the optimal distribution of pulses of digital signal as well as the non-linear mathematical model based on a multiple regression statistical analysis, which are specialized to an effective and low-cost testing of functional parameters in analog electronic circuits. The aim of this concept is to simplify the process of analog circuit specification validation and minimize hardware implementation, time and memory requirements during the testing stage. This strategy requires simulations of the analyzed analog electronic circuit; however, this effort is done only once – before the testing stage. Then, validation of circuit specification can be obtained after a quick, very low-cost procedure without time consuming computations and without expensive external measuring equipment usage. The analyzed test signature is a time response of the analog circuit to the stream of digital pulses for which distributions were determined during evolutionary optimization cycles. Besides, evolutionary computations assure determination of the optimal form and size of the non-linear mathematical formula used to estimate specific functional parameters. Generally, the obtained mathematical model has a structure similar to the polynomial one with terms calculated by means of multiple regression procedure. However, a higher ordered polynomial usage makes it possible to reach non-linear estimation model that improves accuracy of circuit parametric identification. It should be noted that all the evolutionary calculations are made only at the before test stage and the main computational effort, for the analog circuit specification test design, is necessary only once. Such diagnosing system is fully synchronized by a global digital signal clock that precisely determines time points of the slopes of input excitation pulses as well as acquired output signature samples. Efficiency of the proposed technique is confirmed by results obtained for examples based on analog circuits used in previous (and other) publications as test benchmarks.

Cost estimation in the pre-design phase both for the contractor as well as the investor is an important aspect from the point of view of budget planning for a construction project. Constantly growing commercial market, especially the one of public utility constructions, makes the contractor, at the stage of development the design concept, initially estimate the cost of the facade, e.g. office buildings, commercial buildings, etc., which are most often implemented in the form of aluminum-glass facades or ventilated elevations. The valuation of facade systems is of an individual calculation nature, which makes the process complicated, time-consuming, and requiring a high cost estimation work. The authors suggest using a model for estimating the cost of facade systems with the use of statistical methods based on multiple and stepwise regression. The data base used to form statistical models is the result of quantitative-qualitative research of the design and cost documentation of completed public facilities. Basing on the obtained information, the factors that shape the costs of construction façade systems were identified; which then constitute the input variables to the suggested cost estimation models.