Search

The paper introduces an original permutation approach to assessment of the predictive ability of models based on artificial neural networks. Three models based on artificial neural networks (a multilayer perceptron, a network with the radial basis function, and a neural network with the generalized regression) were implemented to illustrate the permutation approach. Data on the spatial distribution of copper in the upper soil layer of Novy Urengoy (Yamalo-Nenets Autonomous Okrug, Russia) were used for modeling. To evaluate the performance of the models, three different methods were used: error indices estimation, a graphical approach (the Taylor diagram), and a randomization estimation of the probability of obtaining the divergence between the observed and predicted series under assumption that both these datasets are taken from the same population. In the permutation approach, two statistics were used: the difference in means and the correlation coefficient. The permutation approach proved to be productive, as it allowed assessing the significance of the divergence between the observed and predicted datasets. [ABSTRACT FROM AUTHOR]

The article presents a study of the current state of the use of digital technologies in the agro-industrial complex. The development of the agro-industrial complex in the conditions of the digital economy involves the use of innovative, digital, cluster, information and computer technologies, as well as new managerial, economic, organizational methods, with fundamentally new properties, which are aimed at increasing the efficiency of production, management and implementation for all areas of the agro-industrial complex. Based on the analysis of foreign experience, the regulatory framework of the Russian Federation, digital standards, types of digital technologies, the paper considers the types of digital technologies in the agro-industrial complex and the possibility of their use for future development. The methodological basis of the study was the works of domestic scientists-economists on the development of innovations and digital technologies in the agro-industrial complex; general scientific research methods (abstract-logical, economic-statistical, etc.) were used, as well as open analytical materials on the development of digital technologies, the use of artificial intelligence and neural networks in the agro-industrial complex, etc. The article reveals that the range of digital technologies currently being created for the agro-industrial complex is diverse and it is aimed at solving a set of problems: growth of output and the efficiency of the agricultural enterprise, introduction of artificial intelligence; creation of insight based on images; determination of the optimal mixture for agronomic products; monitoring the state of crops; methods of automating irrigation and stimulating farmers, etc. As world best practices show, the use of digital technologies in agriculture makes it possible to achieve economic growth in the industry. However, the main problem of introducing digital technologies in agriculture is the high cost of their implementation. [ABSTRACT FROM AUTHOR]

In this paper, several types of artificial neural networks (ANNs) are considered to predict changes in the concentration of one of the main greenhouse gases - carbon dioxide. The study is based on data from the surface CO2 concentration obtained in the course of monitoring the dynamics of changes in the main greenhouse gases on the Arctic island Belyy, Russia. The forecast was obtained using the following types of ANN: Nonlinear Autoregressive Neural Network with an External Input (NARX), Elman Neural Network (ENN), and Multi-Layer Perceptron (MLP). For the study, the time interval of 168 hours was chosen in the summer period (July-August 2015), which is characterized by a pronounced daily cycle of changes in CO2 concentration. The ANN model type NARX most accurately predicted changes in CO2 concentration. [ABSTRACT FROM AUTHOR]