Search

This paper discusses the factors affecting quality of school educational process. It describes the application of artificial intelligence approaches (neural networks) into the educational process. The analysis of methods for tracking changes in the position of the head while studying at school or at home is presented. The results obtained make it possible to evaluate the psycho‐physiological, psycho‐emotional state of primary school students in the process of interacting with a computer when performing educational tasks. The goal is to create computer tools that monitors changes in head position using images from a laptop or tablet digital camera. This article presents the stages of developing a neural network for assessing head turns when performing a school assignment and the results of applying the program. A new tool is proposed for assessing the state of a student in the learning process, to determine the ability to perceive different types of educational information. [ABSTRACT FROM AUTHOR]

The article presents the discussion on competitiveness being a determining factor for the survival of businesses. Topics include sciences research to combine experimental approaches with theoretical modelling tools from systems theory, physics, engineering, computer science; and situation driven organizations on the search for simulation models.

In this paper the relationship between blood glucose concentration and palm perspiration rate is studied as a non-invasive method. A glucose concentration range from 83 mg/dl to 116.5 mg/dl is examined. An artificial neural network (ANN) trained by the Levenberg–Marquardt algorithm is developed to detect the performance indices based on the one- and two-input variables. A data set for 72 volunteers is used for this study. Data of 36 volunteers are used for training the ANN and data of 36 volunteers were reserved for testing. Results of the study are acceptable with an error of 8.38% for the Elman neural network and 8.77% for the multilayer neural network. Therefore, the palm perspiration rate may be used as a good indicator for detecting glucose concentration in blood. This non-invasive method has advantages such as time saving, cost etc. over other methods and it is painless. The results of clinical experiments, follow-up methods and other applications are presented. [ABSTRACT FROM AUTHOR]

We compare log maximum likelihood gradient ascent, root-mean-square error minimizing gradient descent and genetic-algorithm-based artificial neural network procedures for a binary classification problem. We use simulated data and real-world data sets, and four different performance metrics of correct classification, sensitivity, specificity and reliability for our comparisons. Our experiments indicate that a genetic-algorithm-based artificial neural network that maximizes the total number of correct classifications generally fares well for the binary classification problem. However, if the training data set contains inconsistent decisions or noise then the log maximum likelihood maximizing gradient ascent may be the best classification approach to use. The root-mean-square minimizing gradient descent approach appears to overfit training data and has the lowest reliability among the approaches considered for our research. At the end of the paper, we provide a few guidelines, including computational complexity, for selection of an appropriate technique for a given binary classification problem. [ABSTRACT FROM AUTHOR]

The article presents the discussion on research on artificial intelligence (AI). Topics include made possible by the availability of big data and the rapid growth in information and communications technologies (ICT); and AI and blockchain proving to be quite a powerful combination showing the integration of AI and blockchain affecting many aspects.

Recent researches oriented to photovoltaic (PV) systems feature booming interest in current decade. For efficiency improvement, maximum power point tracking (MPPT) of PV array output power is mandatory. Although classical MPPT techniques offer simplified structure and implementation, their performance is degraded when compared with artificial intelligence-based techniques especially during partial shading and rapidly changing environmental conditions. Artificial neural network (ANN) algorithms feature several capabilities such as: (i) off-line training, (ii) nonlinear mapping, (iii) high-speed response, (iv) robust operation, (v) less computational effort and (vi) compact solution for multiple-variable problems. Hence, ANN algorithms have been widely applied as PV MPPT techniques. Among various available ANN-based PV MPPT techniques, very limited references gather those techniques as a survey. Neither classification nor comparisons between those competitors exist. Moreover, no detailed analysis of the system performance under those techniques has been previously discussed. This study presents a detailed survey for ANN based PV MPPT techniques. The authors propose new categorisation for ANN PV MPPT techniques based on controller structure and input variables. In addition, a detailed comparison between those techniques from several points of view, such as ANN structure, experimental verification and transient/steady-state performance is presented. Recent references are taken into consideration for update purpose. [ABSTRACT FROM AUTHOR]

Artificial neural networks are bio-inspired mathematical models that have been widely used to solve complex problems. The training of a neural network is an important issue to deal with, since traditional gradient-based algorithms become easily trapped in local optimal solutions, therefore increasing the time taken in the experimental step. This problem is greater in recurrent neural networks, where the gradient propagation across the recurrence makes the training difficult for long-term dependences. On the other hand, evolutionary algorithms are search and optimization techniques which have been proved to solve many problems effectively. In the case of recurrent neural networks, the training using evolutionary algorithms has provided promising results. In this work, we propose two hybrid evolutionary algorithms as an alternative to improve the training of dynamic recurrent neural networks. The experimental section makes a comparative study of the algorithms proposed, to train Elman recurrent neural networks in time-series prediction problems. [ABSTRACT FROM AUTHOR]

This study proposes the use of a modified genetic algorithm (MGA), a global search technique, as a training method to improve generalizability and to identify relevant inputs in a neural network (NN) model. Generalizability refers to the NN model's ability to perform well on exemplars (observations) that were not used during training (out-of-sample); improved generalizability enhances NN's acceptability as a valid decision-support tool. The MGA improves generalizability by setting unnecessary weights (or connections) to zero and by eliminating these weights. Because the eliminated weights have no further impact on the training (in-sample or out-of-sample data), the relevant variables can be identified from the model. By eliminating unnecessary weights, the MGA is able to search and find a parsimonious model that generalizes well. Unlike the traditional NN, the MGA identifies the model variables that contribute to an outcome, helping decision makers to rationalize output and accept results with greater confidence. The study uses real-life data to demonstrate the use of MGA. [ABSTRACT FROM AUTHOR]

Neural network techniques are widely used in solving pattern recognition or classification problems. However, when statistical data are used in supervised training of a neural network employing the back-propagation least mean square algorithm, the behavior of the classification boundary during training is often unpredictable. This research suggests the application of monotonicity constraints to the back propagation learning algorithm. When the training sample set is preprocessed by a linear classification function, neural network performance and efficiency can be improved in classification applications where the feature vector is related monotonically to the pattern vector. Since most classification problems in business possess monotonic properties, this technique is useful in those problems where any assumptions about the properties of the data are inappropriate. [ABSTRACT FROM AUTHOR]