Hasil untuk "Cybernetics"

Hamid Mirzahossein, Soheil Rezashoar

Customer experience is crucial in the airline industry, as understanding passenger satisfaction helps airlines improve service quality. This study evaluates hyperparameter optimization and feature interpretability in machine learning models for predicting airline passenger satisfaction. Support Vector Machine (SVM) and Multilayer Perceptron (MLP) models were tested for binary classification, labeling passengers as ‘Satisfied’ or ‘Neutral or Dissatisfied’ using a Kaggle dataset with ∼104,000 training and ∼26,000 test records. Hyperparameter tuning used grid search with 10-fold cross-validation. For SVM, the optimal setup included the RBF kernel, C = 10, and gamma = ‘auto’, achieving a mean score of 0.9606. For MLP, the best configuration used no regularization, ''he'' initialization, ReLU activation, 30 epochs, batch size of 32, two hidden layers with 32 neurons each, and a learning rate of 0.001, yielding a mean score of 0.9556. Performance metrics included accuracy, precision, recall, and F1-Score, with SVM achieving a test accuracy of 0.96, precision of 0.97, and F1-Score of 0.95, slightly outperforming MLP by <1 %, though MLP was faster at 0.3 s versus SVM’s 18 s. Both models surpassed baseline models and prior studies, benefiting from robust preprocessing and a large dataset. Permutation importance analysis identified Type of Travel, Inflight Wi-Fi Service, Customer Type, and Online Boarding as key predictors, emphasizing passenger needs for digital connectivity and personalized services. These insights guide airlines to prioritize reliable Wi-Fi and efficient online boarding to enhance satisfaction, loyalty, and competitive positioning.

Vitaliy Pavlyshyn, Eduard Manziuk, Oleksander Barmak et al.

Effective and sustainable planning for future smart transportation systems is hindered by outdated traffic management models that fail to capture real-world dynamics, leading to congestion and significant environmental impact. To address this, advanced machine learning models are required to provide high-fidelity insights into urban mobility. In this work, we propose an adaptive machine learning approach to traffic pattern recognition that synergizes the HDBSCAN and k-means clustering algorithms. By employing a data-driven weighted voting mechanism, our solution provides a robust analytical foundation for sustainable planning, integrating structural analysis with precise cluster refinement. The crafted model was validated using a high-fidelity simulation of the Khmelnytskyi, Ukraine, transport network, where it demonstrated a superior ability to identify distinct traffic modes, achieving a V-measure of 0.79–0.82 and improving cluster compactness by 10–14% over standalone algorithms. It also attained a scenario identification accuracy of 92.8–95.0% with a temporal coherence of 0.94. These findings confirm that our adaptive approach is a foundational technology for intelligent transport systems, enabling the planning and deployment of more responsive, efficient, and sustainable urban mobility solutions.

Simona-Vasilica Oprea, Adela Bara

This study harnesses data from a questionnaire carried out in May 2022 from the Pew Research Center to explore energy sources, climate change issues and factors influencing the adoption of Electric Vehicles (EVs). It was administered to a sample of 10,282 U.S. adults from a total of 11,674, achieving an 88% response rate. The survey was conducted to gauge public opinion on topics related to climate change. Cross-tabulation between the six questions related to EVs and four questions on demographics is performed. The interpretation of the results is essential to support EV adoption. Equally important is the capacity to predict the likelihood of EV adoption. A classification model is proposed, embedding several cutting-edge classifiers. As the neutral segment of population is significant (over 22%), the 1st mapping included this segment associated to the &#x201C;Likely&#x201D; (to adopt) class, whereas the 2nd mapping included this segment associated to the &#x201C;Unlikely&#x201D; class. Several classifiers are tested as baseline (Logistic Regression) or as cutting-edge algorithms (Random Forest-RF, eXtreme Gradient Boost-XGB, Light Gradient Boosting Machine-LGBM, Support Vector Classifier). For the 1st mapping, the RF model shows the best performance AUC=0.83. For the 2nd mapping, RF again performs well with the highest accuracy and AUC (0.86). XGB and LGBM have higher precision, but significantly lower recall compared to RF, which reduces their F1 scores. The most influential features for EV adoption are identified with feature importance: 1) Favor or oppose phasing out new gasoline cars and trucks by 2035; 2) Favor or oppose providing incentives to increase use of hybrid and EV; 3) More important priority for addressing America&#x2019;s energy supply; (4) Family income.

Onur Karadeli, Kıymet Kaya, Şule Gündüz Öğüdücü

The landscape of machine learning constantly demands innovative approaches to enhance algorithms’ performance across diverse tasks. Meta-learning, known as “learning to learn” is a promising way to overcome these diversity challenges by blending multiple algorithms. This study introduces the IHML: Incremental Heuristic Meta-Learner, a novel meta-learning algorithm for classification tasks. By leveraging a variety of base-learners with distinct learning dynamics, such as Gaussian, tree, and instance, IHML offers a comprehensive solution adaptable to different data characteristics. Moreover, the core contributions of IHML lie in its ability to tackle the optimal base-learner and feature sets determination mechanism with the help of Explainable Artificial Intelligence (XAI) and heuristic elbow methods. Existing work in this context utilizes XAI mostly in pre-processing the data or post-analysis of the results, however, IHML incorporates XAI into the learning process in an iterative manner and improves the prediction performance of the meta-learner. To observe the performance of the proposed IHML, we used five different datasets from astrophysics, physics, biology, e-commerce, and economics. The results show that the proposed model achieves more accuracy (in average % 10 and at most % 71 improvements) compared to the baseline machine learning models in the literature.

Kwang-Rim Baek, Sudha Rani Ramakrishnan, Soo-Jung Kim et al.

Mannan and outer structural yeast cell wall polysaccharides have recently garnered attention for their health defense and cosmetic applications. In addition, many studies have confirmed that yeast cell wall mannans exhibit various biological activities, such as antioxidant, immune regulation, reducing hyperlipidemia, and gut health promotion. This paper elucidates yeast cell wall mannan structural features, biological activities, underlying molecular mechanisms, and biosynthesis. Moreover, mannan-overproducing strategies through yeast strain engineering are emphasized and discussed. This review will provide a scientific basis for yeast cell wall mannan research and industrial applications.

I. V. Gladyshev, A. N. Yurasov, M. M. Yashin

Objectives. When measuring the transverse Kerr effect on thin-film structures, interference effects have a great influence on the result obtained. In conference presentations, some researchers have reported on the use of white light in experiments. In their opinion, despite the thickness of the studied layers being much less than the wavelength of light, white light can help avoid interference effects and/or resonant excitation of plasmon waves. The aim of the present work is to verify the validity of such statements using simulation.Methods. In order to solve this problem, the method of computer simulation was used. A numerical solution of equations was compiled for a model structure for various thicknesses and materials of layers.Results. The simulation results show that interference effects in different parts of the spectrum when using white light sources do not neutralize each other. The magnitude of the effect is affected not only by the thickness of the structure layers, but also by the shape of the source emission spectrum, as well as the sensitivity curve of the photodetector. In this case, the output of the measured value of the effect to a plateau at relatively large thicknesses of the magnetooptical film is due to the light being absorbed in the thickness of the magneto-optical film and is negligibility of the back reflection of light from the substrate.Conclusions. The presented technique takes into account the influence of interference effects when measuring the equatorial Kerr effect in white light or using other sources having a wide spectral range, thus improving the interpretation of experimental results. The results are relevant to the development and research of the physical foundations for creating new and improving existing devices in micro-, nano-, and solid-state electronics, as well as quantum devices, including optoelectronic devices and converters of physical quantities.

Leonid N. Kessarinsky

D.A. Savelyev

In this work, the propagation of optical vortices with circular, radial, and azimuthal polarization through subwavelength ring gratings with standard and GRIN substrates using a finite difference time domain method is numerically simulated. It is shown that it is possible to select the polarization of laser radiation and parameters of the element in such a way that a long optical needle (up to 8.04λ, radial polarization), a tight focal spot (up to 0.4λ in diameter, circular polarization), single optical traps, and combinations thereof are generated on the optical axis.

Jano Markhulia, Shalva Kekutia, Vladimer Mikelashvili et al.

In this study, we investigated the cytotoxicity of doxorubicin (DOX)-loaded magnetic nanofluids on 4T1 mouse tumor epithelial cells and MDA-MB-468 human triple-negative breast cancer (TNBC) cells. Superparamagnetic iron oxide nanoparticles were synthesized using sonochemical coprecipitation by applying electrohydraulic discharge treatment (EHD) in an automated chemical reactor, modified with citric acid and loaded with DOX. The resulting magnetic nanofluids exhibited strong magnetic properties and maintained sedimentation stability in physiological pH conditions. The obtained samples were characterized using X-ray diffraction (XRD), transmission electron microscopy (TEM), Fourier-transform infrared spectroscopy, UV-spectrophotometry, dynamic light scattering (DLS), electrophoretic light scattering (ELS), vibrating sample magnetometry (VSM), and transmission electron microscopy (TEM). In vitro studies using the MTT method revealed a synergistic effect of the DOX-loaded citric-acid-modified magnetic nanoparticles on the inhibition of cancer cell growth and proliferation compared to treatment with pure DOX. The combination of the drug and magnetic nanosystem showed promising potential for targeted drug delivery, with the possibility of optimizing the dosage to reduce side-effects and enhance the cytotoxic effect on cancer cells. The nanoparticles’ cytotoxic effects were attributed to the generation of reactive oxygen species and the enhancement of DOX-induced apoptosis. The findings suggest a novel approach for enhancing the therapeutic efficacy of anticancer drugs and reducing their associated side-effects. Overall, the results demonstrate the potential of DOX-loaded citric-acid-modified magnetic nanoparticles as a promising strategy in tumor therapy, and provide insights into their synergistic effects.

V. V. Avagyan

The article identifies the main Russian, European and American developers of digital marketing channels and tools, analyzes the existing digital marketing tools in Russia, highlights the restrictions that apply to Russian businesses and consumers. Special attention was paid to statistical data on the use of digital marketing tools, software, mobile applications, contextual, banner, and targeted advertising on the Internet. The need for the transition of Russian business to domestic digital marketing tools is noted, while the need to improve existing tools is emphasized. In conclusion, recommendations are given, ways of development are highlighted for the further transformation of digital marketing tools in Russia.