Data Machine Learning Research Articles

Formation of Bachelor`s Skills in the Field of Machine Learning and Intelligent Data Analysis

The article presents the content of the educational disciplines “Workshop on Python Programming” and “Fundamentals of Artificial Intelligence”, included in the module “Data Analysis. Machine learning. Artificial Intelligence”, implemented in Ural universities. Modern teachers note that when preparing the next generation of teachers, the priority should be their acquisition of experience in using artificial intelligence tools in teaching. The objectives of mastering the module are developing students’ skills in the field of neural network mathematical modeling, developing knowledge about basic data analysis algorithms, development of skills in visualization and interpretation of data to solve applied problems using object-oriented programming technologies and artificial neural network technologies. As part of project activities, students gain experience in data mining by completing research projects (for example, a project on comparative assessment of the effectiveness of multilayer perceptron learning algorithms) and applied projects (for example, on the development of a neural network system for predicting the success of schoolchildren’s research work; on the development of a neural network system for predicting student attendance training sessions). At USPU, software systems have been developed in Python that implement the presented neural network prognostic systems. Educational and methodological materials for the disciplines of the module include video lectures and presentations for them, laboratory work (including those based on GoogleColab notebooks), datasets, test sets, theory questions and practical tasks for exams and tests. The effectiveness of the developed teaching methodology is confirmed by the results of a pedagogical experiment. Keywords: neural network technologies, data mining, predictive systems, machine learning, content of the academic discipline

Unraveling mudstone compaction at microscale: A combined approach of nanoindentation mapping and machine learning data analysis

Compaction processes within mudstone samples across varying depths (723.5–3213.5 m) in the Vienna Basin were investigated, focusing on porosity changes and the resulting micromechanical response in the clay-rich, fine-grained fraction (“clay matrix”). A novel approach combining nanoindentation mapping with machine learning data analysis was developed to efficiently extract representative micromechanical parameters of the clay matrix. Emphasis was put on capturing the properties of the fine-grained, clay-dominated composite rather than individual mineral phases, enabling the analysis of compaction-induced strength changes at microscale. A significant enhancement in the mechanical strength of the clay matrix with increasing burial depth was observed. Reduced elastic modulus (Er) and hardness (H) increased from 6.8 ± 3.4 to 22.6 ± 7.5 GPa and from 0.2 ± 0.2 to 0.9 ± 0.2 GPa, respectively, over the depth interval of 2490 m. Moreover, a strong correlation between depth and porosity and consequently micromechanical properties of the clay matrix exists. This highlights the substantial influence of intra-clay porosity on mechanical properties, which follow a general compaction trend with depth. Broad ion beam-scanning electron microscopy (BIB-SEM) analysis was used for textural investigations at micro-to nanoscale and indicated that the reduction in porosity predominantly resulted from mechanical compaction rather than mineral diagenesis, which would have been noticeable by signs of significant dissolution or cementation. The correlation coefficient matrix between multiscale porosity measurements and nanoindentation affirmed that the porosity loss was closely linked to the enhanced mechanical properties of the fine-grained composite clay matrix, while other compositional variables like total clay mineral content showed weak correlations. Empirical mathematical equations were derived to describe the mechanical properties as generalized functions of depth and porosity. These may be used as geomechanical input for future basin analysis and geoenergy applications (e.g., seal rock studies in a geological storage context). This study introduces a new approach to unravel compaction processes in fine-grained rocks at microscale, emphasizing the interplay between intra-clay porosity and micromechanical changes during proceeding burial.

Optimizing logistics and supply chain management through advanced analytics: Insights from industries

This review paper explores the transformative role of advanced analytics in optimizing logistics and supply chain management, offering insights into industry applications, best practices, and future trends. As global supply chains become increasingly complex, integrating advanced analytics—encompassing data mining, predictive analytics, machine learning, and big data—has emerged as a critical driver of efficiency, cost reduction, and enhanced decision-making. The paper discusses how various industries, including manufacturing, retail, healthcare, and transportation, leverage advanced analytics to address specific challenges and improve overall supply chain performance. Additionally, it highlights the best practices for successful implementation, such as aligning analytics initiatives with business objectives, investing in the right technology infrastructure, and fostering a culture of data-driven decision-making. The paper also addresses the challenges and barriers to implementation, including technological, organizational, and regulatory hurdles. Finally, it examines emerging technologies like AI, IoT, and blockchain, poised to revolutionize supply chain management further. It identifies key opportunities for growth in areas such as sustainability, risk management, and customization. The paper concludes by emphasizing the importance of advanced analytics in shaping the future of logistics and supply chain strategies, offering recommendations for industry stakeholders to harness these technologies effectively.

Health Benefits and Adverse Effects of Kratom: A Social Media Text-Mining Approach

Background: Kratom is a substance that alters one’s mental state and is used for pain relief, mood enhancement, and opioid withdrawal, despite potential health risks. In this study, we aim to analyze the social media discourse about kratom to provide more insights about kratom’s benefits and adverse effects. Also, we aim to demonstrate how algorithmic machine learning approaches, qualitative methods, and data visualization techniques can complement each other to discern diverse reactions to kratom’s effects, thereby complementing traditional quantitative and qualitative methods. Methods: Social media data were analyzed using the latent Dirichlet allocation (LDA) algorithm, PyLDAVis, and t-distributed stochastic neighbor embedding (t-SNE) technique to identify kratom’s benefits and adverse effects. Results: The analysis showed that kratom aids in addiction recovery and managing opiate withdrawal, alleviates anxiety, depression, and chronic pain, enhances mood, energy, and overall mental well-being, and improves quality of life. Conversely, it may induce nausea, upset stomach, and constipation, elevate heart risks, affect respiratory function, and threaten liver health. Additional reported side effects include brain damage, weight loss, seizures, dry mouth, itchiness, and impacts on sexual function. Conclusion: This combined approach underscores its effectiveness in providing a comprehensive understanding of diverse reactions to kratom, complementing traditional research methodologies used to study kratom.

Development of a pig wean-quality score using machine-learning algorithms to characterize and classify groups with high mortality risk under field conditions

Mortality during the post-weaning phase is a critical indicator of swine production system performance, influenced by a complex interaction of multiple factors of the epidemiological triad. This study leveraged retrospective data from 1723 groups of pigs marketed within a US swine production system to develop a Wean-Quality Score (WQS) using machine learning techniques. The study evaluated three machine learning models, Random Forest, Support Vector Machine, and Gradient Boosting Machine, to classify groups having high or low 60-day mortality, where high mortality groups represented 25 % of the groups among the study population with the highest mortality values (n=431; 60-day mortality=9.98 %), and the remaining 75 % of the groups were of low mortality (n=1292; 60-day mortality=2.75 %). The best-performing model, Random Forest (RF), outperformed the other ML models in terms of accuracy (0.90), sensitivity (0.84), and specificity (0.92) metrics, and was then selected for further analysis, which consisted of creating the WQS and ranking the most important factors for classifying groups as high or low mortality. The most important factors ranked through the RF model to classify groups with high mortality were pre-weaning mortality, weaning age, average parity of litters in sow farms, and PRRS status. Additionally, stocking conditions such as stocking density and time to fill the barn were important predictors of high mortality. The WQS was developed and correlated (r = 0.74) with the actual 60-day mortality of the groups, offering a valuable tool for assessing post-weaning survivability in swine production systems before weaning. This study highlights the potential of machine learning and comprehensive data utilization to improve the assessment and management of weaned pig quality in commercial swine production, which producers can utilize to identify and intervene in groups, according to the WQS.

OPTIMIZATION AND FORECASTING OF A SECURITIES PORTFOLIO BASED ON MACHINE LEARNING METHODS

The article is devoted to the demonstration and analysis of the possibilities of connecting the Markowitz model, which is a classical model for forming a portfolio of securities, and forecasting methods based on machine learning data. The relevance of the chosen topic is justified by the fact that the use of machine learning methods makes it possible to simplify the activities of financial sector workers and the work of professional stock market players, as well as reduce the complexity of interpreting mathematical models and methods of their construction for private investors. The purpose of this work is to use machine learning methods to solve the problem of building an optimal securities portfolio and predicting its behavior. Tasks: - build a Markowitz model based on the selected data; - to build an optimal portfolio based on the proposed optimality criterion; - offer a time series forecast; - to assess the adequacy of the constructed model; - combine the Markowitz model, which belongs to the class of widely known classical models, with the construction of a prediction based on machine learning. Methods and models: machine learning methods are used in the work, the Markowitz model is built based on the selected data, then the Markowitz model is combined with the construction of a forecast based on machine learning. Research results: the optimal portfolio for AXP has been built and its forecast has been fulfilled. The formation of the portfolio was based on the Markowitz model. The data obtained were tested for adequacy and showed fairly good results. A portfolio was built, the optimality criterion of which is the ratio of average profitability and volatility of profitability (maximizing average profitability while minimizing its volatility). The Markowitz model, which belongs to the class of widely known classical models, was combined with the construction of a prediction based on machine learning.

Test and Measure for Partial Mean Dependence Based on Machine Learning Methods

It is of importance to investigate the significance of a subset of covariates W for the response Y given covariates Z in regression modeling. To this end, we propose a significance test for the partial mean independence problem based on machine learning methods and data splitting. The test statistic converges to the standard Chi-squared distribution under the null hypothesis while it converges to a normal distribution under the fixed alternative hypothesis. Power enhancement and algorithm stability are also discussed. If the null hypothesis is rejected, we propose a partial Generalized Measure of Correlation (pGMC) to measure the partial mean dependence of Y given W after controlling for the nonlinear effect of Z. We present the appealing theoretical properties of the pGMC and establish the asymptotic normality of its estimator with the optimal root-N convergence rate. Furthermore, the valid confidence interval for the pGMC is also derived. As an important special case when there are no conditional covariates Z, we introduce a new test of overall significance of covariates for the response in a model-free setting. Numerical studies and real data analysis are also conducted to compare with existing approaches and to demonstrate the validity and flexibility of our proposed procedures. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work.

Blood-based multivariate methylation risk score for cognitive impairment and dementia.

The established link between DNA methylation and pathophysiology of dementia, along with its potential role as a molecular mediator of lifestyle and environmental influences, positions blood-derived DNA methylation as a promising tool for early dementia risk detection. In conjunction with an extensive array of machine learning techniques, we employed whole blood genome-wide DNA methylation data as a surrogate for 14 modifiable and non-modifiable factors in the assessment of dementia risk in independent dementia cohorts. We established a multivariate methylation risk score (MMRS) for identifying mild cognitive impairment cross-sectionally, independent of age and sex (P=2.0×10-3). This score significantly predicted the prospective development of cognitive impairments in independent studies of Alzheimer's disease (hazard ratio for Rey's Auditory Verbal Learning Test (RAVLT)-Learning=2.47) and Parkinson's disease (hazard ratio for MCI/dementia=2.59). Our work shows the potential of employing blood-derived DNA methylation data in the assessment of dementia risk. We used whole blood DNA methylation as a surrogate for 14 dementia risk factors. Created a multivariate methylation risk score for predicting cognitive impairment. Emphasized the role of machine learning and omics data in predicting dementia. The score predicts cognitive impairment development at the population level.

AI integration in supply chain and operations management: Enhancing efficiency and resilience

Artificial intelligence (AI) has become a transformative force in supply chain and operations management, offering significant enhancements in efficiency and resilience. This paper examines the integration of AI technologies such as machine learning, predictive analytics, and real-time data processing in demand forecasting, inventory management, logistics, and risk mitigation. By analyzing diverse data sources, AI improves demand forecasting accuracy, reduces inventory costs, optimizes logistics routes, and enhances supply chain visibility. Case studies and data-driven insights demonstrate how AI-driven systems enable companies to adapt to market dynamics, prevent disruptions, and achieve substantial cost savings. The findings suggest that embracing AI is essential for businesses aiming to optimize their supply chain operations and build robust, resilient frameworks capable of withstanding future challenges.

Unlocking the potential: A review of artificial intelligence applications in wind energy

AbstractThis paper presents a comprehensive review of the most recent papers and research trends in the fields of wind energy and artificial intelligence. Our study aims to guide future research by identifying the potential application and research areas of artificial intelligence and machine learning techniques in the wind energy sector and the knowledge gaps in this field. Artificial intelligence techniques offer significant benefits and advantages in many sub‐areas, such as increasing the efficiency of wind energy facilities, estimating energy production, optimizing operation and maintenance, providing security and control, data analysis, and management. Our research focuses on studies indexed in the Web of Science library on wind energy between 2000 and 2023 using sub‐branches of artificial intelligence techniques such as artificial neural networks, other machine learning methods, data mining, fuzzy logic, meta‐heuristics, and statistical methods. In this way, current methods and techniques in the literature are examined to produce more efficient, sustainable, and reliable wind energy, and the findings are discussed for future studies. This comprehensive evaluation is designed to be helpful to academics and specialists interested in acquiring a current and broad perspective on the types of uses of artificial intelligence in wind energy and seeking what research subjects are needed in this field.

Synthetic Generation of Patient Service Utilization Data: A Scalability Study.

To address privacy and ethical issues in using health data for machine learning, we evaluate the scalability of advanced synthetic data generation methods like GANs, VAEs, copulaGAN, and transformer models specifically for patient service utilization data. Our study examines five models on data from a Canadian health authority, focusing on training and generation efficiency, data resemblance, and practical utility. Our findings indicate that statistical models excel in efficiency, while most models produce synthetic data that closely mirrors real data, and is also useful for real-world applications.

Comparative Analysis of Machine Learning Techniques and Data Sources for Dead Tree Detection: What Is the Best Way to Go?

In Central Europe, the extent of bark beetle infestation in spruce stands due to prolonged high temperatures and drought has created large areas of dead trees, which are difficult to monitor by ground surveys. Remote sensing is the only possibility for the assessment of the extent of the dead tree areas. Several options exist for mapping individual dead trees, including different sources and different processing techniques. Satellite images, aerial images, and images from UAVs can be used as sources. Machine and deep learning techniques are included in the processing techniques, although models are often presented without proper realistic validation.This paper compares methods of monitoring dead tree areas using three data sources: multispectral aerial imagery, multispectral PlanetScope satellite imagery, and multispectral Sentinel-2 imagery, as well as two processing methods. The classification methods used are Random Forest (RF) and neural network (NN) in two modalities: pixel- and object-based. In total, 12 combinations are presented. The results were evaluated using two types of reference data: accuracy of model on validation data and accuracy on vector-format semi-automatic classification polygons created by a human evaluator, referred to as real Ground Truth. The aerial imagery was found to have the highest model accuracy, with the CNN model achieving up to 98% with object classification. A higher classification accuracy for satellite imagery was achieved by combining pixel classification and the RF model (87% accuracy for Sentinel-2). For PlanetScope Imagery, the best result was 89%, using a combination of CNN and object-based classifications. A comparison with the Ground Truth showed a decrease in the classification accuracy of the aerial imagery to 89% and the classification accuracy of the satellite imagery to around 70%. In conclusion, aerial imagery is the most effective tool for monitoring bark beetle calamity in terms of precision and accuracy, but satellite imagery has the advantage of fast availability and shorter data processing time, together with larger coverage areas.

Gravity Wave Momentum Fluxes from 1 km Global ECMWF Integrated Forecast System

Progress in understanding the impact of mesoscale variability, including gravity waves (GWs), on atmospheric circulation is often limited by the availability of global fine-resolution observations and simulated data. This study presents momentum fluxes due to atmospheric GWs extracted from four months of an experimental “nature run", integrated at a 1 km resolution (XNR1K) using the Integrated Forecast System (IFS) model. Helmholtz decomposition is used to compute zonal and meridional flux of vertical momentum from ~1.5 petabytes of data; quantities often emulated by climate model parameterization of GWs. The fluxes are validated using ERA5 reanalysis, both during the first week after initialization and over the boreal winter period from November 2018 to February 2019. The agreement between reanalysis and IFS demonstrates its capability to generate reliable flux distributions and capture mesoscale dynamic variability in the atmosphere. The dataset could be valuable in advancing our understanding of GW-planetary wave interactions, GW evolution around atmospheric extremes, and as high-quality training data for machine learning (ML) simulation of GWs.

Advancements and Challenges in Chemoinformatics: From Drug Discovery to Natural Products Research

Chemoinformatics, an interdisciplinary field combining chemistry, computer science, and information technology, is revolutionising scientific research by using computational methods to analyse, interpret, and predict chemical and biological data. This review explores its pivotal role across drug discovery and natural product research, highlighting key methodologies and challenges. It is critical in drug discovery because it expedites the identification and optimization of drug candidates through techniques such as virtual screening, molecular docking, and quantitative structure-activity relationship (QSAR) studies. It also facilitates the integration of computational chemistry with experimental validation, improving predictions iteratively. Chemoinformatics helps with managing databases, virtually screening bioactive compounds, and figuring out structures through molecular dynamics simulations and metabolomics in the field of natural products research. These tools enable researchers to explore the therapeutic potential of indigenous medicinal plants and other natural sources, facilitating the discovery of novel bioactive molecules with potential pharmaceutical applications. However, challenges such as data quality, computational resources, and ethical considerations persist, particularly in regions with limited infrastructure and expertise. The integration of machine learning and big data analytics promises to further enhance predictive modelling and accelerate discoveries in chemoinformatics. Keywords: Chemoinformatics, Drug Discovery, Natural Products, Research

Hydrogen, Oxygen, and Lead Adsorbates on Al13Co4(100): Accurate Potential Energy Surfaces at Low Computational Cost by Machine Learning and DFT-Based Data.

Intermetallic compounds are promising materials in numerous fields, especially those involving surface interactions, such as catalysis. A key factor to investigate their surface properties lies in adsorption energy maps, typically built using first-principles approaches. However, exploring the adsorption energy landscapes of intermetallic compounds can be cumbersome, usually requiring huge computational resources. In this work, we propose an efficient method to predict adsorption energies, based on a Machine Learning (ML) scheme fed by a few Density Functional Theory (DFT) estimates performed on n sites selected through the Farthest Point Sampling (FPS) process. We detail its application on the Al13Co4(100) quasicrystalline approximant surface for several atomic adsorbates (H, O, and Pb). On this specific example, our approach is shown to outperform both simple interpolation strategies and the recent ML force field MACE [arXiv.2206.07697], especially when the number n is small, i.e., below 36 sites. The ground-truth DFT adsorption energies are much more correlated with the predicted FPS-ML estimates (Pearson R-factor of 0.71, 0.73, and 0.90 for H, O and Pb, respectively, when n = 36) than with interpolation-based or MACE-ML ones (Pearson R-factors of 0.43, 0.39, and 0.56 for H, O, and Pb, in the former case and 0.22, 0.35, and 0.63 in the latter case). The unbiased root-mean-square error (ubRMSE) is lower for FPS-ML than for interpolation-based and MACE-ML predictions (0.15, 0.17, and 0.17 eV, respectively, for hydrogen and 0.17, 0.25, and 0.22 eV for lead), except for oxygen (0.55, 0.47, and 0.46 eV) due to large surface relaxations in this case. We believe that these findings and the corresponding methodology can be extended to a wide range of systems, which will motivate the discovery of novel functional materials.

Drone Technology for Precision Agriculture: Advancements and Optimization Strategies

This paper delves into the development and optimization of an intelligent agricultural monitoring system that makes use of drone technology to enhance agricultural productivity and sustainability. With the world's population on the rise and the scarcity of arable land, precision agriculture becomes essential in guaranteeing food security. By harnessing state-of-the-art sensors and imaging technologies, drone technology offers a new and innovative approach to agricultural monitoring. Efficiently collecting vital data on vegetation indices, soil moisture, and crop health. This study seeks to create a state-of-the-art drone-based monitoring system that uses advanced machine learning algorithms and data processing technologies to analyze agricultural data with exceptional precision. It explores the most efficient methods for operating drones, including flight planning and data collection protocols, with the aim of creating a comprehensive agricultural monitoring system. This platform provides the convenience of automating monitoring processes, leading to lower labor costs, improved resource allocation, and a beneficial impact on agricultural modernization. The article delves into the effects of the latest developments in drone technology and the subsequent decrease in costs on the agriculture industry worldwide. It highlights the ways in which these advancements are transforming conventional farming methods.

Mapping the composite nature of clay matrix in mudstones: integrated micromechanics profiling by high-throughput nanoindentation and data analysis

AbstractMudstones and shales serve as natural barrier rocks in various geoenergy applications. Although many studies have investigated their mechanical properties, characterizing these parameters at the microscale remains challenging due to their fine-grained nature and susceptibility to microstructural damage introduced during sample preparation. This study aims to investigate the micromechanical properties of clay matrix composite in mudstones by combining high-speed nanoindentation mapping and machine learning data analysis. The nanoindentation approach effectively captured the heterogeneity in high-resolution mechanical property maps. Utilizing machine learning-based k-means clustering, the mechanical characteristics of matrix clay, brittle minerals, as well as measurements on grain boundaries and structural discontinuities (e.g., cracks) were successfully distinguished. The classification results were validated through correlation with broad ion beam-scanning electron microscopy images. The resulting average reduced elastic modulus (Er) and hardness (H) values for the clay matrix were determined to be 16.2 ± 6.2 and 0.5 ± 0.5 GPa, respectively, showing consistency across different test settings and indenter tips. Furthermore, the sensitivity of indentation measurements to various factors was investigated, revealing limited sensitivity to indentation depth and tip geometry (when comparing Cube corner and Berkovich tip in a small range of indentation depth variations), but decreased stability at lower loading rates. Box counting and bootstrapping methods were applied to assess the representativeness of parameters determined for the clay matrix. A relatively small dataset (indentation number = 60) is needed to achieve representativeness, while the main challenges is to cover a representative mapping area for clay matrix characterization. Overall, this study demonstrates the feasibility of high-speed nanoindentation mapping combined with data analysis for micromechanical characterization of the clay matrix in mudstones, paving the way for efficient analysis of similar fine-grained sedimentary rocks.

On the use of hydrodynamic modelling and random forest classifiers for the prediction of hypoxia in coastal lagoons

Hypoxia is one of the fundamental threats to water quality globally, particularly for partially enclosed basins with limited water renewal, such as coastal lagoons. This work proposes the combined use of a machine learning technique, field observations, and data derived from a hydrodynamic and heat exchange numerical model to predict, and forecast up to 10 days in advance, the occurrence of hypoxia in a eutrophic coastal lagoon. The random forest machine learning algorithm is used, training and validating a set of models to classify dissolved oxygen levels in the lagoon. The Orbetello lagoon, in the central Mediterranean Sea (Italy), has provided a test case for assessing the reliability of the proposed methodology. Results proved that the methodology is effective in providing a reliable short-term evaluation of DO levels, with a high resolution in both time and space throughout an entire lagoon. An overall classification accuracy of up to 91 % was found in the models, with a score for identifying the occurrence of severe hypoxia - i.e. hourly DO levels lower than 2 mg/l - of 86 %. The use of predictors extracted from a numerical hydrodynamic model allows us to overcome the intrinsic limitation of machine learning modelling approaches which rely on input data from relatively few, local field measurements, i.e. the inability to capture the spatial heterogeneity of DO distributions, unless several measuring points are available. The methodological approach is proposed for application to similar eutrophic environments.

Unleashing Sustainable Efficiency: The Integration of Computer Vision into Industry 4.0

The Internet of Things, artificial intelligence, machine learning, and big data are just a few of the cutting-edge technologies that are being integrated into manufacturing processes as part of the “Industry 4.0” revolution. Computer vision is an essential component of Industry 4.0 regarding sustainability, developed as a disruptive technology that extracts and interprets visual information from digital photos or videos using image processing techniques and advanced models. In the context of Industry 4.0, this article offers an overview of computer vision, including its associated prospects, difficulties, and applications. A particular emphasis is placed on sustainability. It explores computer vision applications in robotics and automation, safety and security, process optimization, augmented reality, robotics and inspection, object identification and tracking, predictive maintenance, and quality control and inspection. The study also identifies the critical approaches used to overcome the difficulties in implementing computer vision solutions. Incorporating computer vision into Industry 4.0 holds promise for unleashing unprecedented levels of efficiency, novelty, and competitiveness in the industrial sector. The manufacturing and industrial sectors may use Industry 4.0‘s prospects and adopt sustainable practices by utilizing computer vision and overcoming its inherent limits. This will help to create an eco-conscious and efficient future.

Forecasting bilateral asylum seeker flows with high-dimensional data and machine learning techniques

Abstract We develop monthly asylum seeker flow forecasting models for 157 origin countries to the EU27, using machine learning and high-dimensional data, including digital trace data from Google Trends. Comparing different models and forecasting horizons and validating out-of-sample, we find that an ensemble forecast combining Random Forest and Extreme Gradient Boosting algorithms outperforms the random walk over horizons between 3 and 12 months. For large corridors, this holds in a parsimonious model exclusively based on Google Trends variables, which has the advantage of near real-time availability. We provide practical recommendations how our approach can enable ahead-of-period asylum seeker flow forecasting applications.