Deploying machine learning for long-term road pavement moisture prediction: A case study from Queensland, Australia

A novel interval decomposition correlation particle swarm optimization-extreme learning machine model for short-term and long-term water quality prediction

This study aims to improve the understanding of sloshing phenomena in partially filled movable storage systems, particularly in rectangular tanks where experimental data remain limited. A multi-approach methodology was employed by combining analytical and numerical solutions validated with experimental work and supported with various machine learning (ML) prediction approaches. The results were validated by conducting experimental investigations at three different liquid heights (150, 100, and 50) mm. The accuracy of the ML model in predicting the fundamental frequency was compared with three different error metrics and cross-validation, where the extreme gradient boost revealed more robust performance compared with support vector regression and artificial neural networks. The parametric investigations are presented, showing the impact of slope on the first fundamental frequency values for various nonstandard shapes of the liquid fillings, sloped from one side, from both sides, symmetrically, and asymmetrically beveled. The outputs revealed that such a combination of physics-based models with ML prediction provides further insight into sloshing phenomena and frequency responses at various circumstances.

There is growing interest in canine behavioral research specifically for working dogs. Here we take advantage of a dataset of a Transportation Safety Administration olfactory detection cohort of 628 Labrador Retrievers to perform Machine Learning (ML) prediction and classification studies of behavioral traits and environmental effects. Data were available for four time points over a 12 month foster period after which dogs were accepted into a training program or eliminated. Three supervised ML algorithms had robust performance in correctly predicting which dogs would be accepted into the training program, but poor performance in distinguishing those that were eliminated (~ 25% of the cohort). The 12 month testing time point yielded the best ability to distinguish accepted and eliminated dogs (AUC = 0.68). Classification studies using Principal Components Analysis and Recursive Feature Elimination using Cross-Validation revealed the importance of olfaction and possession-related traits for an airport terminal search and retrieve test, and possession, confidence, and initiative traits for an environmental test. Our findings suggest which tests, environments, behavioral traits, and time course are most important for olfactory detection dog selection. We discuss how this approach can guide further research that encompasses cognitive and emotional, and social and environmental effects.

The applications of nanomaterials in various fields have led to an urgent need for rapid assessment of the biosafety of nanomaterials. The time-consuming, laborious, and expensive biological experiments are difficult to meet the needs of rapid and comprehensive evaluation of nanotoxicity. Based on a large number of existing toxicity data of nanomaterials, machine learning has played an import role in predicting nanotoxicity. This chapter briefly describes the nanotoxicology and the application of machine learning in predicting the toxicity of nanomaterials such as carbon nanomaterials, metal and metal oxide nanomaterials, and some emerging nanomaterials. We also propose prospects for future development of machine learning in prediction of nanotoxicology.

Petrophysical parameters such as porosity and water saturation are vital in the petroleum industry for reservoir characterization. These aspects are typically assessed through laboratorium measurements of core samples or intricate petrophysical calculations. Machine Learning (ML) offers a cost-effective and efficient approach as an alternative to conventional methods of predicting those parameters. However, developing ML models can be prone to the invisible traps such as overfitting, underfitting, feature selection, and feature importance. This study is intended to share how to identify the traps and its mitigation by establishing a synergistic workflow between ML and petrophysical theory. A model was developed based on data from several wells in X field, where they are randomized and split into test and train data. Well-log normalization preceded data splitting, and input features were normalized with outlier removal. A feature selection function was then employed to choose a specific amount of log data. Finally, the model selection function identified the highest-scoring model. Without a proper workflow, overfitting, irrelevant feature selection, and imprecise ranking issues emerged. However, with the proper workflow, these invisible traps were mitigated, even with a relatively small dataset. The final model could accurately predict porosity and water saturation

With the availability of data and computational technologies in the modern world, machine learning (ML) has emerged as a preferred methodology for data analysis and prediction. While ML holds great promise, the results from such models are not fully unreliable due to the challenges introduced by uncertainty. An ML model generates an optimal solution based on its training data. However, if the uncertainty in the data and the model parameters are not considered, such optimal solutions have a high risk of failure in actual world deployment. This paper surveys the different approaches used in ML to quantify uncertainty. The paper also exhibits the implications of quantifying uncertainty when using ML by performing two case studies with space physics in focus. The first case study consists of the classification of auroral images in predefined labels. In the second case study, the horizontal component of the perturbed magnetic field measured at the Earth’s surface was predicted for the study of Geomagnetically Induced Currents (GICs) by training the model using time series data. In both cases, a Bayesian Neural Network (BNN) was trained to generate predictions, along with epistemic and aleatoric uncertainties. Finally, the pros and cons of both Gaussian Process Regression (GPR) models and Bayesian Deep Learning (DL) are weighed. The paper also provides recommendations for the models that need exploration, focusing on space weather prediction.

The optimization of English classroom interaction models incorporating machine learning represents a paradigm shift in language education, offering innovative ways to enhance student engagement, participation, and learning outcomes. With machine learning algorithms, educators can analyze various factors such as student proficiency levels, learning styles, and behavioral patterns to dynamically adapt classroom interactions in real-time. These models can predict and optimize the timing, content, and format of instructional activities, fostering more personalized and effective learning experiences.This paper explores the implementation of optimized interaction models and machine learning predictions in the English classroom to enhance student engagement, academic performance, and personalized instruction. Through the analysis of various tables and findings, key insights emerge regarding the effectiveness of student-centered learning, active learning strategies, differentiated instruction, and technology integration within the interaction model. Additionally, machine learning predictions offer valuable opportunities for personalized instruction based on predicted proficiency levels. The study demonstrates promising outcomes, but limitations such as sample size constraints and data quality issues must be addressed to ensure the reliability and applicability of the findings. Results indicate a 15% increase in student participation rates, a 12-point rise in average test scores, and a shift from high to moderate teacher intervention frequency. Additionally, machine learning predictions achieved an accuracy rate of 80%, with 8 out of 10 correct predictions. Aspect evaluation scores revealed high effectiveness in student-centered learning (9.2), formative assessment (9.3), and teacher facilitation (9.0). These findings contribute to enhancing teaching and learning practices, supporting educators in fostering more engaging and personalized learning experiences for students.

There has been very limited research conducted to predict rental prices in the German real estate market using an AI-based approach. From a general perspective, conventional approaches struggle to handle large amounts of data and fail to consider the numerous elements that affect rental prices. The absence of sophisticated, data-driven analytical tools further complicates this situation, impeding stakeholders, such as tenants, landlords, real estate agents, and the government, from obtaining the accurate insights necessary for making well-informed decisions in this area. This paper applies novel machine learning (ML) approaches, including ensemble techniques, neural networks, linear regression (LR), and tree-based algorithms, specifically designed for forecasting rental prices in Munich. To ensure accuracy and reliability, the performance of these models is evaluated using the R2 score and root mean squared error (RMSE). The study provides two feature sets for model comparison, selected by particle swarm optimisation (PSO) and CatBoost. These two feature selection methods identify significant variables based on different mechanisms, such as seeking the optimal solution with an objective function and converting categorical features into target statistics (TSs) to address high-dimensional issues. These methods are ideal for this German dataset, which contains 49 features. Testing the performance of 10 ML algorithms on two sets helps validate the robustness and efficacy of the AI-based approach utilising the PyTorch framework. The findings illustrate that ML models combined with PyTorch-based neural networks (PNNs) demonstrate high accuracy compared to standalone ML models, regardless of feature changes. The improved performance indicates that utilising the PyTorch framework for predictive tasks is advantageous, as evidenced by a statistical significance test in terms of both R2 and RMSE (p-values < 0.001). The integration results display outstanding accuracy, averaging 90% across both feature sets. Particularly, the XGB model, which exhibited the lowest performance among all models in both sets, significantly improved from 0.8903 to 0.9097 in set 1 and from 0.8717 to 0.9022 in set 2 after being combined with the PNN. These results showcase the efficacy of using the PyTorch framework, enhancing the precision and reliability of the ML models in predicting the dynamic real estate market. Given that this study applies two feature sets and demonstrates consistent performance across sets with varying characteristics, the methodology may be applied to other locations. By offering accurate projections, it aids investors, renters, property managers, and regulators in facilitating better decision-making in the real estate sector.

The rapid development of urban industrialization has had many negative effects on the quality of water sources around cities. Long-term prediction of water quality can be of great help to the conservation of water environment. This case tries to use several popular deep learning models, such as RNN, LSTM, MLP, and Transformer-based models to predict the long-term integrated water quality index in the Chaohu Lake area. The dataset is derived from daily monitoring data from four monitoring sites within Chaohu Lake from 2019 to 2022, and the long-term prediction performance of the model is evaluated using MAE and MSE as evaluation metrics. The experimental results showed that all models selected in this case achieved good results within the study area, but Informer performed more prominently (MSE = 0.2455, MAE = 0.2449) as the length of the prediction series increased. Our results demonstrate the effectiveness of popular deep learning models in the field of WQI prediction, especially the significant advantage of transformer-based models represented by Informer in long-term water quality prediction, which will further provide an effective modern tool for water quality monitoring and management.

This study explores the use of predictive machine learning techniques to enhance energy management in sustainable transportation systems, with a specific emphasis on electric vehicles (EVs). The analysis of EV specifications has shown a wide variety of battery capacities, ranging from 55 kWh to 75 kWh. These capacities have a direct impact on the energy storage capacity and the possible driving range of the vehicles. The range of vehicle weights, ranging from 1400 kg to 1700 kg, emphasized the possible effects on energy consumption rates and overall efficiency. The performance capabilities were shown with maximum speeds ranging from 160 km/h to 200 km/h. The energy consumption rates ranged from 0.18 kWh/km to 0.25 kWh/km, suggesting different levels of efficiency. An analysis of energy management data revealed that the lengths traveled varied from 180 km to 220 km, while the average speeds ranged from 50 km/h to 60 km/h. These variations directly affected the rates at which energy was used. The vehicles exhibited higher efficiency metrics by attaining energy consumption rates of 4.0 km/kWh to 5.6 km/kWh. The analysis of ambient variables indicated temperature fluctuations ranging from 20°C to 30°C, as well as a variety of terrain types that impact driving conditions and energy requirements. Predictive machine learning models demonstrated high accuracies, with Mean Absolute Error (MAE) values ranging from 0.13 to 0.18 kWh/km, Root Mean Squared Error (RMSE) values ranging from 0.18 to 0.22 kWh/km, and R-squared (R^2) scores ranging from 0.80 to 0.88. These results emphasize the need of using predictive machine learning to estimate energy consumption, optimize energy management systems, and address sustainable transportation concerns in order to improve the efficiency and sustainability of electric vehicles.

The review paper discusses the development and assessment of hybrid machine learning frameworks for early disease diagnosis using multi-source clinical data. It highlights the importance of early disease forecasting in healthcare, as it allows for more efficient illness management, mitigating symptom severity, and decelerating disease development. Traditional disease prediction methods often face challenges, such as reliance on isolated data sources and the existence of imbalanced datasets. Hybrid machine learning models offer a robust approach to address these shortcomings by integrating the advantages of multiple machine learning methods to enhance predictive accuracy and resilience. Case studies have demonstrated the effectiveness of hybrid machine learning models in predicting coronary heart issues, Alzheimer's brain related issues, diabetes, and lung cancer. This review study discusses various studies on the use of deep learning models in medical image analysis, heart disease prediction, and brain tumor detection. The study discusses various methods, such as hybrid machine learning, ensemble machine learning, multi-source transfer learning, convolutional neural networks, and cloud-enabled access control models. It also discusses the role of advanced learning models such as AI and deep learning in early detection of chronic diseases and the spread infection-based diseases. The study also discusses the challenges and opportunities in using machine learning in cardiovascular disease prediction, with some focusing on the use of jellyfish optimization algorithm and others on the use of boosting techniques. The review concludes by focus attention on the importance of understanding the role of machine learning in healthcare and the potential for future advancements in this field.

Introduction: Management of non-small cell lung cancer (NSCLC) depends on the presence and site of distant metastasis. Machine learning (ML) models have shown promise in analyzing clinical and molecular data to predict cancer outcomes and prognosis. Advancements in single-cell RNA sequencing (scRNA) provide deeper understanding of the tumor microenvironment (TME) and in metastasis. We aim to analyze NSCLC metastasis across different organs, investigate the clinical and molecular predictors of distant organ metastasis, and the single-cell landscape driving tumor metastasis. Methods: We used the cBioportal and the Gene Expression Omnibus (GEO) databases to retrieve clinical and scRNA data from the MSKCC Metastasis Cohort and GSE131907. The MSKCC dataset included 4, 601 NSCLC patients with distant organ metastasis to adrenal, bone, CNS/brain, liver, lung, mediastinum, and pleura. Seven random forest classifiers (RFC) were built for each organ metastasis and trained on clinical and genomic features like age, sex, race, histology, overall survival, fraction genome altered (FGA), microsatellite instability (MSI), mutation count, tumor mutational burden (TMB), and tumor purity. Models were evaluated using area under the receiver operating characteristics curve (AUC/ROC), 10-fold cross-validation (CV), accuracy score, and feature importance. Single-cell RNA data was filtered to include primary tumor samples (n=65, 702) and brain metastasis (n=28, 429) for cell distribution analysis. Results: Distant metastasis occurred in 3, 784 (82%) patients. The most common metastatic sites were lung (41%), bone (38%), pleura (35%), CNS (26%), liver (18%), adrenal gland (12%), and mediastinum (8.7%). The RFC-Mediastinum model showed the highest accuracy (0.89), AUC (0.61), and 10-fold CV (0.90), followed by RFC-Adrenal (0.86, AUC=0.62, 10-fold CV=0.86). The lowest performance was seen in RFC-Lung (0.57, AUC=0.56, 10-fold CV=0.52). The most important predictive features across all models were FGA, MSI, TMB, and mutation count, while histology was the least important. Single-cell analysis revealed that brain metastasis had a predominance of stromal cells, particularly epithelial, compared to primary tumor samples. Primary tumor samples consisted mostly of CD8+ T-cells and macrophages. COL1A1/2 genes were significantly expressed in brain metastatic stromal cells, suggesting extracellular matrix remodeling in the TME. Conclusion: Our study highlights the potential of ML models in accurately predicting distant organ metastasis in NSCLC. Variations in TME cellular compositions may be driven by complex intercellular communications in metastatic disease. Citation Format: Zaid Nassar Abu Rjai’, Ahmad Majid Bani-Ata, Ayah N. AL-Bzour. Machine learning prediction and single-cell landscape of distant organ metastasis in non-small cell lung cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 5086.

Digital twin technology is revolutionizing the modeling and simulation of complex systems, such as hydrogen leakage. This paper focuses on integrating digital twin components, starting with experimental measurements. The collected data is then used to calibrate the corresponding physics-based model, improving the accuracy of predictions. A numerical simulator is developed to enable virtual experiments and performance analysis. Comprehensive datasets are generated from real and virtual experiments, enhancing the capabilities of the twin by providing diverse training data. Machine learning prediction utilizes real and artificial datasets to gain insights into system behavior. The bidirectional relationship between the numerical simulator and machine learning prediction enhances accuracy and enriches the overall process. Viewed through the lens of science fiction, the case study on hydrogen leakage can be sculpted into a compelling narrative intertwining advanced technology, human resilience, and the ever-present challenge of harnessing and managing powerful, futuristic energy sources deep within planets.

With the rapid development of information technology in recent years, it has become common for consumers to purchase various products via electric commerce (EC) sites. As a case study, this study focuses on ZOZOUSED, which is engaged in the business of buying used clothes from users, and reselling them as second-hand goods. From the perspective of inventory and management costs, it is desirable for items to be sold as soon as possible after they are listed on EC sites, and the number of listed items has been conventionally controlled, depending on the experience of item managers. However, owing to the subjective assessment of item managers, unnecessary price reductions for sales promotion of items, or opportunity losses triggered by an excessive number of listed items, may occur. For reasonable item management, the demand prediction for items by customers is a crucial task required to develop the optimal listing plan that balances supply and demand. Therefore, this study proposes a forecasting method of sales figures for the actual operation of listing second-hand goods, which comprises two-stage models: the first model is a prior seasonal long-term prediction of sales figures for each item group based on seasonal similarity, and the second model is a short-term fine-tuning prediction for daily operation via residual predictions with recent data. Furthermore, we apply the proposed model to the actual data of past sales figures accumulated in ZOZOUSED, and analyze the obtained results to demonstrate the usefulness of the proposed method. In addition, we empirically demonstrate the effectiveness of the proposed method by designing and performing an empirical experiment on an actual business by applying the output of the proposed method as a new index for determining the number of new items to be listed.

Machine learning methods such as multilayer perceptrons (MLP) and Convolutional Neural Networks (CNN) have emerged as promising methods for genomic prediction (GP). In this context, we assess the performance of MLP and CNN on regression and classification tasks in a case study with maize hybrids. The genomic information was provided to the MLP as a relationship matrix and to the CNN as “genomic images.” In the regression task, the machine learning models were compared along with GBLUP. Under the classification task, MLP and CNN were compared. In this case, the traits (plant height and grain yield) were discretized in such a way to create balanced (moderate selection intensity) and unbalanced (extreme selection intensity) datasets for further evaluations. An automatic hyperparameter search for MLP and CNN was performed, and the best models were reported. For both task types, several metrics were calculated under a validation scheme to assess the effect of the prediction method and other variables. Overall, MLP and CNN presented competitive results to GBLUP. Also, we bring new insights on automated machine learning for genomic prediction and its implications to plant breeding.