Machine learning with applications最新文献

Deep learning-based identification of precipitation clouds from all-sky camera data for observatory safety

Machine learning with applications Pub Date : 2025-03-19 DOI: 10.1016/j.mlwa.2025.100640

Mohammad H. Zhoolideh Haghighi , Alireza Ghasrimanesh , Habib Khosroshahi

{"title":"Deep learning-based identification of precipitation clouds from all-sky camera data for observatory safety","authors":"Mohammad H. Zhoolideh Haghighi , Alireza Ghasrimanesh , Habib Khosroshahi","doi":"10.1016/j.mlwa.2025.100640","DOIUrl":"10.1016/j.mlwa.2025.100640","url":null,"abstract":"<div><div>For monitoring the night sky conditions, wide-angle all-sky cameras are used in most astronomical observatories to monitor the sky cloudiness. In this manuscript, we apply a deep-learning approach for automating the identification of precipitation clouds in all-sky camera data as a cloud warning system. We construct our original training and test sets using the all-sky camera image archive of the Iranian National Observatory (INO). The training and test set images are labeled manually based on their potential rainfall and their distribution in the sky. We train our model on a set of roughly 2445 images taken by the INO all-sky camera through the deep learning method based on the EfficientNet network. Our model reaches an average accuracy of 99% in determining the cloud rainfall’s potential and an accuracy of 96% for cloud coverage. To enable a comprehensive comparison and evaluate the performance of alternative architectures for the task, we additionally trained three models—LeNet, DeiT, and AlexNet. This approach can be used for early warning of incoming dangerous clouds toward telescopes and harnesses the power of deep learning to automatically analyze vast amounts of all-sky camera data and accurately identify precipitation clouds formations. Our trained model can be deployed for real-time analysis, enabling the rapid identification of potential threats, and offering a scaleable solution that can improve our ability to safeguard telescopes and instruments in observatories. This is important now that numerous small- and medium-sized telescopes are increasingly integrated with smart control systems to reduce manual operation.</div></div>","PeriodicalId":74093,"journal":{"name":"Machine learning with applications","volume":"20 ","pages":"Article 100640"},"PeriodicalIF":0.0,"publicationDate":"2025-03-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143683247","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Software engineering meets legal texts: LLMs for auto detection of contract smells

Machine learning with applications Pub Date : 2025-03-18 DOI: 10.1016/j.mlwa.2025.100639

Moriya Dechtiar , Daniel Martin Katz , Hongming Wang

{"title":"Software engineering meets legal texts: LLMs for auto detection of contract smells","authors":"Moriya Dechtiar , Daniel Martin Katz , Hongming Wang","doi":"10.1016/j.mlwa.2025.100639","DOIUrl":"10.1016/j.mlwa.2025.100639","url":null,"abstract":"<div><div>Although there have been many major advances in Artificial Intelligence including its application to a wide variety of tasks, some specialized domains remain difficult to tackle. In this work, we examine parallels between software engineering and legal contract drafting and analysis. Porting well-known code smells principles to various legal contracts, we introduce ”contract smells,” text patterns that are indicative of potentially significant issues within contractual agreements. We leverage semi-auto labeling with GPT-4, prompting and expert spot checks, to create datasets for suitability testing of auto detection of these contract smells. Using transformer-based models, we explore the impact of legal domain knowledge, hyperparameters fine tuning and specific task information on detection success. We achieve high accuracy with further fine-tuning of BERT as well as LEGAL-BERT, while more consistent results were achieved using task-specific data. We further demonstrate that although multi-class detection can boost coverage of rare smells, single-class detection yields better accuracy. While this is an initial foray into the idea of contract smells, this work underscores the feasibility of applying advanced NLP techniques and LLMs to automate aspects of legal contract review, suggesting a scalable path toward standardized, machine-assisted legal drafting and analysis.</div></div>","PeriodicalId":74093,"journal":{"name":"Machine learning with applications","volume":"20 ","pages":"Article 100639"},"PeriodicalIF":0.0,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143683250","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Spatiotemporal Risk Mapping of Statewide Weather-related Traffic Crashes: A Machine Learning Approach

Machine learning with applications Pub Date : 2025-03-15 DOI: 10.1016/j.mlwa.2025.100642

Abimbola Ogungbire , Srinivas S. Pulugurtha

{"title":"Spatiotemporal Risk Mapping of Statewide Weather-related Traffic Crashes: A Machine Learning Approach","authors":"Abimbola Ogungbire , Srinivas S. Pulugurtha","doi":"10.1016/j.mlwa.2025.100642","DOIUrl":"10.1016/j.mlwa.2025.100642","url":null,"abstract":"<div><div>Improving transportation safety statewide is key in upholding a state's economy. However, weather-related crashes, known to be one of the most severe types of crashes, poses a threat to this as lots of money is lost to lives and property damage. The goal of this study is to employ machine learning (ML) to develop a workflow on which weather-related crash risk can be better identified, predicted, and interpreted. Central to this workflow, the effects of spatiotemporal heterogeneity of weather-related crashes are studied. To demonstrate the workflow, weather-related crash events in the state of North Carolina were used. Space-time cubes were created using an optimized 5 mi x 5mi grid size and 1-month time aggregation. Equivalent property damage only (EPDO) scores were computed for each space-time cube to create a risk metric that combines both crash frequency and severity. A two-layered technique was employed for identifying and labelling crash risk zones. Subsequently, XGBoost model was used to predict crash risk zones and identify factors associated with the different risk levels. SHapley Additive exPlanations (SHAP), an explainable AI (XAI) tool, was used to interpret the model and examine the relationship between the explanatory variables and the outcome. Per the results, there are three optimal clusters with distinct variability of the impact of weather conditions that constitute the crash risk levels in the study area. The workflow can be used by transportation safety units within state departments of transportation (DOTs) to evaluate different safety risk levels, and the potential high-risk zones can be flagged for devising countermeasures (i.e., proactive risk mitigation strategies).</div></div>","PeriodicalId":74093,"journal":{"name":"Machine learning with applications","volume":"20 ","pages":"Article 100642"},"PeriodicalIF":0.0,"publicationDate":"2025-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143683249","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

One-shot generative distribution matching for augmented RF-based UAV identification

Machine learning with applications Pub Date : 2025-03-15 DOI: 10.1016/j.mlwa.2025.100638

Amir Kazemi , Salar Basiri , Volodymyr Kindratenko , Srinivasa Salapaka

{"title":"One-shot generative distribution matching for augmented RF-based UAV identification","authors":"Amir Kazemi , Salar Basiri , Volodymyr Kindratenko , Srinivasa Salapaka","doi":"10.1016/j.mlwa.2025.100638","DOIUrl":"10.1016/j.mlwa.2025.100638","url":null,"abstract":"<div><div>This work addresses the challenge of identifying Unmanned Aerial Vehicles (UAV) using radiofrequency (RF) fingerprinting in limited RF environments. The complexity and variability of RF signals, influenced by environmental interference and hardware imperfections, often render traditional RF-based identification methods ineffective. To address these complications, the study introduces the rigorous use of one-shot generative methods for augmenting transformed RF signals, offering a significant improvement in UAV identification. This approach, when utilizing a distributional distance metric, demonstrates significant promise in low-data regimes, outperforming deep generative methods such as conditional generative adversarial networks (GANs) and variational autoencoders (VAEs). The paper provides a theoretical guarantee for the effectiveness of one-shot generative models in augmenting limited data, setting a precedent for their application in limited RF environments. This research also contributes to learning techniques in low-data regime scenarios, which may include complex sequences beyond images and videos. The code and links to datasets used in this study are available at <span><span>https://github.com/amir-kazemi/uav-rf-id</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":74093,"journal":{"name":"Machine learning with applications","volume":"20 ","pages":"Article 100638"},"PeriodicalIF":0.0,"publicationDate":"2025-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143683248","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Hybrid oversampling technique for imbalanced pattern recognition: Enhancing performance with Borderline Synthetic Minority oversampling and Generative Adversarial Networks

Machine learning with applications Pub Date : 2025-03-14 DOI: 10.1016/j.mlwa.2025.100637

Md Manjurul Ahsan , Shivakumar Raman , Yingtao Liu , Zahed Siddique

{"title":"Hybrid oversampling technique for imbalanced pattern recognition: Enhancing performance with Borderline Synthetic Minority oversampling and Generative Adversarial Networks","authors":"Md Manjurul Ahsan , Shivakumar Raman , Yingtao Liu , Zahed Siddique","doi":"10.1016/j.mlwa.2025.100637","DOIUrl":"10.1016/j.mlwa.2025.100637","url":null,"abstract":"<div><div>Class imbalance problems (CIP) are one of the potential challenges in developing unbiased Machine Learning models for predictions. CIP occurs when data samples are not equally distributed between two or multiple classes. Several synthetic oversampling techniques have been introduced to balance the imbalanced data by oversampling the minor samples. One of the potential drawbacks of existing oversampling techniques is that they often fail to focus on the data samples that lie at the border point and give more attention to the extreme observations, ultimately limiting the creation of more diverse data after oversampling, and that is almost the scenario for most of the oversampling strategies. As an effect, marginalization occurs after oversampling. To address these issues, in this work, we propose a hybrid oversampling technique, named Borderline Synthetic Minority Oversampling and Generative Adversarial Network (BSGAN), by combining the strengths of Borderline-Synthetic Minority Oversampling Technique (SMOTE) and Generative Adversarial Networks (GANs). This approach aims to generate more diverse data that follow Gaussian distributions, marking a significant contribution to the field of Artificial Intelligence. We tested BSGAN on ten highly imbalanced datasets, demonstrating its application in engineering, where it outperformed existing oversampling techniques, creating a more diverse dataset that follows a normal distribution after oversampling.</div></div>","PeriodicalId":74093,"journal":{"name":"Machine learning with applications","volume":"20 ","pages":"Article 100637"},"PeriodicalIF":0.0,"publicationDate":"2025-03-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143643032","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Advanced fault detection in photovoltaic panels using enhanced U-Net architectures

Machine learning with applications Pub Date : 2025-03-08 DOI: 10.1016/j.mlwa.2025.100636

Khalfalla Awedat , Gurcan Comert , Mustafa Ayad , Abdulmajid Mrebit

{"title":"Advanced fault detection in photovoltaic panels using enhanced U-Net architectures","authors":"Khalfalla Awedat , Gurcan Comert , Mustafa Ayad , Abdulmajid Mrebit","doi":"10.1016/j.mlwa.2025.100636","DOIUrl":"10.1016/j.mlwa.2025.100636","url":null,"abstract":"<div><div>Fault detection in photovoltaic (PV) panels using thermal images remains a significant challenge due to the complexity of thermal patterns, environmental noise, and the subtle nature of anomalies. This paper introduces an advanced deep learning framework that enhances the U-Net architecture by integrating Residual Blocks, Atrous Spatial Pyramid Pooling (ASPP), and Attention Mechanisms. These enhancements collectively improve feature extraction, contextual understanding, and fault localization, addressing the limitations of traditional segmentation approaches and reducing false positives. Extensive experiments demonstrate that the proposed method significantly outperforms all benchmarked algorithms across key segmentation metrics, including standard U-Net, U-Net with ASPP, and DeepLabV3+. Compared to standard U-Net, the proposed model achieves more than a 29% increase in F1-score and a 62% improvement in Intersection over Union (IoU) while reducing segmentation loss by 71%. Its ability to accurately detect faults under challenging conditions establishes the framework as a state-of-the-art solution for real-time PV monitoring. These results demonstrate the effectiveness of the proposed approach in addressing the challenges of PV fault detection, offering a practical and reliable solution for ensuring the operational performance of renewable energy systems.</div></div>","PeriodicalId":74093,"journal":{"name":"Machine learning with applications","volume":"20 ","pages":"Article 100636"},"PeriodicalIF":0.0,"publicationDate":"2025-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143580284","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Pioneering precision in lumbar spine MRI segmentation with advanced deep learning and data enhancement

Machine learning with applications Pub Date : 2025-03-08 DOI: 10.1016/j.mlwa.2025.100635

Istiak Ahmed , Md. Tanzim Hossain , Md. Zahirul Islam Nahid , Kazi Shahriar Sanjid , Md. Shakib Shahariar Junayed , M. Monir Uddin , Mohammad Monirujjaman Khan

{"title":"Pioneering precision in lumbar spine MRI segmentation with advanced deep learning and data enhancement","authors":"Istiak Ahmed , Md. Tanzim Hossain , Md. Zahirul Islam Nahid , Kazi Shahriar Sanjid , Md. Shakib Shahariar Junayed , M. Monir Uddin , Mohammad Monirujjaman Khan","doi":"10.1016/j.mlwa.2025.100635","DOIUrl":"10.1016/j.mlwa.2025.100635","url":null,"abstract":"<div><div>This study presents an advanced approach to lumbar spine segmentation using deep learning techniques, focusing on addressing key challenges such as class imbalance and data preprocessing. Magnetic resonance imaging (MRI) scans of patients with low back pain are meticulously preprocessed to accurately represent three critical classes: vertebrae, spinal canal, and intervertebral discs (IVDs). By rectifying class inconsistencies in the data preprocessing stage, the fidelity of the training data is ensured. The modified U-Net model incorporates innovative architectural enhancements, including an upsample block with leaky Rectified Linear Units (ReLU) and Glorot uniform initializer, to mitigate common issues such as the dying ReLU problem and improve stability during training. Introducing a custom combined loss function effectively tackles class imbalance, significantly improving segmentation accuracy. Evaluation using a comprehensive suite of metrics showcases the superior performance of this approach, outperforming existing methods and advancing the current techniques in lumbar spine segmentation. These findings hold significant advancements for enhanced lumbar spine MRI and segmentation diagnostic accuracy.</div></div>","PeriodicalId":74093,"journal":{"name":"Machine learning with applications","volume":"20 ","pages":"Article 100635"},"PeriodicalIF":0.0,"publicationDate":"2025-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143609206","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Key technical indicators for stock market prediction

Machine learning with applications Pub Date : 2025-03-01 DOI: 10.1016/j.mlwa.2025.100631

Seyed Mostafa Mostafavi , Ali Reza Hooman

{"title":"Key technical indicators for stock market prediction","authors":"Seyed Mostafa Mostafavi , Ali Reza Hooman","doi":"10.1016/j.mlwa.2025.100631","DOIUrl":"10.1016/j.mlwa.2025.100631","url":null,"abstract":"<div><div>The use of technical indicators for forecasting the stock market is widespread among investors and researchers. It is crucial to determine the optimal number of input technical indicators to predict the stock market successfully. However, there is no consensus on which collection of technical indicators is most suitable. The selection of technical indicators for a given forecasting model continues to be an active area of research. To our knowledge, there is limited published work on the importance of technical indicators in various categories such as momentum, trend, volatility, and volume. To identify the key technical indicators for stock market prediction, we employed XGBoost, Random Forest, Support Vector Regression, and LSTM regression techniques using 88 technical indicators as input data. We also used the PCA method for dimension reduction. The results reveal the most significant technical indicators within the momentum, trend, volatility, and volume categories. Our findings provide evidence that the proposed model is highly effective in predicting daily prices (with and without lag in Close price) on the S&P 500 stock index.</div></div>","PeriodicalId":74093,"journal":{"name":"Machine learning with applications","volume":"20 ","pages":"Article 100631"},"PeriodicalIF":0.0,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143519592","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Corrigendum to “Machine learning for sports betting: should model selection be based on accuracy or calibration?” [Machine Learning with Applications Volume 16, June 2024, 100539]

Machine learning with applications Pub Date : 2025-03-01 DOI: 10.1016/j.mlwa.2025.100627

Conor Walsh, Alok Joshi

引用次数: 0

Machine learning-driven predictive modeling of mechanical properties in diverse steels

Machine learning with applications Pub Date : 2025-02-28 DOI: 10.1016/j.mlwa.2025.100634

Movaffaq Kateb , Sahar Safarian

{"title":"Machine learning-driven predictive modeling of mechanical properties in diverse steels","authors":"Movaffaq Kateb , Sahar Safarian","doi":"10.1016/j.mlwa.2025.100634","DOIUrl":"10.1016/j.mlwa.2025.100634","url":null,"abstract":"<div><div>This study explores the application of machine learning (ML) in steel design using a small dataset of various steel grades that include 13 key elements and three critical mechanical properties. Random forest (RF) models were systematically evaluated for their robustness and effectiveness in predicting the stress-strain of steel properties. Moreover, other alternative approaches, such as support vector machines, extreme gradient boosting machines, and artificial neural networks, were also evaluated to ensure that the predictions made by the RF model are as accurate as possible. To assess the bias-variance trade-off, 1-seed and random 100-seeds with 80/20 train/test split, and leave-one-out cross-validation for all datasets were conducted. The results demonstrated that the RF models are accurate and reliable, achieving low bias and variance while delivering predictions comparable to, and in some cases better than, those obtained in studies with larger datasets. The analysis revealed a trade-off between strength and ductility, with elongation negatively correlated with yield strength and ultimate tensile strength. This study highlights the feasibility of using small, realistic datasets to develop effective ML models for predicting mechanical properties in steel design. The methodology can also be readily extended to investigate processing-property relationships in other systems, offering a versatile approach for advancing materials science through data-driven techniques.</div></div>","PeriodicalId":74093,"journal":{"name":"Machine learning with applications","volume":"20 ","pages":"Article 100634"},"PeriodicalIF":0.0,"publicationDate":"2025-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143551987","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0