检索结果-内蒙古大学图书馆

explainable machine learning for 2D material layer group prediction with automated descriptor selection

MATERIALS TODAY CHEMISTRY 2025年 44卷

作者： Sun, Ruijia Tang, Bijun Liu, Zheng Nanyang Technol Univ Sch Mat Sci & Engn N4 1-01-1050 Nanyang Ave Singapore 639798 Singapore

Crystal symmetry is a fundamental aspect of material properties and plays a pivotal role in the discovery and design of new materials. Existing approaches for predicting the symmetries of two-dimensional (2D) materials typically focus on space groups, which often overlook the distinct in-plane and out-of-plane symmetries inherent to 2D systems. To address this, we present Au2LaP (Automated Descriptor Selection Enhanced 2D Material Layer Group Predictor), the first machine learning framework designed to predict layer groups of 2D materials directly from their chemical composition. Au2LaP integrates Light Gradient Boosting machine (LightGBM) algorithm with SHapley Additive exPlanations (SHAP) for automated descriptor selection, optimizing both predictive accuracy and model interpretability. The explainability of Au2LaP ensures transparency by highlighting the most significant chemical descriptors contributing to layer group classification, thereby enhancing its utility for material discovery and design. Au2LaP outperforms seven state-of-the-art models based on chemical composition, achieving a top-1 accuracy of 0.8102 and a top-3 accuracy of 0.9048. Remarkably, it delivers superior performance using only 20 key descriptors, outperforming models trained with descriptor sets as large as 546. Furthermore, we demonstrate that Au2LaP can effectively predict polymorph structures by identifying multiple possible layer groups for given compositions, further advancing the predictive modelling of new material phases. This work sets a new benchmark for 2D material symmetry prediction, paving the way for more efficient crystal structure prediction, polymorph studies, and material design.

关键词： Two-dimensional materials Crystal symmetry layer group explainable machine learning Descriptor selection

来源：评论

学校读者我要写书评

暂无评论

explainable machine learning for predicting thermogravimetric analysis of oxidatively torrefied spent coffee grounds combustion

引用

ENERGY 2025年 320卷

作者： Pambudi, Suluh Jongyingcharoen, Jiraporn Sripinyowanich Saechua, Wanphut King Mongkuts Inst Technol Ladkrabang Sch Engn Dept Agr Engn Bangkok 10520 Thailand

Understanding the combustion behavior of oxidatively torrefied spent coffee grounds (SCG) is crucial for advancing sustainable fuel technologies. This study introduces a novel, explainable machine learning (ML) framework as a cost-effective alternative to traditional thermogravimetric analysis (TGA) that is designed to accelerate the evaluation of oxidatively torrefied SCG combustion properties. Four ML models: artificial neural network (ANN), k-nearest neighbor (k-NN), random forest (RF), and decision tree (DT), were compared to predict TGA data using proximate analysis and combustion temperature (CT). Among the evaluated models, kNN exhibited the highest performance, achieving near-perfect R2 values that exceeded 0.9904 and RMSE values below 0.9552 on the validation set for both TG (mass loss) and DTG (derivative mass loss). It also accurately predicted key combustion properties, including ignition, peak, and burnout temperature when tested on unknown data. LIME (Local Interpretable Model-agnostic Explanations) analysis revealed that CT was the most influential predictor for TG and DTG, enhancing model interpretability. The results highlight the effectiveness of the k-NN-LIME approach in analyzing the combustion of oxidatively torrefied SCG, offering a robust and explainable model with significant implications for bioenergy research and sustainable fuel development.

关键词： Biomass combustion explainable machine learning Oxidative torrefaction Spent coffee grounds Thermogravimetric analysis

来源：评论

学校读者我要写书评

暂无评论

explainable machine learning framework for predicting concrete abrasion depth

引用

CASE STUDIES IN CONSTRUCTION MATERIALS 2025年 22卷

作者： Moghaddas, Seyed Amirhossein Bao, Yi Stevens Inst Technol Dept Civil Environm & Ocean Engn Hoboken NJ 07030 USA

This paper presents an approach for predicting concrete abrasion depth based on an advanced machine learning-based framework with explainability. The framework integrates multiple data pre-processing efforts, such as feature selection and anomaly detection, into machine learning. Different ensemble learning algorithms XGBoost, CatBoost, LightGBM, Extremely Randomized Trees, and Random Forest are compared using performance metrics, including mean absolute error, root mean squared error, coefficient of determination, mean absolute deviation, mean percentage deviation, and scatter index for regression analysis. Results show that machine learning models achieve high performance. Feature importance analysis and SHAP analysis are performed to assess the effects of factors that affect concrete abrasion depth. The time of testing-to-velocity, water-to-binder ratio, and aggregate compositions are the top three influencing factors. This study demonstrates the potential of machine learning in predicting concrete abrasion depth, providing valuable insights for the design and maintenance of hydraulic concrete structures.

关键词： Abrasion depth explainable machine learning Concrete Ensemble learning Feature importance SHAP analysis

来源：评论

学校读者我要写书评

暂无评论

explainable machine learning models for corn yield prediction using UAV multispectral data☆

引用

COMPUTERS AND ELECTRONICS IN AGRICULTURE 2025年 231卷

作者： Kumar, Chandan Dhillon, Jagman Huang, Yanbo Reddy, Krishna Mississippi State Univ Dept Plant & Soil Sci Starkville MS 39759 USA USDA ARS Crop Prod Syst Res Unit Stoneville MS USA USDA ARS Genet & Sustainable Agr Res Unit Mississippi State MS USA

Accurate and reliable corn (Zea mays L.) yield prediction is essential for optimizing corn production management practices for closing yield gaps. Remote sensing data integrated with machine learning (ML) models have been widely used in crop yield prediction. However, conventional ML models are often criticized for their 'black-box' nature, impeding transparency into their internal processes and the rationale behind their predictions. To address this limitation, this study develops explainable ML models to predict corn yield using high-resolution multispectral data captured by Unmanned Aerial Vehicles. Among thirty variables derived from multispectral data, Simplified Canopy Chlorophyll Content Index and Transformed Chlorophyll Absorption Reflectance Index were found to be the most suitable inputs. To accurately predict corn yield, we evaluated the Generalized Linear Model (GLM), K-Nearest Neighbor (KNN), Principal Component Regression (PCR), Random Forest (RF), Support Vector machine (SVM), and Bayesian Regularized Neural Networks (BRNN). We further evaluated various ensemble models combined using GLM and RF. Among individual models, SVM outperformed others with R2 = 0.67 and RMSE = 1.56 Mg ha-1. Ensemble models combined using RF produced significantly higher prediction accuracy (R2 = 0.90 and RMSE = 0.86 Mg ha-1) than those combined using GLM (R2 = 0.70 and RMSE = 1.52 Mg ha-1). Explanation tools provided valuable insights into three key aspects of ML modeling: (a) the contribution of variables to model performance, (b) the influence of variables on model predictions, and (c) the distribution of residuals. These insights offer a comprehensive understanding of how ML models accurately predict corn yield, fostering trust in ML models, and promoting their adoption in precision agriculture.

关键词： Corn Ensemble model explainable machine learning Multispectral Yield prediction UAV Precision agriculture

来源：评论

学校读者我要写书评

暂无评论

explainable machine learning and feature engineering applied to nanoindentation data

引用

MATERIALS & DESIGN 2025年 253卷

作者： Trost, C. O. W. Zak, S. Schaffer, S. Walch, L. Zitz, J. Kluensner, T. Leitner, H. Exl, L. Cordill, M. J. Austrian Acad Sci Erich Schmid Inst Mat Sci Jahnstr 12 A-8700 Leoben Austria Univ Vienna Wolfgang Pauli Inst Fac Math Oskar Morgenstern Pl 1 A-1090 Vienna Austria Univ Vienna Res Platform MMM Math Magnetism Mat Oskar Morgenstern Pl 1 A-1090 Vienna Austria Leoben Forsch GmbH Mat Ctr Leoben Austria Voestalpine BOHLER Edelstahl GmbH & Co KG Mariazeller Str 25 A-8605 Kapfenberg Austria Univ Leoben Dept Mat Sci Jahnstr 12 A-8700 Leoben Austria

The work aims to challenge the hegemony in the literature of clustering nanoindentation data solely relying on elastic modulus and hardness as features, thereby discarding information provided by the full load-displacement curve. Features based on dimensional analysis initially aimed to solve the inverse nanoindentation problem were adopted to describe the load-displacement curves. More than 3000 indents in high-speed steels were labelled via imaging after indenting. The resulting dataset was used to train and benchmark supervised (classification) and unsupervised (clustering) machine learning models, showing that feature engineering was more impactful than model selection and hyperparameter tuning, increasing the prediction quality in all studied models. The best classifier's predictions were explained via a game theory-based approach, allowing insights into the model's decision-making process and connecting the fields of materials property clustering and materials mechanics.

关键词： Nanoindentation explainable machine learning High-speed steel Composite material

来源：评论

学校读者我要写书评

暂无评论

Impacts of land use characteristics on extreme heat events: Insights from explainable machine learning model

引用

SUSTAINABLE CITIES AND SOCIETY 2025年 120卷

作者： Su, Hangying Qi, Zhuoxu Wang, Qixuan Tongji Univ Coll Architecture & Urban Planning 1239 Siping Rd Shanghai 200000 Peoples R China Southeast Univ Sch Architecture Dept Urban Planning Nanjing 210096 Peoples R China

While previous research has extensively explored how land use characteristics affect urban heat islands, their effects on extreme heat events (EHEs) remain poorly understood. Using Gradient Boosting Decision Tree combined with Shapley Additive exPlanations (SHAP), we assess the relative contribution, as well as the nonlinear and interactive effects of land use characteristics on EHEs by analyzing time-series data from 280 Chinese cities during the period 2012-2020. SHAP is also employed to evaluate the local effects of land use in cities experiencing the most severe EHEs. The results show that: (1) apart from the climate group, the most important factor group influencing EHEs is land use characteristics;(2) the nonlinear relationships and interaction effects between land use characteristics and EHEs are common. For instance, land use intensity exhibits a U-shaped relationship with EHEs. Both high residential and commercial land shares have synergistic effects with high land use intensity that exacerbate EHEs;and (3) in Shantou, the city experiencing the most severe EHEs, land use variables account for half of the top ten factors influencing EHEs. These insights can assist in developing effective land use policies to grapple with the escalating challenges from heat waves in the context of climate change.

关键词： Extreme heat events Urban land use types Land use mix explainable machine learning China

来源：评论

学校读者我要写书评

暂无评论

Evaluating the affecting factors of glacier mass balance in Tanggula Mountains using explainable machine learning and the open global glacier model

引用

Journal of Mountain Science 2025年第2期22卷 466-488页

作者： XU Qiangqiang KANG Shichang HE Xiaobo XU Min Key Laboratory of Cryospheric Science and Frozen Soil Engineering Northwest Institute of Eco-Environment and ResourcesChinese Academy of SciencesLanzhou 730000China Tanggula Mountain Cryosphere and Environment Observation and Research Station of Tibet Autonomous Region Lanzhou 730000China University of Chinese Academy of Sciences Beijing 100049China

Glacier mass balance is a key indicator of glacier health and climate change *** factors include both climatic and nonclimatic elements,forming a complex set of *** is a lack of quantitative analysis of these composite factors,particularly in climate-typical regions like the Tanggula Mountains on the central Tibetan *** collected data on various factors affecting glacier mass balance from 2000 to 2020,including climate variables,topographic variables,geometric parameters,and glacier *** utilized linear regression models,ensemble learning models,and Open Global Glacier Model(OGGM)to analyze glacier mass balance changes in the Tanggula *** indicate that linear models explain 58%of the variance in glacier mass balance,with seasonal temperature and precipitation having significant *** findings show that ensemble learning models made the explanations 5.2%more accurate by including the impact of topographic and geometric factors such as the average glacier height,the slope of the glacier tongue,the speed of the ice flow,and the area of the *** machine learning identified the spatial distribution of positive and negative impacts of these characteristics and the interaction between glacier topography and ice ***,we predicted the responses of glaciers of different sizes to future climate change based on the results of interpretable machine *** was found that relatively large glaciers(>1 km~2)are likely to persist until the end of this century under low emission scenarios,whereas small glaciers(<1 km~2)are expected to nearly disappear by 2080 under any emission *** research provides technical support for improving glacier change modeling and protection on the Tibetan Plateau.

关键词： Glacier mass balance Tanggula Mountains explainable machine learning Open Global Glacier Model Climate change

来源：评论

学校读者我要写书评

暂无评论

Clinical impact of an explainable machine learning with amino acid PET imaging: application to the diagnosis of aggressive glioma

引用

EUROPEAN JOURNAL OF NUCLEAR MEDICINE AND MOLECULAR IMAGING 2025年第6期52卷 1989-2001页

作者： Ahrari, Shamimeh Zaragori, Timothee Zinsz, Adeline Hossu, Gabriela Oster, Julien Allard, Bastien Al Mansour, Laure Bessac, Darejan Boumedine, Sami Bund, Caroline De Leiris, Nicolas Flaus, Anthime Guedj, Eric Kas, Aurelie Keromnes, Nathalie Kiraz, Kevin Kuijper, Fiene Marie Maitre, Valentine Querellou, Solene Stien, Guilhem Humbert, Olivier Imbert, Laetitia Verger, Antoine Univ Lorraine INSERM IADI U1254 Nancy France Univ Lorraine INSERM U1254 & Nancyclotep Imaging Platform Nancy France Univ Lorraine C 1433 Innovat Technol CIC 1433Inserm Nancy France Ctr Hosp Reg Univ Nancy Dept Nucl Med Nancy France Ctr Hosp Valence Dept Med Valence France Hosp Civils Lyon Dept Nucl Med Lyon France ICANS Dept Nucl Med & Mol Imaging Strasbourg France Ctr Antoine Lacassagne Dept Nucl Med Nice France Univ Strasbourg ICube CNRS UMR 7357 Strasbourg France Ctr Hosp Univ Grenoble Alpes Ctr Hospitalier Univ Grenoble Alpes Grenoble France Univ Grenoble Alpes INSERM LRB Grenoble France Lyon Neurosci Res Ctr CNRS UMR5292 INSERM U1028 Lyon France Timone Hosp Dept Nucl Med Marseille France Aix Marseille Univ Inst Fresnel CNRS Cent MarseilleAPHPCERIMED Marseille France Grp Hop Pitie Salpetriere Assistance Publ Hop Paris AP HP Dept Internal Med Paris France Sorbonne Univ Lab Imagerie Biomed INSERM CNRS Paris France Univ Western Brittany UBO Ctr Hosp Reg Univ Brest CHRU Brest INSERMUMR 1304GETBO F-29200 Brest France Univ Cote DAzur INSERM CNRS iBV Nice France CHRU Nancy Hop Brabois Med Nucl Allee Morvan F-54500 Vandoeuvre Les Nancy France

Purpose Radiomics-based machine learning (ML) models of amino acid positron emission tomography (PET) images have shown efficiency in glioma prediction tasks. However, their clinical impact on physician interpretation remains limited. This study investigated whether an explainable radiomics model modifies nuclear physicians' assessment of glioma aggressiveness at diagnosis. Methods Patients underwent dynamic 6-[F-18]fluoro-L-DOPA PET acquisition. With a 75%/25% split for training (n = 63) and test sets (n = 22), an ensemble ML model was trained using radiomics features extracted from static/dynamic parametric PET images to classify lesion aggressiveness. Three explainable ML methods-Local Interpretable Model-agnostic Explanations (LIME), Anchor, and SHapley Additive exPlanations (SHAP)-generated patient-specific explanations. Eighteen physicians from eight institutions evaluated the test samples. During the first phase, physicians analyzed the 22 cases exclusively through magnetic resonance and static/dynamic PET images, acquired within a maximum interval of 30 days. In the second phase, the same physicians reevaluated the same cases (n = 22), using all available data, including the radiomics model predictions and explanations. Results Eighty-five patients (54[39-62] years old, 41 women) were selected. In the second phase, physicians demonstrated a significant improvement in diagnostic accuracy compared to the first phase (0.775 [0.750-0.802] vs. 0.717 [0.694-0.737], p = 0.007). The explainable radiomics model augmented physician agreement, with a 22.72% increase in Fleiss's kappa, and significantly enhanced physician confidence (p < 0.001). Among all physicians, Anchor and SHAP showed efficacy in 75% and 72% of cases, respectively, outperforming LIME (p <= 0.001). Conclusions Our results highlight the potential of an explainable radiomics model using amino acid PET scans as a diagnostic support to assist physicians in identifying glioma aggressiveness.

关键词： Glioma Positron emission tomography Radiomics explainable machine learning Interpretability

来源：评论

学校读者我要写书评

暂无评论

An explainable machine learning method for predicting and designing crashworthiness of multi-cell tubes under oblique load

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2025年 147卷

作者： Xie, Jian Zhang, Junyuan Dou, Zheng Chang, Mengge Su, Chang Jilin Univ Natl Key Lab Automot Chassis Integrat & Bion Changchun 130025 Peoples R China

Multi-cell tubes are widely used in energy-absorbing structures due to their excellent crashworthiness. However, oblique loading in real-world collisions can drastically change their deformation modes, leading to a sharp reduction in energy absorption. This paper proposed an explainable machine learning method combining Artificial Neural Network (ANN) and SHapley Additive exPlanations (SHAP) to quickly predict and design the crashworthiness of multi-cell tubes subjected to oblique loads. explainable models for six cross-section types were developed, with geometry, material, and loading angle as inputs, and three crashworthiness metrics as outputs. These models demonstrate high accuracy, with the Coefficient of Determination (R2) exceeding 0.91 and the Mean Absolute Percentage Error (MAPE) below 15.76, as well as significant computational efficiency, achieving a speed-up of up to 187,500 times relative to finite element (FE) simulations. Explainability analyses reveal critical nonlinear effects of loading angles, with section dimensions and material properties identified as key factors influencing crashworthiness. Based on the explainable global and local effects of design parameters, a fast optimization strategy for crashworthiness in multi-cell tubes was developed. This strategy was demonstrated using a single-cell tube (1CT) as an application example, with its design and optimization results validated by FE simulations. Compared to existing FE simulations and black-box models, this approach effectively combines fast predictions with explainable results, providing actionable insights for optimizing energy absorption in multi-cell tubes under oblique loads.

关键词： Multi-cell tubes Oblique load Crashworthiness explainable machine learning Artificial neural network SHapley additive exPlanations

来源：评论

学校读者我要写书评

暂无评论

An explainable machine learning (XML) approach to determine strength of glass powder concrete

引用

MATERIALS TODAY COMMUNICATIONS 2025年 45卷

作者： Ullah, Wali Bin Inqiad, Waleed Ayub, Bilal Khan, Muhammad Saud Javed, Muhammad Faisal Tongji Univ Coll Civil Engn Dept Bridge Engn Shanghai 200092 Peoples R China Aston Univ Coll Engn & Phys Sci Dept Civil Engn Aston St Birmingham B4 7ET England Natl Univ Sci & Technol NUST Mil Coll Engn MCE Dept Civil Engn Islamabad 44000 Pakistan Univ Manitoba Price Fac Engn Dept Civil Engn Winnipeg MB R3T 5V6 Canada Ghulam Ishaq Khan Inst Engn Sci & Technol Dept Civil Engn Topi 23640 Pakistan Western Caspian Univ Baku Azerbaijan

Glass powder concrete (GPC) holds the potential to reduce the damaging impact of construction industry on the natural environment by cutting down the amounts of cement and natural aggregates used in concrete. However, it requires having reliable ways for predicting key properties of GPC such as compressive strength (CS). The laboratory determination of GPC strength comes with various resource and time limitations due to which this study focused on using advanced machine learning (ML) techniques for predicting CS. Four algorithms including Multi Expression Programming (MEP), AdaBoost, HistGBR, and Extreme Gradient Boosting (XGB) were used for this purpose on a database retrieved from literature published internationally. The mixture composition of GPC like glass powder content, water-to-cement ratio, sand content along with concrete age etc. were used as predictors while CS was the only outcome. The algorithm's performance was validated by several techniques including k-fold cross validation, scatter plots, and residual distribution. Different error evaluation metrics were also utilized which revealed that XGB had the superior accuracy compared to its counterparts with a testing coefficient of determination (R2) value equal to 0.993. In contrast, MEP depicted its outcome as an equation which XGB and all the other algorithms could not provide. Moreover, interpretable ML techniques like shapely and Individual Conditional Expectation (ICE) analysis were employed for investigation of the significance and relationships between inputs and the predicted outcome. Furthermore, a graphical user interface has been developed which will enable the professionals in the civil engineering industry to utilize the findings of this study seamlessly.

关键词： Compressive strength machine learning Glass powder concrete explainable machine learning Prediction

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：