检索结果-内蒙古大学图书馆

Comparing disaggregation approaches DSMART and PPD in disaggregating soil series maps

Pedosphere 2025年第2期35卷 387-404页

作者： Tahmid Huq EASHER Daniel SAURETTE Brandon HEUNG Adam GILLESPIE Richard J.HECK Asim BISWAS School of Environmental Sciences University of Guelph Department of Plant Food and Environmental Sciences Faculty of Agriculture Dalhousie University

Conventional soil maps(CSMs) often have multiple soil types within a single polygon, which hinders the ability of machine learning to accurately predict soils. Soil disaggregation approaches are commonly used to improve the spatial and attribute precision of CSMs. The approach disaggregation and harmonization of soil map units through resampled classification trees(DSMART) is popular but computationally intensive, as it generates and assigns synthetic samples to soil series based on the areal coverage information of CSMs. Alternatively, the disaggregation approach pure polygon disaggregation(PPD) assigns soil series based solely on the proportions of soil series in pure polygons in CSMs. This study compared these two disaggregation approaches by applying them to a CSM of Middlesex County, Ontario, Canada. Four different sampling methods were used: two sampling designs, simple random sampling(SRS) and conditional Latin hypercube sampling(cLHS), with two sample sizes(83 100 and 19 420 samples per sampling plan), both based on an area-weighted approach. Two machine learning algorithms(MLAs), C5.0 decision tree(C5.0) and random forest(RF), were applied to the disaggregation approaches to compare the disaggregation accuracy. The accuracy assessment utilized a set of 500 validation points obtained from the Middlesex County soil survey report. The MLA C5.0(Kappa index = 0.58–0.63) showed better performance than RF(Kappa index = 0.53–0.54) based on the larger sample size,and PPD with C5.0 based on the larger sample size was the best-performing(Kappa index = 0.63) approach. Based on the smaller sample size, both cLHS(Kappa index = 0.41–0.48) and SRS(Kappa index = 0.40–0.47) produced similar accuracy results. The disaggregation approach PPD exhibited lower processing capacity and time demands(1.62–5.93 h) while yielding maps with lower uncertainty as compared to DSMART(2.75–194.2 h). For CSMs predominantly composed of pure polygons, utilizing PPD for soil series disaggregatio

关键词： conditioned Latin hypercube sampling conventional soil map machine learning algorithm processing capacity and time sample size simple random sampling soil map unit soil series disaggregation

来源：评论

学校读者我要写书评

暂无评论

Model Updating of Cultural Heritage Buildings Through Swarm Intelligence algorithms

引用

INTERNATIONAL JOURNAL OF ARCHITECTURAL HERITAGE 2025年第2期19卷 259-275页

作者： Gara, Fabrizio Nicoletti, Vanni Arezzo, Davide Cipriani, Leonardo Leoni, Graziano Univ Politecn Marche Dept Construct Civil Engn & Architecture DICEA Ancona Italy Univ Camerino Sch Architecture & Design SAAD Ascoli Piceno Italy

Masonry buildings constitute a great Italian historical and cultural heritage, but they were also severely damaged by earthquakes over the centuries. Therefore, to assess their structural and seismic performance, it is crucial to gain a refined and trustworthy numerical model adopting model updating techniques, sometimes based on artificial intelligence algorithms. This article deals with the development and the updating of a finite element model of an historical church that account for the presence of both the seismic damage and securing systems. The model updating is performed adopting the particle swarm optimization algorithm and is based on the comparison between numerical and experimental modal parameters, the latter achieved by an extensive dynamic test campaign. The obtained calibrated numerical model has been adopted to support the restoration work design, as well as the design of a structural health monitoring system that has been permanently installed on the church.

关键词： model updating Ambient vibration tests cultural heritage building historical masonry church machine learning algorithm particle swarm optimization

来源：评论

学校读者我要写书评

暂无评论

Predicting social welfare in Madrid neighbourhoods using machine learning

引用

REGIONAL STUDIES REGIONAL SCIENCE 2024年第1期11卷 496-522页

作者： Rodriguez, Carlos Alberto Lastras CUNEF Univ Dept Quantitat Methods Madrid Spain

This research aimed to identify the principal factors predicting social welfare and inequality in the neighbourhoods of Madrid city. A comprehensive dataset representing various socioeconomic metrics of Madrid's neighbourhoods is analysed utilising different linear regression models and the XGBoost machine learning algorithm. The findings indicate that demographic variables play a crucial role in shaping social welfare and inequality in Madrid's neighbourhoods, with the percentage of women, and the percentage of children under 14 years old and adults over 65 years old being the most important variables for predicting social welfare and inequality in the studied neighbourhoods. Contrary to expectations, the average net income per household does not emerge as a significant predictor. Policymakers and urban planners should consider these factors to ensure equitable development and welfare distribution. The low impact of average net income per household on social welfare and inequality observed in this study should be regarded with some caution, but our findings certainly underline the need for a deeper understanding of socioeconomic dynamics in urban settings.

关键词： Social welfare and equality city indicators major global city machine learning algorithm XGBoost gentrification B55 C49 D63 I30 I31 J16

来源：评论

学校读者我要写书评

暂无评论

A hybrid model for improving customer lifetime value prediction using stacking ensemble learning algorithm

COMPUTERS IN HUMAN BEHAVIOR REPORTS

引用

COMPUTERS IN HUMAN BEHAVIOR REPORTS 2025年 18卷

作者： Haddadi, Amir Mohammad Hamidi, Hodjat KN Toosi Univ Technol Dept Ind Engn Tehran Iran KN Toosi Univ Technol Dept Ind Engn Informat Technol Grp Tehran Iran

A significant challenge that analysts and marketing managers often face is predicting future customer buying behavior. Identifying customers who are likely to make purchases down the line and estimating how much they will spend can help companies create more effective marketing campaigns and special offers that not only boost profits but also improve the overall customer experience and contribute to building lasting relationships. This paper outlines a comprehensive, detailed methodology for the forecasting and analysis of customer behavior using advanced predictive models and machine learning techniques. This framework can assist an organization in making more strategic and data-driven decisions to enhance both performance and profitability while securing improved customer satisfaction and long-term loyalty. The primary goal of this research is to increase the accuracy of Customer Lifetime Value (CLV) predictions and to provide deeper insights into customer behaviors. To achieve this objective, a combination of four key metrics is employed, namely: predicted purchases estimated using the BG-NBD model, predicted average value calculated through the Gamma-Gamma model, customer clustering labels assigned via the K-means model, and the customer status index derived from the Markov chain model. These essential metrics are subsequently integrated into the dataset. In order to further improve the accuracy of the predictions, this research uses a Stacking Ensemble model that combines four algorithms, i.e., Elastic Net, Random Forest, XGBoost, and SVM. The results demonstrate that integrating those features and using the Stacking Ensemble model substantially increases the prediction accuracy and decreases the errors. Sensitivity analysis and feature importance also prove the effectiveness of the method.

关键词： Customer lifetime value machine learning algorithm CLV prediction Stacking ensemble learning BG-NBD model K -means algorithm

来源：评论

学校读者我要写书评

暂无评论

Nowcasting Chinese GDP in a data-rich environment: Lessons from machine learning algorithms

引用

ECONOMIC MODELLING 2023年第1期122卷

作者： Zhang, Qin Ni, He Xu, Hao Zhejiang Gongshang Univ Tailong Finance Sch 18 Xuezheng Str Hangzhou 310018 Peoples R China Zhejiang Gongshang Univ Collaborat Innovat Ctr Stat Data Engn Technol & Ap Hangzhou Peoples R China Fintech Res Ctr Zhejiang Lab Shanghai Peoples R China

The ability to estimate current GDP growth before official data are released, known as "nowcasting", is crucial for the Chinese government to effectively implement economic policy and manage economic uncertainties;however, there is limited research on nowcasting China's GDP in a data-rich environment. We evaluate the performance of various machine learning algorithms, dynamic factor models, static factor models, and MIDAS regressions in nowcasting the Chinese annualised real GDP growth rate in pseudo out-of-sample exercise, using 89 macroeconomic variables from years 1995 to 2020. We find that some machine learning methods outperform the benchmark dynamic factor model. The machine learning method that deserves more attention is ridge regression, which dominates all other models not only in terms of nowcast error but also in effective recognition of the impacts of the Global Financial Crisis and Covid-19 shocks. Policy-wise, our study guides practitioners in selecting appropriate nowcasting models for China's macroeconomy.

关键词： Nowcasting China?s macroeconomy machine learning algorithm Dynamic factor model Real GDP

来源：评论

学校读者我要写书评

暂无评论

Enhancing Heart Disease Prediction: A Novel Deep learning Approach with Artificial Neural Networks 2

Enhancing Heart Disease Prediction: A Novel Deep Learning Ap...

引用

2nd International Conference on Sustainable Computing and Smart Systems (ICSCSS)

作者： Shamini, Baby P. Gold, Jemi P. Mahalakshmi, K. Shanker, M. C. Devakadacham, Sterlin Rani Sruthi, S. RMK Engn Coll Dept Comp Sci & Engn Kavaraipettai India RMK Coll Engn & Technol Dept Comp Sci & Engn Kavaraipettai India Vel Tech High Tech Dr Rangarajan Dr Sakunthala En Dept Comp Sci & Engn Chennai Tamil Nadu India Saveetha Inst Med & Tech Sci SIMATS Saveetha Sch Engn Dept Elect & Commun Engn Chennai Tamil Nadu India RMD Engn Coll Dept Comp Sci & Engn Kavaraipettai India

ISBN: (纸本)9798350391558;9798350379990

The crucial role of the heart in all living beings necessitates a high degree of accuracy in diagnosing and predicting heart disease. Even minor errors can lead to serious consequences, highlighting the importance of developing efficient methods for heart disease prediction. Deep learning techniques offer a promising approach in this regard. Early identification of cardiac disease markers is vital for clinical decision-making and paves way for further diagnostic evaluation. However, traditional manual analysis is time-consuming and laborious, especially when dealing with large datasets. Here, this study proposes a novel model using deep learning techniques to achieve rapid and effective heart disease diagnosis based on various physiological markers. The proposed model demonstrates exceptional performance on both training and testing data. We employ Artificial Neural Networks (ANN) to assess the model's efficacy in accurately predicting the presence or absence of heart disease. The proposed approach offers several advantages, including high success rates, superior performance and accuracy compared to traditional methods like Random Forest and SVM, flexibility, and adaptability.

关键词： Deep learning ANN SVM machine learning algorithm

来源：评论

学校读者我要写书评

暂无评论

Room-temperature rapid oxygen monitoring system in high humidity hydrogen gas environment towards water electrolysis application

引用

SENSORS AND ACTUATORS B-CHEMICAL 2025年 422卷

作者： Kwon, Yeongjae Lee, Kichul Kang, Mingu Kim, Cheolmin Ha, Ji-Hwan Han, Hyeonseok Yang, Seungki Yang, Daejong Seo, Jung Hwan Park, Inkyu Korea Adv Inst Sci & Technol KAIST Dept Mech Engn 291 Daehak Ro Daejeon 34141 South Korea Dept Nanomfg Technol Korea Inst Machinery & Mat KI Dept Nanomfg Technol 156 Gajeongbuk Ro Daejeon 34103 South Korea Hyundai Motor Co Water Electrolyzer Engn Design Team 17-5 Mabuk Ro 240 beon Gil Yongin 16891 South Korea Kongju Natl Univ Dept Mech & Automot Engn 1223-24 Cheonan Daero Cheonan Si 31080 Chungcheongnam South Korea Hongik Univ Dept Mech & Syst Design Engn 94 Wausan Ro Seoul South Korea

Water electrolysis is one of the key processes for carbon-free hydrogen (H-2) generation, yet it inherently presents the risk of oxygen (O-2) permeation, a significant concern that must be addressed. Contrary to external atmospheric conditions, the environment within a water electrolysis system is characterized exclusively by high-purity H-2, little O-2, and moisture (H2O). However, there has been limited research on sensors capable of detecting O-2 crossover in such a pure H-2 atmosphere under high humidity conditions. In this study, we developed an electrolysis O-2 monitoring (E-OM) sensor based on a semiconducting metal oxide (SMO) that can operate at a room-temperature with such high humidity through photoactivation with 365 nm LED light. A porous In2O3 nanofilm, prepared by glancing angle deposition (GLAD), was used as the gas sensing material, and copper nanoparticles (Cu NPs) were decorated on the In2O3 surface through electron beam evaporation to accelerate the O-2 adsorption. Remarkably, the E-OM sensor showed a humidity-independent response to O-2 gas in the H-2 atmosphere. Meanwhile, since the response of the E-OM sensor itself was not fast enough for water electrolysis system application, convolutional neural network (CNN)-based signal processing in the time domain was applied. As a result, our E-OM sensor system could predict O-2 concentrations from 0 to 2 vol% in an H-2 environment with a relative humidity (RH) of 30-90 %, and the detection time was under 5 s. This development could enable safe, rapid, and cost-effective monitoring of O-2 crossover in H-2 production infrastructure.

关键词： Photoactivated gas sensor Metal oxide Water electrolysis system Glancing angle deposition machine learning algorithm

来源：评论

学校读者我要写书评

暂无评论

Porous Silicon Particle-Assisted Mass Spectrometry Technology Unlocks Serum Metabolic Fingerprints in the Progression From Chronic Hepatitis B to Hepatocellular Carcinoma

引用

ACS APPLIED MATERIALS & INTERFACES 2025年第4期17卷 5893-5908页

作者： Jiang, Xinrong Tao, Liye Cao, Shuo Xu, Zhengao Zheng, Shuang Zhang, Huafang Xu, Xinran Qu, Xuetong Liu, Xingyue Yu, Jiekai Chen, Xiaoming Wu, Jianmin Liang, Xiao Zhejiang Univ Sir Run Run Shaw Hosp Sch Med Hangzhou 310016 Zhejiang Peoples R China Zhejiang Univ Dept Gen Surg Zhejiang Key Lab Multi Precis Diag & Treatment Liv Sir Run Run Shaw HospSch Med Hangzhou 310016 Peoples R China Zhejiang Univ Inst Analyt Chem Dept Chem Hangzhou 310058 Zhejiang Peoples R China Well Healthcare Technol Co Hangzhou 310051 Peoples R China Zhejiang Univ Biomed Res Ctr Sir Run Run Shaw Hosp Sch Med Hangzhou 310016 Zhejiang Peoples R China Taizhou First Peoples Hosp Taizhou 318020 Zhejiang Peoples R China Wuyi First Peoples Hosp Jinhua 321200 Zhejiang Peoples R China Zhejiang Univ Sch Med Key Lab Canc Prevent & Intervent Canc InstChina Natl Minist EducAffiliated Hosp 2 Hangzhou 310000 Zhejiang Peoples R China Shaoxing Univ Sch Med Shaoxing 312000 Zhejiang Peoples R China Hangzhou Med Coll Sch Basic Med Sci & Forens Med Hangzhou 310000 Zhejiang Peoples R China Zhejiang Univ Childrens Hosp Sch Med Hangzhou 310052 Peoples R China

Hepatocellular carcinoma (HCC) is a common malignancy and generally develops from liver cirrhosis (LC), which is primarily caused by the chronic hepatitis B (CHB) virus. Reliable liquid biopsy methods for HCC screening in high-risk populations are urgently needed. Here, we establish a porous silicon-assisted laser desorption ionization mass spectrometry (PSALDI-MS) technology to profile metabolite information hidden in human serum in a high throughput manner. Serum metabolites can be captured in the pore channel of APTES-modified porous silicon (pSi) particles and well-preserved during storage or transportation. Furthermore, serum metabolites captured in the APTES-pSi particles can be directly detected on the LDI-MS without the addition of an organic matrix, thus greatly accelerating the acquisition of metabolic fingerprints of serum samples. The PSALDI-MS displays the capability of high throughput (5 min per 96 samples), high reproducibility (coefficient of variation <15%), high sensitivity (LOD similar to 1 pmol), and high tolerance to background salt and proteins. In a multicenter cohort study, 1433 subjects including healthy controls (HC), CHB, LC, and HCC volunteers were enrolled and nontargeted serum metabolomic analysis was performed on the PSALDI-MS platform. After the selection of feature metabolites, a stepwise diagnostic model for the classification of different liver disease stages was constructed by the machine learning algorithm. In external testing, the accuracy of 91.2% for HC, 71.4% for CHB, 70.0% for LC, and 95.3% for HCC was achieved by chemometrics. Preliminary studies indicated that the diagnostic model constructed from serum metabolic fingerprint also displays good predictive performance in a prospective observation. We believe that the combination of PSALDI-MS technology and machine learning may serve as an efficient tool in clinical practice.

关键词： porous silicon laser desorption ionization serum metabolic fingerprinting liver disease diagnosis machine learning algorithm

来源：评论

学校读者我要写书评

暂无评论

Modelling pinyon pine distribution in the northern Great Basin, USA

引用

ENVIRONMENTAL CONSERVATION 2024年第3期51卷 188-195页

作者： Russell, Jacob M. Miller, Robert A. Carlisle, Jay D. Kaltenecker, Gregory S. Boise State Univ Intermt Bird Observ Boise ID 83725 USA

Pinyon-juniper woodlands are dry ecosystems defined by the presence of juniper (Juniperus spp.) and pinyon pine (Pinus spp.), which stretch over 400 000 km2 across 10 US states. Certain areas have become unnaturally dense and have moved into former shrub and grasslands, while others have experienced widespread mortality. To properly manage these woodlands, sites must be evaluated individually and decisions made based on scientific information that is often not available. Many species utilize pinyon-juniper woodlands, including the pinyon jay (Gymnorhinus cyanocephalus), named for its mutualism with pinyon pine, whose population has declined by c. 2.2% per year from 1966 to 2022, an overall decrease of c. 71%. To increase the likelihood of further research progress, we propose a tool to model the distribution of pinyon pine at a finer scale than current woodland classification tools in the northern US Great Basin: a random forest model using geographical, ecological and climate variables. Our results achieved an accuracy of 93.94%, indicating high predictive power to identify locations of pinyon pine in north-eastern Nevada, the south-eastern corner of Oregon and southern Idaho. These findings can inform managers and planners researching pinyon pine, pinyon-juniper woodlands and potentially the pinyon jay.

关键词： Gymnorhinus cyanocephalus Idaho machine learning algorithm Nevada Oregon Pinus edulis Pinus monophylla pinyon jay random forest Utah

来源：评论

学校读者我要写书评

暂无评论

Development of adaptive neuro-fuzzy inference system and stochastic simulation for mean droplet size modeling in rotatory agitated columns under different mass transfer scenarios

引用

INTERNATIONAL COMMUNICATIONS IN HEAT AND MASS TRANSFER 2024年 158卷

作者： Shakib, Benyamin Khiadani, Mehdi Petranikova, Martina Jyothi, Rajesh Kumar Lee, Jin-Young Korea Inst Geosci & Mineral Resources KIGAM Resources Utilizat Res Div 124 Gwahak Ro Daejeon 34132 South Korea Korea Univ Sci & Technol UST Dept Resources Engn 217 Gajeong Ro Daejeon 34113 South Korea Edith Cowan Univ Sch Engn 270 Joondalup Dr Perth WA 6027 Australia Chalmers Univ Technol Dept Chem & Chem Engn Kemivagengen 4 SE-41296 Gothenburg Sweden Australian Mineral Res Ctr AMRC CSIRO Mineral Resources CMR 7 Conlon St Waterford Perth WA 6102 Australia

The liquid-liquid extraction types of equipment are of substantial significance in industrial fields due to their effectiveness in facilitating the separation and purification of substances. Different types of rotatory agitated columns, such as RDC, ARDC, PRDC, Oldshue-Rushton, K & uuml;hni, and Scheibel, are evaluated for their performance. The nonlinearity and complexity of the mechanisms in these columns present challenges for accurately predicting the droplet behavior using available empirical models. Hence, this paper developed machine learning (ML) models, adaptive neuro-fuzzy inference system (ANFIS), artificial neural network (ANN), support vector regression (SVR), and gradient boosting algorithm (GBA) utilizing a dataset of 1135 experimental points to predict mean droplet size (d32) 32 ) under different mass transfer conditions. The ANFIS model surpasses ANN, SVR, and GBA models, demonstrating enhanced precision and resilience. The mean squared error (MSE) values for the ANFIS were recorded at 0.004 and 0.009, while the root mean squared error (RMSE) values stood at 0.065 and 0.094 during both the training and validation stages, respectively. Subsequently, the Monte Carlo simulation (MCS) was utilized, employing the PERT distribution function to model uncertain variables within the ANFIS model, considering their asymmetric distributions. The simulations revealed a mean d 32 of 2.0482 mm with a 90% confidence interval of [2.0398 mm, 2.0567 mm], a standard deviation of 0.5154 mm, skewness of-0.5882, and kurtosis of 3.6297. The sensitivity analysis based on permutation importance highlighted agitation speed as the most influential parameter, followed by fractional free area and interfacial tension, indicating their significant impact on d32. 32 . The optimized model's capacity to efficiently manage the intricacies and non-linearities present in dispersed phase droplets represents the first comprehensive progression from conventional empirical approaches, openi

关键词： Extraction columns Hydrodynamic Mass transfer Mean droplet size machine learning algorithm Risk analysis

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：