检索结果-内蒙古大学图书馆

A Novel Approach for Disaster Victim Detection Under Debris Environments Using decision tree algorithms With Deep Learning Features

引用

IEEE ACCESS 2023年 11卷 54760-54772页

作者： Seeja, G. Doss, Arockia Selvakumar Arockia Hency, V. Berlin Vellore Inst Technol Sch Elect Engn Chennai 600127 Tamil Nadu India Vellore Inst Technol Sch Mech Engn Chennai 600127 Tamil Nadu India

Search and Rescue operations for victim identification in an unstructured collapsed building are high-risk and time-consuming. The possibility of saving a victim is high only during the first 48 hours, and then the prospect tends to zero. The faster the response and identification, the sooner the victim can be taken to medical assistance. Combining mobile robots with practical Artificial Intelligence (AI) driven Human Victim Detection (HVD) systems managed by professional teams can considerably reduce this problem. In this paper, we have developed a Transfer Learning-based Deep Learning approach to identify human victims under collapsed building environments by integrating machine learning classification algorithms. A custom-made human victim dataset was created with five class labels: head, hand, leg, upper body, and without the body. First, we extracted the class-wise features of the dataset using fine-tuning-based transfer learning on ResNet-50 deep learning model. The learned features of the model were then extracted, and then a feature selection was performed using J48 to study the impact of feature reduction in classification. Several decision tree algorithms, including decision stump, hoeffiding tree, J48, Linear Model tree (LMT), Random Forest, Random tree, Representative (REP) tree, J48 graft, and other famous algorithms like LibSVM, Logistic regression, Multilayer perceptron, BayesNet, Naive Bayes are then used to perform the classification. The classification accuracy of the abovementioned algorithms is compared to recommend the optimal approach for real-time use. The random tree approach outperformed all other tree-based algorithms with a maximum classification accuracy of 99.53% and a computation time of 0.02 seconds.

关键词： Search and rescue disaster victim detection collapsed building deep learning decision tree algorithms machine learning ResNet-50 decision tree classifiers

来源：评论

学校读者我要写书评

暂无评论

Cash Holdings Prediction Using decision tree algorithms and Comparison with Logistic Regression Model

引用

CYBERNETICS AND SYSTEMS 2021年第8期52卷 689-704页

作者： Wu, Hsu-Che Chen, Jen-Hsiang Wang, Pei-Wen Natl Chung Cheng Univ Dept Accounting & Informat Technol Chiayi Taiwan Shih Chien Univ Dept Informat Management Taipei Taiwan

Cash is a very important property of an enterprise, but offers the lowest return of any asset. Nevertheless, firms do not invest all assets into higher-paying assets and hold part of their cash, especially firms in the high-tech electronics industry. Firms in this industry hold considerable amounts of cash because they have high expenses and recovery is uncertain, potentially leaving the firm without sufficient funds. It is therefore necessary to prepare a certain amount of cash. This study uses a decision tree algorithm including J48, LMT, Random Forest, REP tree, Simple CART, Extra tree, and BF tree to measure the performance of predictions. This study has three experiments: (1) testing the predictive ability of the decision tree algorithm, (2) testing the decision tree algorithm with performance improvements, and (3) determining the best decision tree forecast rate comparison using the logistic regression model. The experiments indicate that the random forest has the highest and best prediction rate comparison with the logistic regression model.

关键词： Cash holdings decision tree algorithms logistic regression model

来源：评论

学校读者我要写书评

暂无评论

Estimation of the Double-Cage Model for Three-Phase Induction Machines Using decision tree-Based algorithms

IEEE OPEN JOURNAL OF THE INDUSTRIAL ELECTRONICS SOCIETY

引用

IEEE OPEN JOURNAL OF THE INDUSTRIAL ELECTRONICS SOCIETY 2025年 6卷 915-926页

作者： Oliveira, Eduardo Ferreira Rios Santos, Rafael Simoes, Marcelo Godoy Paredes, Helmo Morales Sao Paulo State Univ UNESP Inst Sci & Technol BR-18087180 Sorocaba SP Brazil Univ Vaasa Sch Technol & Innovat Dept Elect Engn FI-65101 Vaasa Finland

This article presents a novel methodology for estimating the double-cage model (DCM) for three-phase induction machines (TIMs) using decision tree-based algorithms. Validated on a diverse dataset of 860 machines spanning a power range from 0.12 to 370 kW, the proposed method stands out by requiring fewer input parameters than traditional techniques like the modified Newton method. Moreover, the proposed approach remains effective even when the input data exhibits statistical deviations, a common challenge in practical scenarios. The main contributions of this work are the reduction of the number of parameters necessary for the estimation of the DCM equivalent circuit and employing three distinct decision tree-based algorithms, whose effectiveness was confirmed through simulations and experimental tests, thereby providing an accurate representation of the dynamics of real TIMs. The results indicate that by using only basic and readily available data from machine nameplates, such as nominal current, power, speed, voltage, and torque, the proposed methodology provides a reliable and efficient framework for incorporating the real dynamics of TIMs into computational models.

关键词： Integrated circuit modeling Torque Equivalent circuits Newton method Analytical models Accuracy Stator windings Parameter estimation Industrial electronics decision trees decision tree algorithms double-cage model (DCM) induction machines parameter estimation

来源：评论

学校读者我要写书评

暂无评论

WT Model & Applications in Loan Platform Customer Default Prediction Based on decision tree algorithms 14th

WT Model & Applications in Loan Platform Customer Default Pr...

引用

14th International Conference on Intelligent Computing (ICIC)

作者： Pang, Sulin Yuan, Jinmeng Jinan Univ Inst Finance Engn Sch Management Guangzhou 510632 Guangdong Peoples R China Guangdong Emergency Technol Res Ctr Risk Evaluat Guangzhou 510632 Guangdong Peoples R China

ISBN: (纸本)9783319959290;9783319959306

Due to the huge losses caused by the bad credit customers, loan platforms attach great attention to testing and forecasting of bad loans. From perspectives of both loan customer type identification and loan default prediction, we initially constructed a WT early warning model for loan default client prewarning based on C5.0 decision tree, CART decision tree and CHAID decision tree in this paper. WT model is set with weighted calculating algorithms. Considering the data characteristics of loan platform, we designed a posteriori combination algorithm of three sub-models: C5.0, CART, CHAID, and performance test indicators: sensitivity, accuracy, warning rate, false alarm rate. In empirical research, we used the real loan transaction dataset of a bank in Taiwan to construct the WT model of the bank, and found that WT model overcomes the shortcomings of each sub-model respectively and achieves effective prewarning of customer default. The experimental results show that the alarm rate of test data set is 26.93% and the false alarm rate is 18.33% and the accuracy rate is 81.67% when applying the WT model. Loan platforms can acquire both high customer default prediction accuracy and high alarm rate by applying WT prewarning model. Both the research method and experiment results in this paper are meaningful to loan platform operation.

关键词： WT model decision tree algorithms Customer default prewarning Loan platforms

来源：评论

学校读者我要写书评

暂无评论

Potential Assessment of Neural Network and decision tree algorithms for Forecasting Ambient PM2.5 and CO Concentrations: Case Study

引用

JOURNAL OF HAZARDOUS TOXIC AND RADIOACTIVE WASTE 2016年第4期20卷 1-8页

作者： Sekar, Chandrra Gurjar, B. R. Ojha, C. S. P. Goyal, Manish Kumar Indian Inst Technol Dept Civil Engn Roorkee 247667 Uttar Pradesh India Indian Inst Technol Dept Civil Engn Gauhati 781039 India

Air pollution in megacities have caught attention of both researchers and policymakers because of increasing emissions, poor air quality, and potential adverse health impacts on densely inhabited populations. Oxides of nitrogen, particulate matter, carbon monoxide, and hydrocarbons are the major air pollutants of vehicular emissions near major intersections and arterials in megacities. The present study is mainly aimed at predicting PM2.5 and CO concentrations at an income tax office (ITO) intersection in the megacity of Delhi. Artificial neural networks (ANNs) and decision tree algorithms (e.g., REPtree and M5P algorithm techniques) are used to predict hourly fine particulate matter (PM2.5) and carbon monoxide (CO) pollutant concentrations at the ITO intersection. Factors and parameters, such as meteorological conditions, traffic, and vehicular emissions, that affect pollutant concentrations are used in different combinations for the model development. Performance evaluation of ANN, REPtree, and M5P algorithms for hourly PM2.5 and CO concentration prediction is carried out, and the effects of the aforementioned factors are discussed. The M5P algorithm performs better than ANN and REPtree algorithms in that it precisely captures the relationships among the predictor variables and pollutant concentrations. (C) 2015 American Society of Civil Engineers.

关键词： Artificial neural network decision tree algorithms PM2.5 CO Megacity Air quality

来源：评论

学校读者我要写书评

暂无评论

Predicting maritime accident consequence scenarios for emergency response decisions using optimization-based decision tree approach

引用

MARITIME POLICY & MANAGEMENT 2023年第1期50卷 19-41页

作者： Li, Baode Lu, Jing Lu, Han Li, Jing Dalian Maritime Univ Transportat Engn Coll Dalian Peoples R China

Emergency response decision-making for maritime accidents needs to consider the possible consequences and scenarios of an accident to develop an effective emergency response strategy to reduce the severity of the accident. This paper proposes a novel machine learning-based methodology for predicting accident scenarios and analysing its factors to assist emergency response decision-making from an emergency rescue perspective. Specifically, the accident data used are collected from maritime accident investigation reports, and then two types of decision tree (DT) algorithms, classification and regression tree (CART) and random forest (RF), are used to develop scenario prediction models for three accident consequences including ship damage, casualty, and environmental damage. The hyper-parameters of these two DT algorithms are optimized using two state-of-the-art optimization algorithms, namely random search (RS) and Bayesian optimization (BO), respectively, aiming to obtain the prediction model with the highest accuracy. Experimental results reveal that BO-RF algorithm produces the best accuracy as compared to others. In addition, an analysis of feature importance shows that the number of people involved in an accident is the most important driving factor affecting the final accident scenario. Finally, decision rules are generated from the obtained optimal prediction model, which can provide decision support for emergency response decisions.

关键词： Maritime accidents emergency response decision tree algorithms hyper-parameters optimization decision rule generation

来源：评论

学校读者我要写书评

暂无评论

Predicting Hospital No-Shows: Interpretable Machine Learning Models Approach

引用

IEEE ACCESS 2024年 12卷 166058-166067页

作者： Toffaha, Khaled M. Simsekler, Mecit Can Emre Alshehhi, Aamna Omar, Mohammed Atif Khalifa Univ Dept Management Sci & Engn Abu Dhabi U Arab Emirates Khalifa Univ Dept Biomed Engn & Biotechnol Abu Dhabi U Arab Emirates

Healthcare systems face significant challenges and financial burdens due to patient no-shows, highlighting the need for accurate and interpretable predictive models. This study evaluated the efficacy of twelve classification algorithms that can generate self-explanatory results illustrations across four categories: Rule Set classifiers, Rule List classifiers, Rule tree classifiers, and Algebraic Models, using a real-world dataset, "Brazilian Medical Appointment No Shows". Analysis across multiple performance metrics revealed significant differences among the algorithms. Advanced models like tree- Generalized Additive Model (GAM), Fast Interpretable Greedy-tree Sums (FIGS), tree Alternating Optimization (TAO) tree, and RuleFit demonstrated superior predictive capabilities using Over-Sampling and feature selection, achieving an accuracy of 87.53%, AUC 0.87, and F1-score of 0.86, compared to basic tree algorithms like Greedy tree and C4.5. While tree-GAM showed high accuracy, it had a significantly longer runtime of approximately 101 seconds. FIGS and TAO tree offered compelling alternatives with comparable accuracy but significantly reduced computational demands, with runtimes under 1 second. These findings highlight the trade-offs between predictive power, computational efficiency, and practical implementation in healthcare settings. The study also revealed the value of flexible, adaptive architectures in capturing nuanced factors influencing patient no-shows. Overall, these advanced algorithms present accurate and interpretable solutions for forecasting patient no-shows, with FIGS and TAO tree emerging as particularly effective choices that offer a good balance between predictive insight and practical viability. These insights aim to guide health systems in optimizing patient access and reliability while addressing the complex issue of no-shows, underscoring the importance of considering multiple performance metrics when selecting algorithms for real-world applicati

关键词： Patient no-shows decision tree algorithms greedy tree classifiers additive model trees hierarchical shrinkage trees hierarchical shrinkage trees healthcare analytics healthcare analytics machine learning machine learning hierarchical shrinkage trees healthcare analytics machine learning

来源：评论

学校读者我要写书评

暂无评论

Integrating between Taguchi methodology and boosted decision trees machine learning: a case study in enhancing quality electrical conductor manufacturing

引用

JORDAN JOURNAL OF MECHANICAL AND INDUSTRIAL ENGINEERING 2024年第4期18卷 847-859页

作者： Mansour, Hamdia Abohashima, Hanaa Elkhouly, Heba, I Fayoum Univ Fac Engn Mech Engn Dept Faiyum 65311 Egypt Beni Suef Univ Fac Engn Mech Engn Dept Bani Suwayf 62511 Egypt

This study introduces an integrated approach that combines Taguchi methodology and machine learning techniques to enhance production quality in electrical cable manufacturing. The Taguchi method was employed to identify critical factors such as compaction percentage, wire diameter, raw materials, assembly procedures, and operating voltage, converging on an regression, and k-star, were utilized alongside evaluation metrics like sensitivity, F1-score, and accuracy. The integration of Taguchi and machine learning facilitated the identification of key process parameters and their optimal settings, significantly improving the quality and efficiency of cable manufacturing. The optimal solution achieved included a 666 kg/km weight, 2.64 cm diameter, and a 30% compaction rate, reducing the poor quality cost from 5% to 1.7%. This synergistic approach allowed for the optimization of critical process factors, resulting in significant improvements in product quality and reductions in defects and costs.

关键词： Keywords : machine learning decision tree algorithms electric cables doe ANOVA Taguchimethod.

来源：评论

学校读者我要写书评

暂无评论

decision tree-based identification of important molecular fragments for protein-ligand binding

引用

CHEMICAL BIOLOGY & DRUG DESIGN 2024年第1期103卷 e14427-e14427页

作者： Li, Baiyi Wang, Yunsong Yin, Zuode Xu, Lei Xie, Liangxu Xu, Xiaojun Jiangsu Univ Technol Inst Bioinformat & Med Engn Changzhou 213001 Peoples R China Jiangsu Univ Technol Sch Elect & Informat Engn Changzhou Peoples R China

Fragment-based drug design is an emerging technology in pharmaceutical research and development. One of the key aspects of this technology is the identification and quantitative characterization of molecular fragments. This study presents a strategy for identifying important molecular fragments based on molecular fingerprints and decision tree algorithms and verifies its feasibility in predicting protein-ligand binding affinity. Specifically, the three-dimensional (3D) structures of protein-ligand complexes are encoded using extended-connectivity fingerprints (ECFP), and three decision tree models, namely Random Forest, XGBoost, and LightGBM, are used to quantitatively characterize the feature importance, thereby extracting important molecular fragments with high reliability. Few-shot learning reveals that the extracted molecular fragments contribute significantly and consistently to the binding affinity even with a small sample size. Despite the absence of location and distance information for molecular fragments in ECFP, 3D visualization, in combination with the reverse ECFP process, shows that the majority of the extracted fragments are located at the binding interface of the protein and the ligand. This alignment with the distance constraints critical for binding affinity further supports the reliability of the strategy for identifying important molecular fragments. Identifying important molecular fragments through ECFP fingerprints for fragment encoding and decision tree algorithms for feature importance ***

关键词： binding affinity decision tree algorithms important molecular fragments molecular fingerprints protein-ligand complex

来源：评论

学校读者我要写书评

暂无评论

Enhancing intrusion detection recursive feature elimination with resampling in WSN

引用

INTERNATIONAL JOURNAL OF SYSTEM ASSURANCE ENGINEERING AND MANAGEMENT 2023年第6期14卷 2642-2660页

作者： Abhale, Ashwini B. Avulapalli, Jayaram Reddy Vellore Inst Technol Sch Comp Sci Engn & Informat Syst SCORE Vellore India

With the proliferation of technologies such as the Internet of Things, Cloud computing, and Social Networking, large quantities of network traffic and data are generated, necessitating the use of effective Intrusion Detection Systems (IDS). IDS play an essential role in detecting hazards within a system and generating alerts for potential attacks. This paper examines the use of machine learning algorithms for intrusion detection with the NSL KDD dataset. However, not all features contribute equally to performance enhancement in large datasets. Therefore, it becomes essential to reduce the feature set to a subset that improves both speed and accuracy. Within the IDS framework, we employ machine learning algorithms such as Random Forest, Support Vector Machine, K Nearest Neighbour, and decision tree through rigorous experimentation. Recursive Feature Elimination and Resampling techniques are utilised to select and evaluate the impact of feature reduction. The comparative evaluation of the model's performance is demonstrated, emphasising the performance enhancements attained through feature selection.

关键词： Wireless sensor network Intrusion detection system (IDS) NSL KDD dataset Machine learning Random forest Support vector machine K nearest neighbor decision tree algorithms

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

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

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：