检索结果-内蒙古大学图书馆

USE a variety of meteorological and geographic data to build a LightGBM temperature predicted model and quantify the impact of feature on temperature changes based on SHAP

USE a variety of meteorological and geographic data to build...

引用

2024 international conference on Remote Sensing, Mapping, and Geographic Information Systems, RSMG 2024

作者： Xu, Lei Qin, Fen Du, Jinjin Ren, Jiwei College of Geography and Environmental Science Henan University Kaifeng475004 China Henan Industrial Technology Academy of Spatial-Temporal Big Data Zhengzhou450046 China

ISBN: (纸本)9781510685826

Temperature change is a complex atmospheric phenomenon. Even the ERA5-Land atmospheric reanalysis temperature dataset with the highest accuracy currently has errors with the actual observed temperature. This study captures the nonlinear relationship between ERA5-Land temperature and various geographic and climate data such as population density, land use type, DEM, air pressure, and relative humidity by constructing a LightGBM model. The experimental results show that compared with ERA5-Land, the constructed model has a Mean Absolute Error is reduced by 5% and Root Mean Squared Error is reduced by 13%, Mean Squared Error is reduced by24%, Pearson correlation coefficient increased by 2%, R-squared is increased by 3%. A machine learning model with strong versatility and accurate prediction has been obtained, and the correction of daily temperature data in the Yellow River Basin from 2015 to 2019 has been realized. After the model training is completed, SHAP is used to analyze the impact of various feature data on the prediction of temperature by the machine learning model, sort them according to the degree of impact, and analyze the interaction between various feature values to study how different features affect the model prediction of temperature. The results show that in the process of model prediction, except for the ERA5-Land temperature data used for correction, the greatest impact on the model prediction is air pressure and DEM. © 2024 SPIE.

关键词： Atmospheric pressure

来源：评论

学校读者我要写书评

暂无评论

Research on Unknown Distributed Denial-of-Service Attacks Intrusion Detection Based on Deep learning 9

Research on Unknown Distributed Denial-of-Service Attacks In...

引用

9th international conference on Intelligent Computing and Signal Processing, ICSP 2024

作者： Zhang, Liumei An, Zhaoyang Hu, Yuanjiao Xi'an Shiyou University School of Computer Science Xi'an China

ISBN: (纸本)9798350376548

The increasing prevalence of Distributed Denial-of- Service (DDoS) attacks poses a serious threat to network security, emphasizing the crucial need for effective intrusion detection methods. The continuous evolution of attack strategies has exposed the limitations of traditional feature engineering and statistical analysis in detection mechanisms. To address this challenge, this study proposes an innovative DDoS attack detection framework based on deep learning. At the core of this framework lies a Temporal Convolutional Network (TCN) structure, cleverly integrated with an attention mechanism to enhance the model's ability to identify and extract crucial information. This integration not only aids in capturing long- term dependencies and global patterns in time-series data but also significantly improves detection accuracy and system robustness. To validate the effectiveness of the proposed method, comprehensive experimental evaluations were conducted using the CICDDoS2019 dataset. The results demonstrate a notable achievement, with our model achieving a high classification accuracy of 98.14%. In comparison to traditional machine learning methods, the model exhibits significantly reduced false positive and false negative rates. © 2024 IEEE.

关键词： Network intrusion

来源：评论

学校读者我要写书评

暂无评论

Cost-Effective Peak Shaving Strategy Based on Clustering and XGBoost Algorithm 5

Cost-Effective Peak Shaving Strategy Based on Clustering and...

引用

5th international conference on Artificial Intelligence in Information and Communication (ICAIIC)

作者： Lim, Sol Gantassi, Rahma Choi, Yonghoon Chonnam Natl Univ Dept Elect Engn Gwangju 61186 South Korea

ISBN: (纸本)9781665456456

In a cost-effective peak shaving strategy, clustering and machine learning algorithm can be used to set optimal peak shaving time zone for each load. Energy Storage System (ESS) charge amount is determined with load prediction data through machine learning model, and the peak shaving time zone is adjusted flexibly according to load patterns for each cluster. It is possible to prevent ESS from being overcharged or undercharged through load prediction. In addition, rather than applying peak shaving collectively at the on-peak time, efficient operation of the power grid can be expected by adjusting the time zone flexibly for each power usage pattern. The effectiveness of the proposed system model is to be proved through changes in electricity cost depending on whether it is introduced or not.

关键词： clustering cost minimization K-means machine learning peak shaving XGBoost

来源：评论

学校读者我要写书评

暂无评论

CRFL: A novel federated learning scheme of client reputation assessment via local model inversion

引用

international JOURNAL OF INTELLIGENT SYSTEMS 2022年第8期37卷 5457-5471页

作者： Zheng, Jiamin Huang, Teng Huang, Jiahui Beijing Technol & Business Univ Sch E Business & Logist Beijing Peoples R China Guangzhou Univ Inst Artificial Intelligence & Blockchain Guangzhou 510006 Peoples R China

Federated learning (FL) is gradually becoming a key learning paradigm in Privacy-preserving machine learning (ML) systems. In FL, a large number of clients cooperate with a central server to learn a shared model without sharing their own data sets. However, since there is a great disparity between the client data sets, standard FL is often hard to tune and suffers from performance degradation due to the inharmony among local models. To this end, in this paper we propose a novel FL scheme, termed client reputation federated learning (CRFL), which dynamically assesses the reputation of the clients participating in FL. Our method leverages techniques from model explanation, and aims at precisely measure each client's impact to the global model. To be specific, we first calculate the saliency-weighted variance on pixelwise relevance scores as the quality factor of a single sample. Then we extract activation function values at the last hidden layer to compute the divergence factor of individual data set. Finally, the server integrates these two factors as an assessment of the client reputation. By leveraging such assessment, CRFL can dynamically adjust the weights of the clients in each aggregation round, thus leading to a significant improvement over the baseline method in terms of model accuracy and convergence rate. Intensive experiments are conducted on the MNIST and CIFAR-10 data sets, and experimental results demonstrate the efficacy of the proposed method.

关键词： federated learning machine learning model explanation model inversion reputation assessment

来源：评论

学校读者我要写书评

暂无评论

Dark matter search towards the Sun using machine learning reconstructions of single-line events in ANTARES 38

Dark matter search towards the Sun using Machine Learning re...

引用

38th international Cosmic Ray conference, ICRC 2023

作者： Méndez, Juan García Ardid, Salva Ardid, Miguel Universitat Politècnica de València Research Institute for Integrated Management of Coastal Areas Carrer Paranimf 1 Gandia Spain Universitat Politècnica de València Manufacture and Management of Industrial Projects Spain Universitat Politècnica de València Department of Applied Physics Spain

The ANTARES neutrino telescope stopped gathering data in February 2022, after nearly 16 years of operation. The detector consisted of 12 vertical lines forming a 3D array of photo-sensors, which instrumented about 10 megatons of Mediterranean seawater. We present a method using deep learning that improves the direction reconstruction of low-energy single-line events, for which the reconstruction of the azimuth angle of the incoming neutrino is particularly difficult. We also present a combination of machine learning techniques to reconstruct the energy of the same kind of events. Our results enhance the resolution of former reconstruction techniques, at least doubling our sensitivity in the range of energy of tens of GeV, which is highly relevant for dark matter searches and other physics studies. Here, we propose a binned Dark Matter (DM) search towards the Sun for ANTARES single-line events using the new reconstruction methods. We compute the neutrino flux sensitivity for different DM annihilation channels and particle candidate masses. In this first trial, the methodology is applied to a subset of ANTARES data: these results anticipate better sensitivities for low masses of DM candidates (below ∼ 150 GeV) and/or soft spectrum channels compared to those obtained based on standard reconstruction techniques once the method is applied to the whole ANTARES dataset. © Copyright owned by the author(s) under the terms of the Creative Commons.

关键词： Dark Matter

来源：评论

学校读者我要写书评

暂无评论

Interpreting Black-box machine learning Models for High Dimensional datasets 10

Interpreting Black-box Machine Learning Models for High Dime...

引用

10th IEEE international conference on data science and Advanced Analytics, DSAA 2023

作者： Karim, Md. Rezaul Shajalal, Md Gras, Alexander Dohmen, Till Chala, Sisay Adugna Boden, Alexander Beecks, Christian Decker, Stefan Rwth Aachen University Computer Science 5 - Information Systems and Databases Germany Fraunhofer - Institute for Applied Information Technology Fit Germany University of Siegen Germany Bonn-Rhein-Sieg University of Applied Sciences Germany University of Hagen Germany

ISBN: (纸本)9798350345032

Many datasets are of increasingly high dimension- ality, where a large number of features could be irrelevant to the learning task. The inclusion of such features would not only introduce unwanted noise but also increase computational complexity. Deep neural networks (DNNs) outperform machine learning (ML) algorithms in a variety of applications due to their effectiveness in modelling complex problems and handling high-dimensional datasets. However, due to non-linearity and higher-order feature interactions, DNN models are unavoidably opaque, making them black-box methods. In contrast, an interpretable model can identify statistically significant features and explain the way they affect the model's outcome. In this paper, we propose a novel method to improve the interpretability of blackbox models in the case of high-dimensional datasets. First, a black-box model is trained on full feature space that learns useful embeddings on which the classification is performed. To decompose the inner principles of the black-box and to identify top-k important features (global explainability), probing and perturbing techniques are applied. An interpretable surrogate model is then trained on top-k feature space to approximate the black-box. Finally, decision rules and counterfactuals are derived from the surrogate to provide local decisions. Our approach outperforms tabular learners, e.g., TabNet and XGboost, and SHAP-based interpretability techniques, when tested on a number of datasets having dimensionality between 54 and 20,53111GitHub: https://***/rezacsedu/DeepExplainHidim. © 2023 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Enhancing Electrical Fault Detection in Transmission Lines Through machine learning 3

Enhancing Electrical Fault Detection in Transmission Lines T...

引用

3rd IEEE international conference on Electrical Power and Energy Systems, ICEPES 2024

作者： Kulshrestha, Deepansh Chaudhary, Vishal Madhav Institute of Technology & Science Department of Electrcial Engineering Gwalior India

ISBN: (纸本)9798350390728

The purpose of this study is to apply machine learning techniques to improve the detection of electrical irregularities in transmission lines. The major objective is to provide a reliable framework that reliably identifies and categorises different electrical anomalies, improving power distribution networks' reliability and efficiency. The models were trained and assessed using a range of machine learning methods, such as Support Vector machines (SVM), Random Forest Classifiers, and Decision Tree Classifiers, XGB Classifier, and Logistic Regression. The dataset replicated real-world electrical transmission line issues. The major conclusions demonstrate that every model-aside from SVM maintains a high level of accuracy when evaluated on fresh data and that every model has a training accuracy that is always high, frequently surpassing 100%. Evaluation criteria like recall, F1-score, and accuracy show how well the ensemble techniques Decision Tree, Random Forest, and XGB Classifier identify electrical defects. According to the study, ensemble techniques can enhance fault identification in power distribution networks when used with logistic regression. The efficacy and interpretability of logistic regression are noteworthy. This research provides practical solutions to enhance the resilience and dependability of electrical transmission networks while also advancing our understanding of machine learning applications for problem identification. © 2024 IEEE.

关键词： Power distribution faults

来源：评论

学校读者我要写书评

暂无评论

Structural damage detection based on fundamental Bayesian two-stage model considering the modal parameters uncertainty

引用

STRUCTURAL HEALTH MONITORING-AN international JOURNAL 2023年第4期22卷 2305-2324页

作者： Zhang, Feng-Liang Gu, Dong-Kai Li, Xiao Ye, Xiao-Wei Peng, Huayi Harbin Inst Technol Sch Civil & Environm Engn Shenzhen 518055 Peoples R China Zhejiang Univ Dept Civil Engn Hangzhou Peoples R China

In structural health monitoring (SHM), damage detection is a final target to know the real status of the objective structure. Vibration-based damage detection is a commonly used method, since it makes full use of the dynamic characteristics. Improving the efficiency of this kind of methods has attracted increasing attentions. The existing uncertainty of identified modal parameters using measured data may significantly affect the detection accuracy. Furthermore, an optimization algorithm with a better convergence speed can improve the detection accuracy and reduce the computational time. This article presents the work to develop a novel damage detection method based on fundamental Bayesian two-stage model and sparse regularization. In this method, the most probable value of modal parameters and the associated posterior uncertainty are combined to investigate the effect of uncertainty on damage detection. The usage of the sparse regularization in the objective function can decrease the complexity of modeling and avoid the overfitting problem. A machine learning method combining intelligent swarm optimization algorithm with K-means clustering was used to carry out the optimization. Finally, a method combining three existing theory, that is, fundamental Bayesian two-stage model, sparse regularization, and I-Jaya algorithm, was developed. To investigate the efficiency of the proposed method, the traditional objective functions with and without the sparse regularization were also used for the comparison. The proposed method was verified by an ASCE benchmark example, and then it is applied into an experimental structure. The results show that due to the consideration of uncertainty, the objective function based on the fundamental Bayesian model and sparse regularization has a better performance.

关键词： Vibration-based damage detection ambient vibration modal parameters posterior uncertainty machine learning sparse regularization

来源：评论

学校读者我要写书评

暂无评论

FedSig: A Federated Graph Augmentation for Class-Imbalanced Node Classification 29th

FedSig: A Federated Graph Augmentation for Class-Imbalanced ...

引用

29th international conference on database Systems for Advanced Applications (DASFAA)

作者： Bi, Bei Zhang, Zhiwei Qiao, Pengpeng Yuan, Ye Wang, Guoren Beijing Inst Technol Beijing Peoples R China

ISBN: (纸本)9789819755516;9789819755523

Federated learning (FL) is a learning paradigm that collaboratively trains machine learning models among distributed clients without leaking data privacy. A key challenge in federated learning is to handle the class imbalance issue across clients. Although significant efforts have been dedicated to addressing this challenge, the effect of the imbalanced classification on graphs is still not satisfactory. In this paper, we propose FedSig, a novel Federated graph data augmentation that employs Synthetic minority oversampling techniques for Imbalanced Graphs. Specifically, we adopt a shared encoder and private decoder architecture to support data heterogeneity. Building upon FedSig, we propose a two-stage synthetic minority node mechanism that can effectively capture both local and global features for generating minority nodes. In addition, we propose an edge generator to simulate the relationship between synthetic minority nodes and real nodes to balance the original imbalanced graph. To mitigate drift between local and global models, we propose a novel regularization term that enforces proximity between the distribution of local minority node embeddings and that of global minority node embeddings. Extensive experiments on Pubmed, Github, and Amazon-cs demonstrate that FedSig outperforms state-of-the-art approaches for imbalanced node classification.

关键词： Federated learning Imbalanced classification GNNs

来源：评论

学校读者我要写书评

暂无评论

The Effect of Balancing Methods on Model Behavior in Imbalanced Classification Problems 15

The Effect of Balancing Methods on Model Behavior in Imbalan...

引用

5th international Workshop on learning with Imbalanced Domains - Theory and Applications / European conference on machine learning and Principles and Practice of Knowledge Discovery in databases (ECML PKDD)

作者： Stando, Adrian Cavus, Mustafa Biecek, Przemyslaw Warsaw Univ Technol Fac Math & Informat Sci Warsaw Poland Eskisehir Tech Univ Dept Stat Eskisehir Turkiye

Imbalanced data poses a significant challenge in classification as model performance is affected by insufficient learning from minority classes. Balancing methods are often used to address this problem. However, such techniques can lead to problems such as overfitting or loss of information. This study addresses a more challenging aspect of balancing methods - their impact on model behavior. To capture these changes, Explainable Artificial Intelligence tools are used to compare models trained on datasets before and after balancing. In addition to the Variable Importance method, this study uses Partial Dependence and Accumulated Local Effects profiles. Real and simulated datasets are tested, and an open-source Python package edgaro is developed to facilitate this analysis. The results obtained show significant changes in model behavior due to balancing methods, which can lead to biased models toward a balanced distribution. These findings confirm that balancing analysis should go beyond model performance comparisons to achieve higher reliability of machine learning models. Therefore, we propose a new method performance gain plot for informed data balancing strategy to make an optimal selection of balancing method by analyzing the measure of change in model behavior versus performance gain.

关键词： Imbalanced learning Explainable artificial intelligence data balancing Model behavior change Performance gain plot

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：