检索结果-内蒙古大学图书馆

Comprehensive analysis of various imputation and forecasting models for predicting PM2.5 pollutant in Delhi

Neural Computing and Applications 2025年第17期37卷 11441-11458页

作者： Karnati, Hemanth Soma, Anuraag Alam, Adnan Kalaavathi, B. School of Computer Science and Engineering Vellore Institute of Technology Tamil Nadu Vellore632014 India Department of IoT School of Computer Science and Engineering Vellore Institute of Technology Tamil Nadu Vellore632014 India

Understanding and predicting air quality is pivotal for public health and environmental management, especially in urban areas like Delhi. This study utilizes a comprehensive dataset from the Central Pollution Control Board, detailing air pollutant concentrations at the ITO (Income Tax Office), Delhi station, from 2017 to 2023. Our focus on pollutants such as PM2.5, PM10, NO2, NH3, SO2, ozone, and CO highlights the challenges posed by missing data in environmental studies. Through the application of several imputation models, including KNN, linear regression, forward fill + backward fill, Fourier + KNN, linear interpolation, and statistical methods, we identified combination of forward fill and backward fill as the most effective method for addressing data gaps, specifically for ozone measurements. Building on this imputed dataset, we employ various forecasting models to predict PM2.5 levels, a critical pollutant. Our exploration includes time series and deep learning approaches, such as LSTM, Bi-LSTM, CNN-LSTM, GRU, Deep AR, WaveNet, TCN, ARIMA, SARIMA, and Prophet. The performance of these models is evaluated using daily data, with Bi-LSTM and LSTM with attention emerging as the top performers due to their accuracy in predicting PM2.5 concentrations. The implications of our findings extend beyond academic interest, offering practical insights for environmental policy and health advisories in Delhi. By demonstrating the efficacy of specific imputation and forecasting techniques, our research contributes to the broader effort of improving air quality monitoring and prediction. The success of Bi-LSTM and LSTM with attention models, in particular, suggests a promising avenue for future studies aiming to enhance the precision of environmental forecasts. © The Author(s) 2025.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

APWF: A Parallel Website Fingerprinting Attack with Attention Mechanism

引用

computers, Materials & Continua 2025年第2期82卷 2027-2041页

作者： Dawei Xu Min Wang Yue Lv Moxuan Fu Yi Wu Jian Zhao College of Computer Science and Technology Changchun UniversityChangchun130022China School of Cyberspace Science and Technology Beijing Institute of TechnologyBeijing100081China Key Laboratory of Intelligent Rehabilitation and Barrier-Free for the Disabled Changchun UniversityChangchun130022China China Academy of Information and Communications Technology Beijing100083China Key Laboratory of Mobile Application Innovation and Governance Technology Ministry of Industry and Information TechnologyBeijing100846China

Website fingerprinting (WF) attacks can reveal information about the websites users browse by de-anonymizing encrypted traffic. Traditional website fingerprinting attack models, focusing solely on a single spatial feature, are inefficient regarding training time. When confronted with the concept drift problem, they suffer from a sharp drop in attack accuracy within a short period due to their reliance on extensive, outdated training data. To address the above problems, this paper proposes a parallel website fingerprinting attack (APWF) that incorporates an attention mechanism, which consists of an attack model and a fine-tuning method. Among them, the APWF model innovatively adopts a parallel structure, fusing temporal features related to both the front and back of the fingerprint sequence, along with spatial features captured through channel attention enhancement, to enhance the accuracy of the attack. Meanwhile, the APWF method introduces isomorphic migration learning and adjusts the model by freezing the optimal model weights and fine-tuning the parameters so that only a small number of the target, samples are needed to adapt to web page changes. A series of experiments show that the attack model can achieve 83% accuracy with the help of only 10 samples per category, which is a 30% improvement over the traditional attack model. Compared to comparative modeling, APWF improves accuracy while reducing time costs. After further fine-tuning the freezing model, the method in this paper can maintain the accuracy at 92.4% in the scenario of 56 days between the training data and the target data, which is only 4% less loss compared to the instant attack, significantly improving the robustness and accuracy of the model in coping with conceptual drift.

关键词： Website fingerprinting attack transfer learning concept drift

来源：评论

学校读者我要写书评

暂无评论

Experiment and modeling of coke formation and catalyst deactivation in n-heptane catalytic cracking over HZSM-5 zeolites

引用

Chinese Journal of Chemical engineering 2023年第3期55卷 165-172页

作者： Zhenzhou Ma Xu Hou Bochong Chen Liu Zhao Enxian Yuan Tingting Cui School of Chemical Engineering Changchun University of TechnologyChangchun 130012China Advanced Institute of Materials Science Changchun University of TechnologyChangchun 130012China School of Chemistry and Chemical Engineering Yangzhou UniversityYangzhou 225009China Department of Chemistry Tsinghua UniversityBeijing 100084China

Since paraffins catalytic cracking was of significant importance to light olefins and aromatics production,this work was intended to gain insights into the feature and model of coke formation and catalyst deactivation in n-heptane catalytic cracking over HZSM-5 zeolites. 18 tests of n-heptane catalytic cracking were designed and carried out over HZSM-5 zeolites in a wide range of operating conditions. A particular attention was paid to the measurement of the conversion, product distribution, coke content, and the porosity and acidity of the fresh and spent HZSM-5 zeolites. It was found that alkene and aromatic promoted coke formation, and it reduced the pore volume and acid site of HZSM-5 zeolites, tailoring its performance in n-heptane catalytic cracking. The specific relationship between HZSM-5 zeolites, n-heptane conversion, product distribution and coke formation was quantitively characterized by the exponential and linear function. Based on the reaction network, the coupled scheme of coke formation and catalyst deactivation were specified for n-heptane catalytic cracking. The dual-model was proposed for the process simulation of n-heptane catalytic cracking over HZSM-5 zeolites. It predicted not only the conversion and product distribution but also coke content with the acceptable errors.

关键词： n-Heptane HZSM-5 Catalytic cracking Coke Deactivation Dual-model

来源：评论

学校读者我要写书评

暂无评论

Effect of Nitrogen and Sulfur Co‑Doped Graphene on the Electrochemical Hydrogen Storage Performance of Co_(0.9)Cu_(0.1)Si Alloy

引用

Acta Metallurgica Sinica(English Letters) 2023年第6期36卷 1023-1037页

作者： Wenhao Fan Jianxun Zhao Dayong Liu Qingcheng Liang Wanqiang Liu Qingshuang Wang Heng Liu Peng Chen Shang Gao Xinlong Bao Yong Cheng Xinwei Wang Xin Guo School of Materials Science and Engineering Changchun University of Science and TechnologyChangchun 130022China Engineering Research Center of Optoelectronic Functional Materials Ministry of EducationChangchun 130022China College of Electronic Science and Engineering Jilin UniversityChangchun 130022China School of Life Science and Technology Changchun University of Science and TechnologyChangchun 130022China Research Center for Nanotechnology Changchun University of Science and TechnologyChangchun 130022China Research Institute of Changchun University of Science and Technology in Chongqing Chongqing 401120China State Key Laboratory of Rare Earth Resource Utilization Changchun Institute of Applied ChemistryCASChangchun 130022China

Co_(0.9)Cu_(0.1)Si alloy was prepared by mechanical alloying ***-doped graphene(NG)and nitrogen–sulfur codoped graphene(NSG)were prepared by hydrothermal method.5 wt%graphene oxide,NG and NSG were doped into Co_(0.9)Cu_(0.1)Si alloy,respectively,by ball milling to improve the electrochemical hydrogen storage performance of the composite material.X-ray diffraction and scanning electron microscopy were used to characterize the structure and morphology of the composite material,and the LAND battery test system and three-electrode battery system were used to test the electrochemical performance of the composite *** composite material showed better discharge capacity and better cycle stability than the pristine *** addition,in order to study the optimal ratio of NSG,3%,5%,7%and 10%of NSG were doped into Co_(0.9)Cu_(0.1)Si alloy,***_(0.9)Cu_(0.1)Si alloy doped with 5%NSG had the best performance among all the *** best discharge capacity was 580.1 mAh/g,and its highest capacity retention rate was 64.1%.The improvement in electrochemical hydrogen storage performance can be attributed to two *** the one hand,the electrocatalytic performance of graphene is improved by co-doping nitrogen and sulfur,on the other hand,graphene has excellent electrical conductivity.

关键词： Co_(0.9)Cu_(0.1)Si alloy Nitrogen-sulfur co-doped graphene(NSG) Composite material Electrochemical hydrogen storage

来源：评论

学校读者我要写书评

暂无评论

Border Sensitive Knowledge Distillation for Rice Panicle Detection in UAV Images

引用

computers, Materials & Continua 2024年第10期81卷 827-842页

作者： Anitha Ramachandran Sendhil Kumar K.S. School of Computer Science and Engineering VIT(Vellore Institute of Technology)VelloreVellore632014India

Research on panicle detection is one of the most important aspects of paddy phenotypic analysis.A phenotyping method that uses unmanned aerial vehicles can be an excellent alternative to field-based ***,it entails many other challenges,including different illuminations,panicle sizes,shape distortions,partial occlusions,and complex *** detection algorithms are directly affected by these *** work proposes a model for detecting panicles called Border Sensitive Knowledge Distillation(BSKD).It is designed to prioritize the preservation of knowledge in border areas through the use of feature *** feature-based knowledge distillation method allows us to compress the model without sacrificing its *** imitation mask is used to distinguish panicle-related foreground features from irrelevant background features.A significant improvement in Unmanned Aerial Vehicle(UAV)images is achieved when students imitate the teacher’s *** the UAV rice imagery dataset,the proposed BSKD model shows superior performance with 76.3%mAP,88.3%precision,90.1%recall and 92.6%F1 score.

关键词： Rice panicle detection UAV border sensitive method deep learning knowledge distillation

来源：评论

学校读者我要写书评

暂无评论

Nonparametric Statistical Feature Scaling Based Quadratic Regressive Convolution Deep Neural Network for Software Fault Prediction

引用

computers, Materials & Continua 2024年第3期78卷 3469-3487页

作者： Sureka Sivavelu Venkatesh Palanisamy School of Computer Science Engineering and Information Systems Vellore Institute of TechnologyVellore632014India

The development of defect prediction plays a significant role in improving software quality. Such predictions are used to identify defective modules before the testing and to minimize the time and cost. The software with defects negatively impacts operational costs and finally affects customer satisfaction. Numerous approaches exist to predict software defects. However, the timely and accurate software bugs are the major challenging issues. To improve the timely and accurate software defect prediction, a novel technique called Nonparametric Statistical feature scaled QuAdratic regressive convolution Deep nEural Network (SQADEN) is introduced. The proposed SQADEN technique mainly includes two major processes namely metric or feature selection and classification. First, the SQADEN uses the nonparametric statistical Torgerson–Gower scaling technique for identifying the relevant software metrics by measuring the similarity using the dice coefficient. The feature selection process is used to minimize the time complexity of software fault prediction. With the selected metrics, software fault perdition with the help of the Quadratic Censored regressive convolution deep neural network-based classification. The deep learning classifier analyzes the training and testing samples using the contingency correlation coefficient. The softstep activation function is used to provide the final fault prediction results. To minimize the error, the Nelder–Mead method is applied to solve non-linear least-squares problems. Finally, accurate classification results with a minimum error are obtained at the output layer. Experimental evaluation is carried out with different quantitative metrics such as accuracy, precision, recall, F-measure, and time complexity. The analyzed results demonstrate the superior performance of our proposed SQADEN technique with maximum accuracy, sensitivity and specificity by 3%, 3%, 2% and 3% and minimum time and space by 13% and 15% when compared with the two sta

关键词： Software defect prediction feature selection nonparametric statistical Torgerson-Gower scaling technique quadratic censored regressive convolution deep neural network softstep activation function nelder-mead method

来源：评论

学校读者我要写书评

暂无评论

Distributed Voltage Regulation of Distribution Network Based on Partition Control 2

Distributed Voltage Regulation of Distribution Network Based...

引用

2nd International Conference on Cyber-Energy Systems and Intelligent Energy, ICCSIE 2024

作者： Jine, Wang Hui, Chen Liying, Zhao Guangxin, Zhang School of Electrical Information Changchun University of Architecture and Civil Engineering Changchun City China School of Computer Technology and Engineering Changchun Institute of Technology Changchun City China School of Information and Engineering College of Science & Technology Ningbo University Ningbo City China

ISBN: (纸本)9798350373318

As China's social economy progresses, the amount of equipment in distribution networks continues to grow, leading to increasingly complex system operations. In this environment, traditional centralized control methods consume excessive computing resources and fail to achieve the processing speeds required by modern intelligent distribution networks. To address these challenges, this article introduces the concept of electrical distance to segment the distribution network and proposes voltage regulation algorithms based on the alternating direction multiplier method and dual ascent method for efficient distributed voltage regulation. © 2024 IEEE.

关键词： DC distribution systems

来源：评论

学校读者我要写书评

暂无评论

Semantic segmentation of ultra-high resolution remote sensing images based on fully convolutional neural networks 4

Semantic segmentation of ultra-high resolution remote sensin...

引用

4th International Conference on computer Vision, Image and Deep Learning, CVIDL 2023

作者： Zhang, Hua Jiang, Zhengang Xu, Jun Yao, Xuekun School of Computer Science and Technology Changchun University of Science and Technology Changchun China Changchun Institute of Technology Jilin Key Laboratory of Changbai History and Culture and VR Technology Reconstruction Changchun China School of Computer Technology and Engineering Changchun Institute of Technology Changchun China

ISBN: (纸本)9798350326444

Semantic segmentation has been widely used in various fields of remote sensing;however, ultra-high resolution remote sensing images, due to their extremely high resolution, exhibit more spatial details and cannot be directly segmented using manual segmentation or traditional semantic segmentation methods. We applied fully convolutional deep learning-based neural networks to the semantic segmentation of ultra-high-resolution remote sensing images and at the same time designed an efficient image segmentation strategy to reduce the difficulty of training while ensuring the integrity of the dataset. We conducted experiments on the ultra-high resolution remote sensing dataset Potsdam and demonstrated that the semantic segmentation method based on fully convolutional neural networks can achieve 93.78% MIoU and 97.39% PA on the training set. © 2023 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Enhancing Multicriteria-Based Recommendations by Alleviating Scalability and Sparsity Issues Using Collaborative Denoising Autoencoder

引用

computers, Materials & Continua 2024年第2期78卷 2269-2286页

作者： S.Abinaya K.Uttej Kumar School of Computer Science&Engineering Vellore Institute of TechnologyChennaiTamilnadu600127India

A Recommender System(RS)is a crucial part of several firms,particularly those involved in *** conventional RS,a user may only offer a single rating for an item-that is insufficient to perceive consumer ***,businesses in industries like e-learning and tourism enable customers to rate a product using a variety of factors to comprehend customers’*** the other hand,the collaborative filtering(CF)algorithm utilizing AutoEncoder(AE)is seen to be effective in identifying user-interested ***,the cost of these computations increases nonlinearly as the number of items and users *** triumph over the issues,a novel expanded stacked autoencoder(ESAE)with Kernel Fuzzy C-Means Clustering(KFCM)technique is proposed with two *** the first phase of offline,the sparse multicriteria rating matrix is smoothened to a complete matrix by predicting the users’intact rating by the ESAE approach and users are clustered using the KFCM *** the next phase of online,the top-N recommendation prediction is made by the ESAE approach involving only the most similar user from multiple *** the ESAE_KFCM model upgrades the prediction accuracy of 98.2%in Top-N recommendation with a minimized recommendation generation *** experimental check on the Yahoo!Movies(YM)movie dataset and TripAdvisor(TA)travel dataset confirmed that the ESAE_KFCM model constantly outperforms conventional RS algorithms on a variety of assessment measures.

关键词： Recommender systems multicriteria rating collaborative filtering sparsity issue scalability issue stacked-autoencoder Kernel Fuzzy C-Means Clustering

来源：评论

学校读者我要写书评

暂无评论

Revolutionizing signature scheme: the enhanced Edward Elgamal extreme performance accumulate signature approach for IoT and blockchain applications

引用

Soft Computing 2025年第3期29卷 1473-1496页

作者： Anusha, R. Saravanan, R. School of Computer Science and Engineering Vellore Institute of Technology Tamilnadu Vellore632014 India

Digital signatures, essential for establishing trust in the digital realm, have evolved in their application and importance alongside emerging technologies such as the Internet of Things (IoT), Blockchain, and cryptocurrency. These advancements necessitate improvements in performance, security, and efficiency. This article examines and compares the Elliptic Curve Digital Signature Algorithm with the Hyper Elliptic Curve Digital Signature Algorithm and the Edwards Curve Digital Signature Algorithm. We highlight its superior capabilities for blockchain and IoT applications and advocate for its potential to deliver immediate enhancements in security and performance. Our study introduces a novel digital signature scheme specifically designed to enhance non-repudiation in blockchain ecosystems. Utilizing the Optimized Extreme Performance Edwards Curve Accumulated Signature scheme, our approach significantly reduces signing and verification times by 10% and 13%, respectively, compared to traditional signatures. Additionally, it offers a 10% boost in transaction throughput and block validation efficiency. Experiments conducted within various blockchain-integrated IoT setups demonstrate the scheme's effectiveness, consistently achieving improvements across diverse IoT sensor data. This highlights the innovative contribution of our scheme to the efficiency and security of blockchain technology. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Blockchain Edward curve Elliptic curve Hyperelliptic curve IoT Optimization

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：