检索结果-内蒙古大学图书馆

CLIP-AE: A Multi-modal Unsupervised Images Enhancement Method based on High-order Adaptive Curve for visual Disbalance Defects

引用

IEEE Transactions on Multimedia 2025年

作者： Wu, Jiaqi Zhang, shihao Hou, Mingshuo Wang, Zehua Chen, Wei Tian, Zijian Yu, F. Richard Leung, victor C. M. School of Artificial Intelligence Beijing100083 China University of British Columbia Department of Electrical and Computer Engineering 2332 Main Mall Vancouver VancouverBCV6T 1Z4 Canada The University of British Columbia Department of Electrical and Computer Engineering 2332 Main Mall Vancouver VancouverBCV6T 1Z4 Canada China University of Mining and Technology School of Computer Science & Technology Jiangsu Xuzhou221116 China Carleton University Department of Systems and Computer Engineering OttawaONK1S 5B6 Canada The University of British Columbia Department of Electrical and Computer Engineering VancouverBCV6T 1Z4 Canada

For visual disbalance defects (vDDs) in low-light images, such as brightness unevenness and color imbalance, existing enhancement methods struggle to extract defect features from local regions and apply adaptive enhancement based on varying degrees of these defects. To address these challenges, we propose an unsupervised multi-modal enhancement method based on a high-order adaptive curve, named CLIP-AE. specifically, we introduce a multi-modal recurrent optimization approach utilizing contrastive language-image pre-training (CLIP). This method iteratively optimizes variable embedded prompts and an Adaptive Enhancement Module (AEM) to establish dependencies between the prompts and detailed style features in the images, guiding the AEM to perform adaptive image enhancement. Additionally, we implement a progressive feature alignment strategy to enhance the model's ability to perceive style features and improve optimization efficiency by using multiple enhanced images with identical content features and incremental style features. In the AEM, the optimized Hyperparameters Generative Network (HGN) generates the optimal hyperparameters, which drive a High-Dimensional Nested Gamma correction (HDN-Gamma) to perform pixel-wise adaptive enhancement for vDDs. HDN-Gamma further maps pixel values using specific enhancement curves to avoid artifacts. Extensive experiments demonstrate that our method effectively improves visual disbalance defects and reduces artifacts. Compared to seven state-of-the-art algorithms, our method shows significant improvements (PsNR: 16.46%, 16.89%, and 15.14%;ssIM: 9.26%, 8.02%, and 9.85%;MUsIQ: 6.37%, 6.54%, and 7.45%) on the LOL, sICE, and MIT-Adobe FiveK datasets. Our approach offers a novel solution for applying multimedia technology in low-light image enhancement tasks. © 1999-2012 IEEE.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Personalized search engines on mining user preferences using clickthrough data

Personalized search engines on mining user preferences using...

引用

2014 International Conference on Information Communication and Embedded systems, ICICEs 2014

作者： Preetha, s. vimal shankar, K.N. Department of Computer Science and Engineering V.S.B Engineering College Karur India

ISBN: (纸本)9781479938346

In an web based application, different users may have different search goals when they submit it to a search engine. For a broad-topic and ambiguous query it is difficult. Here we Propose a novel approach to infer user search goals by analyzing search engine query logs. A major deficiency of generic search engines is that they follow the 'one size fits all' model and are not adaptable to individual users. This is typically shown in cases such as these: Different users have different backgrounds and interests. However, effective personalization cannot be achieved without accurate user profiles. We address the problem of learning the user profile within the user's ongoing behaviors by using the user search. We propose a framework that enables large-scale evaluation of personalized search. User interest is employed in the clustering process to achieve personalization effect. The goal of personalized IR (information retrieval) is to return search results that better match the user intent. First, we propose a framework to discover different user search goals for a query by clustering the proposed feedback sessions. Feedback sessions are get constructed from user click-through logs and can efficiently reflect the information needs of users. second, we propose an approach to generate pseudo-documents to better represent the feedback sessions for clustering. Most document-based methods focus on analyzing users' clicking and browsing behaviors recorded at the users' clickthrough data. In the Web search engines, clickthrough data are important implicit feedback mechanism from users. An example of clickthrough data for the query 'apple,' which contains a list of ranked search results presented to the user, which contains identification on the results that was previously clicked by the user. The bolded documents that have been clicked by the user have been ranked. several personalized systems that employ clickthrough data to capture users' interest have been proposed. © 2014 IE

关键词： search engines

来源：评论

学校读者我要写书评

暂无评论

Enhanced framework for credit card fraud detection using robust feature selection and a stacking ensemble model approach

Results in Engineering

引用

Results in engineering 2025年 26卷

作者： Gupta, Rahul Kumar Hassan, Asmaul Majhi, samir Kumar Parveen, Nikhat Zamani, Abu Taha Anitha, Raju Ojha, binayak singh, Abhinav Kumar Muduli, Debendra Department of Computer Science and Engineering C. V. Raman Global University Odisha Bhubaneswar India Department of Information Technology Western Governors University 4001 S 700 E 300 MillcreekUT84107 United States Department of Artificial Intelligence College of Computing and Information Technology University of Bisha Al Khuzama Saudi Arabia Department of Computer Science Faculty of Science Northern Border University Arar73213 Saudi Arabia Department of Computer Science & Engineering Koneru Lakshmaiah Education Foundation Green Fields A.P. Vaddeswaram522302 India

Credit card fraud is an emerging global issue that causes substantial financial losses and undermines consumer trust in digital transactions. With the increase in online payment volumes, conventional fraud detection technologies are increasingly confronted by the complexity of fraudulent strategies that require intelligent and scalable alternatives. This study introduces an innovative machine learning-based fraud detection framework that incorporates sophisticated preprocessing methods like sMOTE-ENN for class imbalance mitigation, autoencoder for dimensionality reduction, and TOPsIs for optimal feature selection. A stacking ensemble model is developed with support vector machine (svM), K-nearest neighbors (KNN), and extreme learning machine (ELM) to enhance forecast accuracy. The particle swarm optimization (PsO) algorithm is employed to optimize ELM parameters, enhancing generalization and model convergence. Extensive tests with standard datasets show outstanding results, achieving 99.95% accuracy, 99.93% precision, and 99.97% recall in detecting fraud. The outcomes highlight the model's proficiency in properly identifying fraudulent transactions while reducing false positives. The proposed method provides a viable alternative for secure and efficient credit card fraud detection in the contemporary digital economy, characterized by high accuracy and real-time scalability. © 2025

关键词： Dimensionality reduction

来源：评论

学校读者我要写书评

暂无评论

Developing a Robust Intrusion Detection system Using sMOTE and Hybrid svNN Model

Developing a Robust Intrusion Detection System Using SMOTE a...

引用

sentiment Analysis and Deep Learning (ICsADL), International Conference on

作者： b. Kiranmayee M. sharmila Devi Kalva susheela Raghu Dhumpati Kiran Kumar Reddy Penubaka U Ganesh Naidu Department of CSE (AI&ML) G Pulla Reddy Engineering College Kurnool India Department of CSE Santhiram Engineering College Nandyal India Department of Computer Science Sir C R Reddy College for Women Eluru India Department of CSE Bahrain Polytechnic College Bahrain Department of CSE-AIML MLR Institute of Technology Hyderabad India B V Raju Institute of Technology Medak Telangana India

ISBN: (数字)9798331523923

ISBN: (纸本)9798331523930

This paper proposes an integrated approach to build up IDs with proper effectiveness toward the rising need for strong network security. Network traffic anomaly detection and classification are one of the major aims and enhance the security layer against various types of cyber threats. This study is a methodical approach in which a diverse set of data is first extracted from Kaggle. The collected dataset is a comprehensive one that includes various kinds of network traffic data. The first step includes preprocessing the data, i.e., handling missing values, removing erroneous entries, and dealing with outliers using the Interquartile Range (IQR) method. To counter class imbalance, synthetic Minority Oversampling Technique (sMOTE) is utilized in generating synthetic samples in underrepresented classes for generalization of models. The feature selection algorithm relies on the Chi-square Lasso Feature selection method. This method unites Chi-square tests to select statistically significant features with Lasso regularization for feature dimension reduction with the goal of overfitting minimization. For building models, a hybrid of support vector Machine and K-Nearest Neighbors, which is termed as svNN, is used. by using soft voting to aggregate the outputs of svM and KNN, the proposed svNN model improves the prediction accuracy, the degree of robustness, and adaptability. The presented methodology has achieved high classification accuracy 93.28%, precision 93.49%, recall 93.28%, F1 score 93.23%, and low RMsE of 0.2592. These metrics demonstrate that the model is reliable and of practical use for intrusion detection in dynamic, high-dimensional environments of networks, providing a proper solution to modern security network challenges.

关键词： support vector machines Adaptation models Analytical models Accuracy Intrusion detection Telecommunication traffic Nearest neighbor methods Network security Feature extraction Data models

来源：评论

学校读者我要写书评

暂无评论

Network Traffic Prediction using Deep Learning Models: Reducing Latency in High-Demand scenarios

Network Traffic Prediction using Deep Learning Models: Reduc...

引用

Computing for sustainability and Intelligent Future (COMP-sIF), International Conference on

作者： b.J. Job Karuna sagar Divya M v Namita Kale Neetu Anand Kumar Gaurav shrividya shashidhara Department of CSE Joginpally B.R Engineering College Hyderabad Telangana Department of Computer Science East West Institute of Technology Bangalore India Department of Information Technology MET'S Institute Of Engineering Nashik India Department of Computer Applications Maharaja Surajmal Institute Delhi India Department of CSE BMS Institute of Technology and Management Bangalore India

ISBN: (数字)9798331538538

ISBN: (纸本)9798331538545

The availability of the internet and enhanced applications, more specifically, the utilization of more devices in networks, and improved computational applications has created complexity in the control of network traffic. Network traffic forecasting is an essential component in low-latency networks, as well as in real-time applications that deserve excellent service quality and efficient resource allocation. This study therefore focuses on identifying whether deep learning models can further improve the accuracy in forecasting traffic flows in a network. Therefore, with the help of LsTM networks and Convolutional Neural Networks, the work will build temporal dependencies and spatial correlations of network traffic data. There are major and minor adaptations to some classic methodologies that conform the proposed framework; moreover, it is applied to two real datasets to highlight the capacity of the model in anticipating traffic congestion with a higher level of accuracy than the traditional methodologies. Evaluation of models in dynamic network conditions is done through common parameters such as prediction accuracy, lowered latency, and scalability. saved models show that deep learning-based solutions reduce latency at high loads much more effectively, which will help network administrators in finding efficient solutions. This study demonstrate how deep learning can aid in the prediction of traffics, and the manner in which it is changing the efficiency and effectiveness of modern-day network systems. It makes the foundation for improved understanding of traffic patterns in order to apply it to optimize future generation of communication networks.

关键词： Deep learning Adaptation models Accuracy Telecommunication traffic Predictive models Throughput Convolutional neural networks Forecasting Long short term memory Load modeling

来源：评论

学校读者我要写书评

暂无评论

vehicle-Assisted service Caching for Task Offloading in vehicular Edge Computing

引用

IEEE Transactions on Mobile Computing 2025年

作者： Jiang, Hongbo Cai, Jianghao Xiao, Zhu Yang, Kehua Chen, Hongyang Liu, Jiangchuan Hunan University College of Computer Science and Electronic Engineering Hunan Changsha China Hunan University Shenzhen Research Institute Shenzhen518055 China Zhejiang Laboratory Research Center for Graph Computing Hangzhou311100 China Simon Fraser University School of Computing Science BurnabyBCV5A 1S6 Canada

The development of artificial intelligence (AI) enables vehicular edge computing (vEC) servers to be able to provide more intelligent services. However, the limited storage resources of vEC servers constrain the deployment of intelligent service contents, which greatly restricts the intelligence level of the vEC network. To resolve this problem, we first design a novel vehicle-assisted vEC network architecture and further propose vaCo, a vehicle-assisted Collaborative caching system. vaCo allows vEC servers to download the cached service content from any vehicle in the vEC network to support task offloading. vaCo mainly considers the real-time scheduling problem of vehicle storage resources under the dynamic vEC network and the benefit problem caused by invoking vehicle resources under the highly dynamic load environment. vaCo models the vehicle storage resources as an independent resource pool and deploys a cross-vEC server content retrieval mechanism to achieve unified and efficient management of the storage resources of the vehicle cluster and the vEC server cluster. Then, we propose a multi-swarm collaborative optimization scheme to jointly optimize the service failure rate and cost, and further propose a Pareto-based optimization scheme to ensuring that vaCo can correctly evaluate the benefits of invoking vehicle resources in a dynamic vEC network. Finally, we implement vaCo and conduct extensive evaluations on real-world dataset. The experimental results on the real trajectory dataset show that vaCo can effectively utilize vehicle resources and ensure the benefits of both vehicles and vEC servers simultaneously. © 2002-2012 IEEE.

关键词： Mobile edge computing

来源：评论

学校读者我要写书评

暂无评论

Leveraging an Optimized Deep Convolutional Neural Network Architecture for the Automated Diagnosis of Pathological Conditions Through Acoustic Analysis of Human Cough signatures

Leveraging an Optimized Deep Convolutional Neural Network Ar...

引用

smart systems for applications in Electrical sciences (ICssEs), International Conference on

作者： Tamilarasan M Keshavamurthy T G vidya K v Nicy siby sriraksha Devaraj Revanna b R Department of Computer Science and Engineering Dr.N.G.P Institute of Technology Coimbatore Anna University Tamil Nadu India Department of Electronics and Communication Engineering Sri Siddhartha Institute of Technology Sri Siddhartha Academy of Higher Education Tumkur Karnataka India Department of Electronics and Communication and Engineering B.M.S College of Engineering Affiliated to Visvesvaraya Technological University Belgavi Bangalore Karnataka India

ISBN: (数字)9798331507671

ISBN: (纸本)9798331507688

The usage of artificial intelligence (AI) techniques particularly deep learning algorithms, has recently increased dramatically and the healthcare sector is only one of many industries affected Using deep convolutional neural networks (CNNs) enhanced with human auditory inputs we present a novel approach to disease detection in this paper The proposed system aims to provide an effective economical and non invasive method of early disease detection we employ a deep convolutional neural network (CNN) architecture that has been optimized for audio input processing. Further more we use data augmentation methodologies to strengthen the model's capacity to generalize and handle varied patient groups and diseases more effectively. Our all encompassing approach incorporates denoising techniques using Wavelet transform denoising segmentation via K-mean clustering, feature extraction using Recursive Feature Elimination (RFE) and classification using Enhanced Deep Convolutional Neural Networks (EDCNN) which improves the accuracy and dependability of the system in identifying diseases based on human sound signatures.

关键词： Deep learning Accuracy Noise reduction Medical services Feature extraction Data augmentation Data models Convolutional neural networks Artificial intelligence Diseases

来源：评论

学校读者我要写书评

暂无评论

Joint Adaptation for Mobile 360-Degree video streaming and Enhancement

引用

IEEE Transactions on Mobile Computing 2025年

作者： Guo, Haotian Wang, Feng Zhang, Wei Zhu, Yifei Cui, Laizhong Liu, Jiangchuan Yu, F. Richard Zhang, Lei Shenzhen University College of Computer Science and Software Engineering Guangdong Shenzhen518060 China The University of Mississippi Department of Computer and Information Science MS38677 United States ByteDance Guangdong Shenzhen518060 China University of Michigan-Shanghai Jiao Tong University Joint Institute Shanghai200240 China School of Computing Science Simon Fraser University BurnabyBCV5A 1S6 Canada Carleton University School of Information Technology OttawaONK1S 5B6 Canada

Tile-based streaming and super resolution (sR) are two representative technologies adopted to improve bandwidth efficiency of 360° video streaming. The former allows selective downloading of contents in the user viewport by splitting the video into multiple independently decodable tiles. The latter leverages client-side computation to enhance the received video to higher quality using advanced neural network models. In this work, we propose a Collaborated streaming and Enhancement (CsE) adaptation framework for mobile 360° videos, which integrates super resolution with tile-based streaming to optimize the user experience with dynamic bandwidth and limited computing capability. To effectively enhance the tile-based video streaming through sR, we propose to adaptively group the tiles for quality enhancement adapting to the content similarity. We also identify and address several key design issues to integrate sR into tile-based video streaming including unified video quality assessment, computational complexity model for super resolution, and buffer analysis considering the interplay between transmission and enhancement. We further formulate the quality-of-experience (QoE) maximization problem for mobile 360° video streaming and propose a rate adaptation algorithm to make the best decisions for download and for enhancement based on the Lyapunov optimization theory. Extensive evaluation results validate the superiority of our proposed approach, which demonstrates stable performance with considerable QoE improvement, while enabling a trade-off between playback smoothness and video quality. © 2002-2012 IEEE.

关键词： video streaming

来源：评论

学校读者我要写书评

暂无评论

Assessing synergies between thermal energy and torrefaction severity index of wood spruce sawdust via machine learning algorithms

引用

Journal of Analytical and Applied Pyrolysis 2025年 190卷

作者： Šafář, Michal Raclavska, Helena Růžičková, Jana Ali, Imtiaz Naqvi, salman Raza scala, Fabrizio CEET/ENET Centre VŠB – Technical University of Ostrava Ostrava-Poruba708 00 Czech Republic Department of Chemical Engineering College of Engineering Prince Mohammad Bin Fahd University Al Khobar Saudi Arabia Department of Engineering and Chemical Sciences Karlstad University Karlstad Sweden Faculty of Electrical Engineering and Computer Science VŠB-Technical University of Ostrava 17. Listopadu 2172/15 Poruba Ostrava708 00 Czech Republic DICMaPI Università di Napoli Federico II P.le Tecchio 80 Napoli80125 Italy STEMS CNR P.le Tecchio 80 Napoli80125 Italy

Assessing the synergies between thermal energy and the torrefaction severity index by elucidating the effects of biomass torrefaction conditions on product characteristics is still a relevant research question. This study explores the optimization of torrefaction of spruce wood sawdust by analyzing the chemical and physical features of the resultant material. The study employs thermogravimetric analysis and Thermal Desorption-Gas Chromatography/Mass spectrometry (TD-GC/Ms) to examine the changes in the components during torrefaction. The torrefaction process has a profound effect on the composition, causing conversion of up to 30 % of the initial mass to volatile organic compounds and incondensable gases when subjected to a temperature of 300 °C. The correlation matrix highlights the relationships between variables, including time, temperature, heating value, mass yield, energy yield, and ratios of C, H, and O. The matrix visually represents the interplay between these factors during the torrefaction process. Principal component analysis (PCA) and hierarchical cluster analysis (HCA) revealed that torrefaction severity index, cellulose, lignin, torrefaction time, and temperature correlate positively, while H/C, O/C, and hemicellulose content correlate negatively. The ANN model exhibited superior predictive accuracy (R² = 0.99982, RMsE = 0.00359), surpassing C&RT (R² = 0.99483, RMsE = 0.01943), KNN (R² = 0.99467, RMsE = 0.01974), and svM (R² = 0.99022, RMsE = 0.02674), thus validating the efficacy of machine learning for precise torrefaction severity index (TsI) prediction. This finding enhances the efficiency of biomass processing and provides reliable tools for future research in the field, thereby informing and guiding future studies and industrial applications. © 2025 The Authors

关键词： Hierarchical clustering

来源：评论

学校读者我要写书评

暂无评论

Advancing HRM with AI Powered Recruitment Performance Management and Employee Engagement Using One Class svMs

Advancing HRM with AI Powered Recruitment Performance Manage...

引用

Integrated Circuits and Communication systems (ICICACs), IEEE International Conference on

作者： Rafiya banu Ruchi Tripathi saumya sharma Loganathan Chenniappan K. Chandrasekaran Umanesan R Department of Commerce B. S. Abdur Rahman Crescent Institute of Science and Technology Chennai India Faculty of Management Studies SRM-IST Ghaziabad India Department of Management & Commerce The Northcap University Gurugram India Business Administration IBRI Sultanate of Oman Department of Mechanical Engineering K.S.R College of Engineering Tiruchengode India Department of Computer Science and Engineering School of Computing Vel Tech Rangarajan Dr Sagunthala R&D Institute of Science and Technology Chennai India

ISBN: (数字)9798331508456

ISBN: (纸本)9798331508463

AI is making labour-intensive HRM operations like hiring, performance review, and employee engagement more efficient and effective. The incorporation of AI technology in HRM could change several sectors. Through efficient applicant sourcing, screening, and selection, AI-driven solutions simplify hiring processes and ensure fair and accurate picks. Performance management using AI-powered solutions provides data-driven feedback and customized development plans to align individual progress with organizational goals. Predictive analytics improve employee engagement by personalizing experiences and meeting needs. The method focuses on preprocessing and model training. Data are cleansed, ordinal and categorical variables encoded, scaling characteristics and engineering features applied, and dimensional homogeneity assured before processing. For model training, the simple and robust one-class svM technique is used. One-class svM model outperformed others with 95.28%. This method worked best compared to GAN and GRU models. AI could improve HRM by improving decision-making and offering data-driven solutions. The findings show that the proposed model outperforms current options, providing the framework for HRM AI applications.

关键词： support vector machines Training Productivity Accuracy Uncertainty Reviews Artificial intelligence Integrated circuit modeling Predictive analytics Recruitment

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：