检索结果-内蒙古大学图书馆

How graph convolutions amplify popularity bias for recommendation?

Frontiers of computer Science 2024年第5期18卷 121-132页

作者： Jiajia CHEN Jiancan WU Jiawei CHEN Xin XIN Yong LI Xiangnan HE School of Information Science and Technology University of Science and Technology of ChinaHefei 230026China School of Computer Science and Technology Zhejiang UniversityHangzhou 310058China School of Computer Science and Technology Shandong UniversityQingdao 250100China Department of Electronic Engineering Tsinghua UniversityBeijing 100084China

Graphconvolutional networks(GCNs)have become prevalent in recommender system(RS)due to their superiority in modeling collaborative *** improving the overall accuracy,GCNs unfortunately amplify popularity bias-tail items are less likely to be *** effect prevents the GCN-based RS from making precise and fair recommendations,decreasing the effectiveness of recommender systems in the long *** this paper,we investigate how graph convolutions amplify the popularity bias in *** theoretical analyses,we identify two fundamental factors:(1)with graph convolution(i.e.,neighborhood aggregation),popular items exert larger influence than tail items on neighbor users,making the users move towards popular items in the representation space;(2)after multiple times of graph convolution,popular items would affect more high-order neighbors and become more *** two points make popular items get closer to almost users and thus being recommended more *** rectify this,we propose to estimate the amplified effect of popular nodes on each node's representation,and intervene the effect after each graph ***,we adopt clustering to discover highly-influential nodes and estimate the amplification effect of each node,then remove the effect from the node embeddings at each graph convolution *** method is simple and generic-it can be used in the inference stage to correct existing models rather than training a new model from scratch,and can be applied to various GCN *** demonstrate our method on two representative GCN backbones LightGCN and UltraGCN,verifying its ability in improving the recommendations of tail items without sacrificing the performance of popular *** are open-sourced^(1)).

关键词： recommendation graph convolution networks popularity bias

来源：评论

学校读者我要写书评

暂无评论

A biased edge enhancement method for truss-based community search

引用

Frontiers of computer Science 2024年第3期18卷 251-253页

作者： Yuqi LI Tao MENG Zhixiong HE Haiyan LIU Keqin LI School of Computer and Information Engineering Central South University of Forestry and TechnologyChangsha 410082China School of Business Central South University of Forestry and TechnologyChangsha 410082China College of Information Engineering Changsha Medical UniversityChangsha 410219China Department of Computer Science State University of New YorkNew York 12561USA

1 Introduction A related study called community search,whose target is to find dense subgraphs containing the given node,has drawn a growing amount of attention recently[1].To explore the higher-order structure of complex networks,truss-based community search methods[2]have been ***,the truss-based hypergraph constructed from the original graph is frequently fragmented and consists of numerous subgraphs and isolated nodes[3],which boils down to the fact that these methods often pay only attention to the truss connections but ignore the lower-order connectivity of the original graph.

关键词： graph. truss connectivity

来源：评论

学校读者我要写书评

暂无评论

Understanding Time Series Analysis in Early Detection of Unclassified Endocrine Tumors

Understanding Time Series Analysis in Early Detection of Unc...

引用

2024 International Conference on Optimization Computing and Wireless Communication, ICOCWC 2024

作者： Seth, Jyoti Prakash, Febin Singh, Vaishali Vivekananda Global University Department of Computer Science & Engineering Jaipur India Department of Computer Science and Information Technology Bangalore India Maharishi School of Engineering & Technology Maharishi University of Information Technology Uttar Pradesh India

ISBN: (纸本)9798350383348

Endocrine tumors are malignant tumors that get up inside the endocrine system, a network of glands and organs responsible for producing hormones that affect the physiological procedures of the frame. Although endocrine tumors are usually unusual, early detection is critical for successful treatment. Due to the lack of reliable medical markers for endocrine tumors, early detection is mainly predicated on imaging and laboratory tests. However, these checks may be luxurious and can be hard to interpret. In recent years, time collection analysis (TSA) has been gaining popularity as a powerful device for the early detection of unclassified endocrine tumors. Time series analysis is a form of statistical evaluation that applies mathematical fashions to datasets by converting data values over a hard and fast time frame. It's miles used to discover trends in the facts, making it possible to discover abnormal behavior that could indicate contamination. This method has been shown to have a high degree of accuracy. It may provide insight into, in any other case, unclassified endocrine tumors, making an allowance for the set-off detection and remedy. In this paper, we discuss the capacity of time collection evaluation inside the early detection of unclassified endocrine tumors, alongside the demanding situations and opportunities associated with its use. Time series evaluation has been established to be a powerful tool for the early detection of unclassified endocrine tumors. By presenting perception into patterns and developments in temporal information, this method allows studies to pinpoint regions of molecular change from which ability biomarkers can be diagnosed. The technique is predicated on measuring and analyzing adjustments inside the gene expression levels of regulation networks over time. This method has been used to become aware of adjustments within the expression of pathways related to G protein-coupled receptors and pathways associated with endocrinology. Moreove

关键词： detection Endocrine illnesses improvement reliable

来源：评论

学校读者我要写书评

暂无评论

Air Predictive Modeling for Air Quality: A Comparative Study of Machine Learning and Deep Learning Techniques

Air Predictive Modeling for Air Quality: A Comparative Study...

引用

2024 IEEE International Students' Conference on Electrical, Electronics and computer Science, SCEECS 2024

作者： Dhongade, Shubhangi Tiwaskar, Shweta Koturwar, Abhijeet Motiramani, Raksha Zarekar, Devyani Vishwakarma Institute of Information Technology Department of Computer Engineering Maharashtra Pune India

ISBN: (纸本)9798350348460

Air quality emerges as a critical global concern affecting millions of individuals. This paper explores into the development and application of data-driven models for air quality prediction, Motivated by air quality's profound impact on both the environment and human well-being, we leverage deep learning and machine learning methodologies to enhance the precision of air quality forecasts. Dataset used comprising over 33,360 entries. This dataset encompasses diverse characteristics, including PM2.5, PM10, SO2, NO2, CO, O3, etc., with pollutant levels documented on an hourly basis. To gauge air quality levels, we employ various models such as the Random Forest model, Long Short-Term Memory (LSTM), Artificial Neural Network (ANN), Support Vector Machine (SVM), and K-Nearest Neighbors (K-NN). A thorough evaluation and performance comparison of these models is conducted using Kaggle's air quality dataset. The study seeks to not only address the immediate challenge of air quality prediction but also holds implications for broader environmental and public health considerations on a global scale. © 2024 IEEE.

关键词： Artificial Neural Network Deep learning K-Nearest Neighbors and Support Vector Machine Long Short-Term Memory Random Forest model

来源：评论

学校读者我要写书评

暂无评论

A novel computer-Aided Estimation of Breast Cancer via Time Series Analysis

A novel Computer-Aided Estimation of Breast Cancer via Time ...

引用

2024 International Conference on Optimization Computing and Wireless Communication, ICOCWC 2024

作者： Bhambu, Pawan Haripriya Yadav, Dhyan Chandra Vivekananda Global University Department of Computer Science and Engineering Jaipur India Department of Computer Science and Information Technology Bangalore India Maharishi School of Engineering & Technology Maharishi University of Information Technology Uttar Pradesh India

ISBN: (纸本)9798350383348

Introduces a estimation based totally on time series evaluation. among ladies, and it's far more critical to stumble on and deal with it in the early ranges. However, due to time consumption within the analysis method and its cost, there's a need to increase pc-aided structures to automate the analysis. It proposes a time series evaluation method based on machine mastering for detecting and estimating breast most cancers degree (early or past due) at the preliminary ranges of prognosis. The proposed device is predicated on the type of records amassed in an early stage of breast most cancers analysis, along with age, tumor size, lymph node involvement, and other laboratory values, which includes estrogen and progesterone receptor degrees. The proposed machine changed into tested using a dataset of patients with breast cancers. The device's performance was evaluated using the region under the curve and F1 rating. a unique method estimating possibility of diagnosis through time series evaluation. The proposed method applies a sparse illustration model for pre-processing, observed via the utility of a random wooded area classifier to evaluate the chance of breast cancer. public dataset of 961 female sufferers with breast cancer from the UCI machine studying repository. The proposed approach is faster and more correct than the associated methods. The concept of using time collection evaluation for estimating breast cancers is, for this reason promising as it could be deployed in medical exercise to offer reliable detection of breast cancer. © 2024 IEEE.

关键词： Time series analysis

来源：评论

学校读者我要写书评

暂无评论

Study of adsorption of Congo red on ferric oxide@straw magnetic composites 3

Study of adsorption of Congo red on ferric oxide@straw magne...

引用

2023 3rd International Conference on Advanced Materials and Chemical engineering, AMCE 2023

作者： Qu, Yanping Ma, Huimin Liu, Jiabin Luo, Yuyue Yue, Yinqi Wang, Jingqi Cheng, Ying School of Computer and Information Engineering Qilu Institute of Technology Shandong Jinan China

In this paper, citric acid-modified corn stalk and Fe3O4 were used to make magnetic straw adsorbent to realize rapid separation of adsorbent and wastewater. The influences of pH, dosage, reaction time, and initial stress on the Congo red adsorption rate were investigated. Then, we used adsorption kinetics and isothermal linear adsorption adjustment to explore the adsorption mechanism. Studies have shown the increasing doses of magnetic adsorbents on straw. When the dose was 8 g/L, the commercial adsorption rate was 95%. The adsorption process of magnetic straw adsorbent for Congo red was exothermic. Adsorption was mainly chemisorption, and the adsorption process was relatively fast, reaching adsorption equilibrium after about 90 min. © Published under licence by IOP Publishing Ltd.

关键词： Adsorption

来源：评论

学校读者我要写书评

暂无评论

Fortifying Healthcare Data Security in the Cloud:A Comprehensive Examination of the EPM-KEA Encryption Protocol

引用

computers, Materials & Continua 2024年第5期79卷 3397-3416页

作者： Umi Salma Basha Shashi Kant Gupta Wedad Alawad SeongKi Kim Salil Bharany Computer Science and Information Technology Jazan UniversityJazanSaudi Arabia CSE Eudoxia Research UniversityNew CastleUSA Department of Information Technology College of Computer ScienceQassim UniversityBuraydahSaudi Arabia National Center of Excellence in Software Sangmyung UniversitySeoulKorea Department of Computer Science and Engineering Lovely Professional UniversityPhagwaraPunjabIndia

A new era of data access and management has begun with the use of cloud computing in the healthcare *** the efficiency and scalability that the cloud provides, the security of private patient data is still a majorconcern. Encryption, network security, and adherence to data protection laws are key to ensuring the confidentialityand integrity of healthcare data in the cloud. The computational overhead of encryption technologies could leadto delays in data access and processing rates. To address these challenges, we introduced the Enhanced ParallelMulti-Key Encryption Algorithm (EPM-KEA), aiming to bolster healthcare data security and facilitate the securestorage of critical patient records in the cloud. The data was gathered from two categories Authorization forHospital Admission (AIH) and Authorization for High Complexity *** use Z-score normalization forpreprocessing. The primary goal of implementing encryption techniques is to secure and store massive amountsof data on the cloud. It is feasible that cloud storage alternatives for protecting healthcare data will become morewidely available if security issues can be successfully fixed. As a result of our analysis using specific parametersincluding Execution time (42%), Encryption time (45%), Decryption time (40%), Security level (97%), and Energyconsumption (53%), the system demonstrated favorable performance when compared to the traditional *** suggests that by addressing these security concerns, there is the potential for broader accessibility to cloudstorage solutions for safeguarding healthcare data.

关键词： Cloud computing healthcare data security enhanced parallel multi-key encryption algorithm(EPM-KEA)

来源：评论

学校读者我要写书评

暂无评论

Motion Retargeting from Human in Video to 3D Characters with Different Skeleton Topology 4

Motion Retargeting from Human in Video to 3D Characters with...

引用

4th International Conference on Consumer Electronics and computer engineering, ICCECE 2024

作者： Guo, Hao Li, Shujie Jia, Wei Hefei University of Technology Department of Computer Science and Information Engineering Hefei China

ISBN: (纸本)9798350384062

Motion retargeting from videos to 3D virtual character is a challenging task in computer vision and computer graphics. A solution is first to extract the 3D motion sequences from videos using human pose estimation algorithm and then retarget to 3D virtual character by skeleton motion retargeting algorithm. However, manual skeleton mapping is often necessary because of the difference in skeleton topology between pose estimation and 3D virtual character, introducing uncertainty and potentially compromising the semantics of the original motion. Additionally, the results of pose estimation contain noise while existing skeleton motion retargeting algorithms do not consider the influence of noise inputs in their design. In response to the above issues, we propose a novel motion retargeting framework and introduce a new motion retargeting network named CTBRnet (Combined Transformer Bi-LSTM Retargeting network). The framework leverages existing algorithm to generate paired training data, providing supervision to CTBRnet and enabling CTBRnet to learn potential skeleton mapping. The framework utilizes human motion videos as input and generates retargeting results for virtual characters, which are driven by joint rotations aligned with the virtual character's skeletal topology. Our proposed network combines transformer and Bi-LSTM, effectively capturing stable high-dimensional motion features from noisy results of pose estimation in both global and local perspectives. The experimental results show that our framework can achieve motion retargeting from videos to 3D virtual character with different skeleton topology, and compared with existing skeleton motion retargeting networks, CTBRnet reduces joint position errors while enhancing the smoothness. © 2024 IEEE.

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

Intelligent Cost-Effective Electric Vehicle Charging Solutions for Cities and Highway Networks 17

Intelligent Cost-Effective Electric Vehicle Charging Solutio...

引用

17th International Conference on COMmunication Systems and NETworkS, COMSNETS 2025

作者： Lohia, Akshat Pochhi, Aadesh Mishra, Shivansh Chakraborty, Nilotpal Indian Institute of Information Technology Department of Computer Science and Engineering Assam Guwahati India

ISBN: (纸本)9798331531195

In this work, we address the strategic placement and optimal sizing of electric vehicle charging stations for cities as well as highway traffic to minimize overall cost. We formulate the problem as a Mixed Integer Linear Programming (MILP) model and propose a set of greedy algorithms to obtain scalable solutions quickly. For city areas, we propose a systematic approach considering factors like population density and traffic patterns to maximize customer satisfaction and minimize costs. For highways and expressways, we account for deterministic inter-arrival times and maximum waiting times. Additionally, we propose an integrated approach that, when run for two cities and a connecting highway, demonstrates a total cost reduction compared to using separate algorithms for cities and highways. © 2025 IEEE.

关键词： Integer linear programming

来源：评论

学校读者我要写书评

暂无评论

Steel Surface Defect Detection Algorithm Based on Improved YOLOv8 21

Steel Surface Defect Detection Algorithm Based on Improved Y...

引用

21st International Conference on Networking, Sensing and Control, ICNSC 2024

作者： Che, Jie Yao, Xinrui Wang, Yuqi Li, Bohan Xu, Hangrui Liu, Jian School of Computer Science and Information Engineering Hefei University of Technology Hefei China

ISBN: (纸本)9798350365221

Surface defect detection technology is a vital component of the steel industry that has garnered significant attention from the academic community in recent times. While modern methods with deep learning-based object detection provide better detection accuracy than traditional approaches, production requirements are often not met by their inference time. This study suggested an improved method based on the industry-leading object detection algorithm YOLOv8 to address this issue and effectively detect steel surface defects while keeping a reasonable balance between speed and accuracy. First, a Stem module based on YOLOv8s is added to downsample the image and lessen the effect of redundant features in the original image. In the meantime, the backbone network gains a large kernel depthwise convolution module called InceptionNeXt+, which improves the model's ability to extract features and has a larger effective receptive field. Lastly, the base model's SPPF module is changed to a RsimSPPF module, which increases the model's accuracy and speeds up inference. The improved YOLOv8 model confirms the efficacy of the improved method with experimental results on the NEU-DET dataset achieving a mAP of 76.9%, increasing by 3.3 compared to the baseline model. © 2024 IEEE.

关键词： Steelmaking

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：