检索结果-内蒙古大学图书馆

international conference on System Modeling & Advancement in Research Trends (SMART)

作者： Sofia Goel Sudhansh Sharma RC Tripathi Indira Gandhi National Open University New Delhi India Faculty of Engineering and Computing Sciences TMU Moradabad UP

ISBN: (纸本)9781665439718

Diabetes mellitus is a chronic disease that produces blood glucose abnormalities. this needs to be monitored at regular intervals. Many machine learning models made significant contribution in the prediction of diabetes. One such deep learning model is the convolution neural network. It is a multi-layered network architecture that comprises convolution, pooling, output layers, and activation functions. Activation functions are used to learn and approximate the relationships between various features in the model for final output. this research paper focus on comparing the performance of the CNN model using four non-linear activation functions; sigmoid, tanh, ReLU, ELU for blood glucose prediction. the model uses a simglucose UVA/Padova dataset of 30 patients for experimental analysis. Additionally, the model presents a three-class classification problem, where the patient is classified as euglycemic, hypoglycaemic, and hyperglycaemic. Statistical measures like recall, precision, F-score, and accuracy are used to assess the performance of CNNs employing various activation functions. the experimental results show that CNN gives the maximum accuracy of 97.3% when applied with the ELU activation function.

关键词： Deep learning Training Convolution Neural networks Predictive models Glucose Diabetes

来源：评论

学校读者我要写书评

暂无评论

Robust Zero Watermarking Algorithm for Medical Images Using Local Binary pattern and Discrete Cosine Transform 8th

Robust Zero Watermarking Algorithm for Medical Images Using ...

引用

8th international conference on Artificial Intelligence and Security , ICAIS 2022

作者： Liu, Wenyi Li, Jingbing Shao, Chunyan Ma, Jixin Huang, Mengxing Bhatti, Uzair Aslam School of Information and Communication Engineering Hainan University Haikou570228 China School of Computing and Mathematical Sciences University of Greenwich LondonSE10 9LS United Kingdom State Key Laboratory of Marine Resource Utilization in the South China Sea Hainan University Haikou570228 China

ISBN: (纸本)9783031067631

thanks to the rapid development of the telemedicine, medical treatment has turned in great convenience. However, as the internet network can be addressed by any user, the medical images cannot avoid the defects of suffering from information leakage and require information encryption techniques. To protect the privacy of patients in medical images, a robust watermarking method, which combines medical images with watermark information, is proposed in this paper. First, the low-frequency feature vector of the medical image is extracted by combining local binary pattern (LBP) and transformation with discrete cosine transform (DCT). Second, hash sequence is implemented on to convert feature vector into binary sequences. At last, the medical image is embedded with hash sequence and watermark preprocessed by chaotic encryption technology using the zero-watermark technology. the experimental results show that the method can encrypt the medical images’ watermark effectively and has good robustness against geometric attacks. © 2022, the Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Local binary pattern

来源：评论

学校读者我要写书评

暂无评论

Digital technologies as a promotor of well-being and inclusion of people with intellectual and developmental disabilities: What is the current situation? 10

Digital technologies as a promotor of well-being and inclusi...

引用

10th international conference on Software Development and Technologies for Enhancing Accessibility and Fighting Info-Exclusion, DSAI 2022

作者： Tymoshchuk, Oksana Martins, Isabel Catarina Almeida, Ana Margarida Pisco Cartaxo, Cátia Raquel Albuquerque, Eulália DigiMedia Department of Communication and Art University of Aveiro Portugal AVISPT21 PAIS-EM-REDE Portugal

ISBN: (纸本)9781450398077

Digital technologies, including computers, mobile phones, the internet, and social networks, are widely used worldwide. these devices and technologies can be of great use helping people with intellectual and developmental disabilities (IDD) due to their potential to access information and make social contacts easier, promoting their well-being and social inclusion. Digital technologies can help removing barriers to their use and promoting equal rights and opportunities for all. However, little research has been done to explore the patterns of use and difficulties in using digital technologies experienced by people with IDD. this study seeks to overcome this gap by surveying information on the digital skills of people with IDD, their habits of using digital technologies, and the factors that facilitate or hinder this use. this study surveyed 78 people with IDD about their knowledge, frequency of use, goals, and autonomy in using digital technologies. the results demonstrate the need to implement new approaches to promote the digital inclusion of people with IDD, ensuring that they have access to digital technologies as well as the possibility of learning to use them flexibly and adapt to their conditions. Moreover, the results of this study highlight the importance of developing support networks that involve school, family, and community that can offer a range of initiatives to promote the digital and social inclusion of individuals with IDD. © 2022 ACM.

关键词： Social sciences computing

来源：评论

学校读者我要写书评

暂无评论

Layer-2 Routing Analytic Model by Linear Programming

Layer-2 Routing Analytic Model by Linear Programming

引用

international conference on Advanced Communication Technology (ICACT)

作者： Changjin Suh Shin-woong Jung School of Computing Soongsil University Seoul South Korea

Carrier Ethernet integrates distributed layer-2 bridge networks to a backbone network. Since carrier Ethernets use simple layer-2 routing, its performance is lower than that of layer-3 routing methods. We introduce an analytic model that evaluates the routing methods in backbone networks. the model solves the multi- input and multi-output maximum flow problem in backbone networks using linear programming. Our method can be used as follows. Firstly, the model suggests appropriate link capacity if network topology and traffic pattern is given. Secondly, the model shows and compares the performance of layer-2 routing methods including perfect routing and our proposed routing called ENDIST(Edge Node Divided Spanning Tree). In this study, the performance is the maximum throughput that a given routing method can reach. We additionally found many good results about ENDIST. Our analytic model proved that ENDIST outperforms the existing layer-2 standard protocols such as STP, MSTP for any offered load and performs perfectly under medium and light traffic load in the given network, thus enough to say that ENDIST is as good as layer 3 routing from the point of throughput.

关键词： Routing Linear programming Spine Ethernet networks Telecommunication traffic throughput Bridges Network topology Traffic control Performance analysis

来源：评论

学校读者我要写书评

暂无评论

Analysis of the Effect pattern of Mixing Ratio and Age on the Conductivity of Concrete 10

Analysis of the Effect Pattern of Mixing Ratio and Age on th...

引用

10th international conference on Advances in Machinery, Material Science and Engineering Application, MMSE 2024

作者： Chi, Hongyou Zhang, Shenghang Tang, Lei Li, Jun Li, Po Pan, Wenming Guangdong GDH Pearl River Delta Water Supply Co. Ltd Guangzhou510000 China Nanjing Hydraulic Research Institute Nanjing210029 China

ISBN: (纸本)9781643685465

Concrete is the most widely used building material, and its electrical index test has received more attention in recent years. the samples of different mix proportions (water-cement ratios, sand rate, accelerator) were made and tested by impedance analysis. the results show that the impedance analysis method can eliminate the polarization effect and test the electrical properties more stably;With the increase of age, the relationship between electrical conductivity and water cement ratio gradually changed from positive correlation to negative correlation;the conductivity of the sample with accelerator was slightly higher than that without accelerator;the electrical conductivity of concrete is basically negatively correlated with the sand content. there is no order of magnitude difference in the electrical index of concrete in the concrete mix ratio, and the difference gradually decreases with the increase of concrete age. the experimental results can provide reference for electromagnetic detection work. © 2024 the Authors.

关键词： Cements

来源：评论

学校读者我要写书评

暂无评论

Partial pattern fulfillment and its application in event processing: poster 16

Partial pattern fulfillment and its application in event pro...

引用

Proceedings of the 10th ACM international conference on Distributed and Event-based Systems

作者： Suad Sejdovic FZI Research Center for Information Technology Karlsruhe Germany

ISBN: (纸本)9781450340212

Within the last few years, event processing has been gaining a lot of attention as it represents a powerful possibility to establish a real-time monitoring and situation detection. the real-time detection of situations allows a timely reaction and helps to reduce the damage made by harmful situations or increase the benefit of opportunities. the so called situation of interest (SOI) is described as a pattern, which represents a combination of events describing their temporal causality. Instead of just detecting SOIs, we search for possibilities to predict them. As patterns are the basis for reactive event processing, we also want to exploit them for proactive event processing. therefore, we immerse into the world of pattern management to gain a better understanding of patterns, their structure and expressiveness.

关键词： pattern management proactive event-driven computing complex event processing

来源：评论

学校读者我要写书评

暂无评论

3D Warehousing: Enabling Intelligent Warehousing Visualization Based on *** 10th

3D Warehousing: Enabling Intelligent Warehousing Visualizati...

引用

10th international conference on Signal and Information Processing, Network and Computers, ICSINC 2022

作者： Shan, Wenqian Wan, Wenbo Chen, Anyi Ren, Leichao Sun, Jiande Fang, Meng Zhan, Yantong School of Information Science and Engineering Shandong Normal University Jinan China Shandong Port Land-Sea International Logistics Group Co. Ltd. Jinan China Shandong Luruan Digital Technology Co. Ltd. Smart Energy Branch Jinan China

ISBN: (纸本)9789811999673

In the industry 4.0 era, the 3D visualization and real-time automatic monitoring of intelligent warehousing supported are becoming increasingly important. In intelligent warehousing system, 2D chart mode is key component for visualization design. However, the requirements of 3D panoramic scene for warehousing data and the gap between interactive operation and data visualization challenge the development of visual warehousing. In order to solve the problems of untimely information sharing in the warehousing system, the inability to clearly and intuitively display the stockpiling of goods, and the difficulty of overall management and control of the warehouse. In this paper, we present a ***-based warehousing system which can enable 3D warehousing visualization from the perspective of intelligent management. the proposed system includes five key components: warehousing data visualization, device data visualization, security monitoring visualization, visualization of roaming in the warehouse area, and visualization of inventory increases and decreases. Finally, the core technologies are identified to provide a visual depiction for warehousing function, particularly in intelligent warehousing visualization applications. © 2023, the Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Visualization

来源：评论

学校读者我要写书评

暂无评论

PROPEL: Probabilistic Parametric Regression Loss for Convolutional Neural Networks

PROPEL: Probabilistic Parametric Regression Loss for Convolu...

引用

international conference on pattern recognition

作者： Muhammad Asad Rilwan Basaru S M Masudur Rahman Al Arif Greg Slabaugh Imagination Technologies Kings Langley UK Onaria Technologies London UK ASML Veldhoven Netherlands City University of London London UK

ISBN: (纸本)9781728188089;9781728188096

In recent years, Convolutional Neural Networks (CNNs) have enabled significant advancements to the state-of-the-art in computer vision. For classification tasks, CNNs have widely employed probabilistic output and have shown the significance of providing additional confidence for predictions. However, such probabilistic methodologies are not widely applicable for addressing regression problems using CNNs, as regression involves learning unconstrained continuous and, in many cases, multi-variate target variables. We propose a PRObabilistic Parametric rEgression Loss (PROPEL) that facilitates CNNs to learn parameters of probability distributions for addressing probabilistic regression problems. PROPEL is fully differentiable and, hence, can be easily incorporated for end - to-end training of existing CNN regression architectures using existing optimization algorithms. the proposed method is flexible as it enables learning complex unconstrained probabilities while being generalizable to higher dimensional multi-variate regression problems. We utilize a PROPEL-based CNN to address the problem of learning hand and head orientation from uncalibrated color images. Our experimental validation and comparison with existing CNN regression loss functions show that PROPEL improves the accuracy of a CNN by enabling probabilistic regression, while significantly reducing required model parameters by 10×, resulting in improved generalization as compared to the existing state-of-the-art.

关键词： Training Computer architecture Propulsion Predictive models Probabilistic logic Probability distribution pattern recognition

来源：评论

学校读者我要写书评

暂无评论

Towards Robust Learning with Different Label Noise Distributions

Towards Robust Learning with Different Label Noise Distribut...

引用

international conference on pattern recognition

作者： Diego Ortego Eric Arazo Paul Albert Noel E. O'Connor Kevin McGuinness Insight Centre for Data Analytics Dublin City University (DCU)

Noisy labels are an unavoidable consequence of labeling processes and detecting them is an important step towards preventing performance degradations in Convolutional Neural Networks. Discarding noisy labels avoids a harmful memorization, while the associated image content can still be exploited in a semi-supervised learning (SSL) setup. Clean samples are usually identified using the small loss trick, i.e. they exhibit a low loss. However, we show that different noise distributions make the application of this trick less straightforward and propose to continuously relabel all images to reveal a discriminative loss against multiple distributions. SSL is then applied twice, once to improve the clean-noisy detection and again for training the final model. We design an experimental setup based on ImageNet32/64 for better understanding the consequences of representation learning with differing label noise distributions and find that non-uniform out-of-distribution noise better resembles real-world noise and that in most cases intermediate features are not affected by label noise corruption. Experiments in CIFAR-10/100, ImageNet32/64 and WebVision (real-world noise) demonstrate that the proposed label noise Distribution Robust Pseudo-Labeling (DRPL) approach gives substantial improvements over recent state-of-the-art. Code is available at https://***/JJ0PV.

关键词： Training Degradation Semisupervised learning Robustness pattern recognition Noise measurement Labeling

来源：评论

学校读者我要写书评

暂无评论

A Deep Learning-Based Method for Identification of Bacteriophage-Host Interaction

引用

IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS 2021年第5期18卷 1801-1810页

作者： Li, Menglu Wang, Yanan Li, Fuyi Zhao, Yun Liu, Mengya Zhang, Sijia Bin, Yannan Smith, A. Ian Webb, Geoffrey I. Li, Jian Song, Jiangning Xia, Junfeng Anhui Univ Key Lab Intelligent Comp & Signal Proc Minist Educ Inst Phys Sci & Informat Technol Hefei 230601 Anhui Peoples R China Anhui Univ Sch Comp Sci & Technol Hefei 230601 Anhui Peoples R China Monash Univ Biomed Discovery Inst Melbourne Vic 3800 Australia Monash Univ Dept Biochem & Mol Biol Melbourne Vic 3800 Australia Monash Univ Fac Informat Technol Monash Ctr Data Sci Melbourne Vic 3800 Australia Monash Univ Ctr Impact AMR Dept Microbiol Melbourne Vic 3800 Australia Monash Univ Dept Biochem & Mol Biol Monash Ctr Data Sci Fac Informat Technol Melbourne Vic 3800 Australia Monash Univ Ctr Impact AMR Melbourne Vic 3800 Australia

Multi-drug resistance (MDR) has become one of the greatest threats to human health worldwide, and novel treatment methods of infections caused by MDR bacteria are urgently needed. Phage therapy is a promising alternative to solve this problem, to which the key is correctly matching target pathogenic bacteria with the corresponding therapeutic phage. Deep learning is powerful for mining complex patterns to generate accurate predictions. In this study, we develop PredPHI (Predicting Phage-Host Interactions), a deep learning-based tool capable of predicting the host of phages from sequence data. We collect >3000 phage-host pairs along with their protein sequences from PhagesDB and GenBank databases and extract a set of features. then we select high-quality negative samples based on the K-Means clustering method and construct a balanced training set. Finally, we employ a deep convolutional neural network to build the predictive model. the results indicate that PredPHI can achieve a predictive performance of 81 percent in terms of the area under the receiver operating characteristic curve on the test set, and the clustering-based method is significantly more robust than that based on randomly selecting negative samples. these results highlight that PredPHI is a useful and accurate tool for identifying phage-host interactions from sequence data.

关键词： Stakeholders Job shop scheduling Routing Schedules Medical services Delays Phage-host interaction bioinformatics sequence analysis deep learning pattern recognition multi-drug resistance

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：