检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Calafà, Matteo Hovad, Emil Engsig-Karup, Allan P. Andriollo, Tito Dept. of Mechanical and Production Engineering Aarhus University Aarhus Denmark AI Lab Alexandra Instituttet Copenhagen Denmark Dept. of Applied Mathematics and Computer Science Technical University of Denmark Kongens Lyngby Denmark

We propose physics-informed holomorphic neural networks (PIHNNs) as a method to solve boundary value problems where the solution can be represented via holomorphic functions. Specifically, we consider the case of plane linear elasticity and, by leveraging the Kolosov-Muskhelishvili representation of the solution in terms of holomorphic potentials, we train a complex-valued neural network to fulfill stress and displacement boundary conditions while automatically satisfying the governing equations. This is achieved by designing the network to return only approximations that inherently satisfy the Cauchy-Riemann conditions through specific choices of layers and activation functions. To ensure generality, we provide a universal approximation theorem guaranteeing that, under basic assumptions, the proposed holomorphic neural networks can approximate any holomorphic function. Furthermore, we suggest a new tailored weight initialization technique to mitigate the issue of vanishing/exploding gradients. Compared to the standard PINN approach, noteworthy benefits of the proposed method for the linear elasticity problem include a more efficient training, as evaluations are needed solely on the boundary of the domain, lower memory requirements, due to the reduced number of training points, and C∞ regularity of the learned solution. Several benchmark examples are used to verify the correctness of the obtained PIHNN approximations, the substantial benefits over traditional PINNs, and the possibility to deal with non-trivial, multiply-connected geometries via a domain-decomposition strategy. Copyright © 2024, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Comparative analysis of OFDM in AWGN & Local Acoustic Channel

Comparative analysis of OFDM in AWGN & Local Acoustic Channe...

引用

Humanitarian Technology Conference (KHI-HTC), IEEE Karachi Section

作者： Areeb Ahmed Aish Siddqui Rehan Khan Ghulam Nabi Murtaza Ali Dept. Electrical and Computer Systems Engineering Mohammad Ali Jinnah University Karachi Pakistan Centers of Excellence in Science and Applied Technologies (CESAT) Islamabad Pakistan

In this research study, a comparative analysis is presented that analyze the Orthogonal Frequency Division Multiplexing (OFDM) in two distinct communication systems Additive White Gaussian Noise (AWGN) and a Local Acoustic Channel. This study incorporates insights from the BELLHOP model to enhance understanding of sound propagation in dept. of water and guide the adaptation of OFDM for optimized communication. From simulated experiments, evaluate the performance of OFDM is assessed in the presence of local acoustic channel and AWGN, the important parameters like signal-to-noise ratio (SNR) and bit error rate (BER). The system of BELLHOP provides valuable context in the comparative analysis, highlighting the result of environmental factors on the performance of OFDM. The study not only contributes to the ongoing efforts in advancing underwater communication technique but also emphasizes the significance of tailored approach’s in different underwater scenarios, ultimately having the way for more robust and efficient communication systems beneath the waves. Through a comparative analysis between AWGN and a local acoustic channel, results show that the system achieves a BER of 10 −3 at SNR of 10 in local acoustic channel, while in the AWGN, a similar BER is achieved at SNR of 16.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Exact algebraic blind source separation using side information 28

Exact algebraic blind source separation using side informati...

引用

28th European Signal Processing Conference, EUSIPCO 2020

作者： Weiss, Amir Yeredor, Arie Dept. of Computer Science and Applied Mathematics Weizmann Institute of Science Israel School of Electrical Engineering Tel-Aviv University Israel

ISBN: (纸本)9789082797053

Classical Blind Source Separation (BSS) methods rarely attain exact separation, even under noiseless conditions. In addition, they often rely on particular structural or statistical assumptions (e.g., mutual independence) regarding the sources. In this work we consider a (realistic) "twist" in the classical linear BSS plot, which, quite surprisingly, not only enables perfect separation (under noiseless conditions), but is also free of any assumptions (except for regularity assumptions) regarding the sources or the mixing matrix. In particular, we consider the standard linear mixture model, augmented by a single ancillary, unknown linear mixture of some known linear transformations of the sources. We derive a closed-form expression for an exact algebraic solution, free of any statistical considerations whatsoever, attaining perfect separation in the noiseless case. In addition, we propose a well-behaved solution for the same model in the presence of noise or other measurement inaccuracies. Our derivations are corroborated by several simulation results. © 2021 European Signal Processing Conference, EUSIPCO. All rights reserved.

关键词： Blind source separation

来源：评论

学校读者我要写书评

暂无评论

Wireless Security and IoT Device Identification using RF Fingerprinting and Deep Learning

Wireless Security and IoT Device Identification using RF Fin...

引用

IEEE Conference on Vehicular Technology (VTC)

作者： Nordine Quadar Abdellah Chehri Benoit Debaque Dept. Math. and Computer Science Royal Military College of Canada Kingston Canada Research and Technology Center Thales-Canada Quebec Canada

ISBN: (数字)9798331517786

ISBN: (纸本)9798331517793

Enhancing the security of wireless networks involves implementing a user authentication method when the fingerprint of a network device is unknown or considered a potential threat. This technique is known as radio frequency (RF) fingerprinting. This paper presents a novel method for RF fingerprinting of Internet of Things (IoT) devices, addressing the challenges of the radio frequency spectrum. The proposed architecture integrates a feature generator module that transforms time-series I/Q samples into a multi-dimensional matrix and a deep learning module inspired by the ResNet-50-1D model. We assess the effectiveness of our approach by analyzing a real-world dataset of BT emissions obtained from 10 commercial IoT devices in two challenging indoor environments. The datasets, made publicly accessible on IEEE Dataport, were gathered using a USRP X300 software-defined radio (SDR) in both line-of-sight (LoS) and rich multipath propagation scenarios. Our method showcases excellent results in the TTS scenario and shows promise in the challenging TTD scenario, considering the complex nature of frequency hopping. The evaluation results emphasize the significance of evaluating RF fingerprinting models in various scenarios and offer valuable insights into the strengths and limitations of our approach in handling radio frequency waveforms.

关键词： Radio frequency Deep learning Vehicular and wireless technologies Wireless networks Transfer learning Transforms Fingerprint recognition Internet of Things Security Signal to noise ratio

来源：评论

学校读者我要写书评

暂无评论

A Cluster-based Analysis for Targeting Potential Customers in a Real-world Marketing System

A Cluster-based Analysis for Targeting Potential Customers i...

引用

IEEE International Conference on Intelligent computer Communication and Processing (ICCP)

作者： Sheikh Sharfuddin Mim Doina Logofatu Dept. of Computer Science Frankfurt University of Applied Sciences Frankfurt Germany

ISBN: (纸本)9781665464383

In today’s fast-paced, customer-focused marketing world, customer service management is key to revenue growth and profitability. Customer behavior knowledge can help marketing managers reevaluate existing customer tactics and plan to improve and increase effective strategies. Maintaining a productive relationship with business customers is crucial because business transactions require more decision-making and professional buying effort than consumer purchases. Most customer segmentation methods based on customer value fail to account for time and value changes. Today’s business is based on new concepts because so many customers are unsure of what to buy. Businesses can’t identify target customers. Machine learning algorithms detect hidden data patterns to make better decisions. Customer segmentation groups customers by gender, age, interests, and spending habits. Customers with unique needs are segmented by companies. Companies want to know their customers. Their goals must be clear and individualized. By analyzing data, businesses can better understand customer preferences and identify profitable segments. This helps them strategize their marketing while minimizing investment risk. Customer segmentation depends on many factors. Demographic, geographic, economic, and behavioural data help the company approach different sectors. Customer segmentation uses clustering to determine which consumer segment to target. This paper demonstrates machine learning-based customer segmentation. This is the unsupervised clustering problem, and we apply four of the most popular clustering algorithms: K-Means, Affinity Propagation, DBSCAN, and Hierarchical Clustering. We find the optimal number of clusters from a shopping mall dataset using the Elbow method and Silhouette score with the key factors like age, spending score, and annual income of customers. It helps shopping malls improve business by finding customer behavioural patterns. We compare the results using the Silhouette and

关键词： Machine learning algorithms Profitability Biological system modeling Customer services Decision making Clustering algorithms Companies

来源：评论

学校读者我要写书评

暂无评论

Market Study and User-Friendly Enablement of 4G/5G LAN-Like Connectivity for SOHO Customers

Market Study and User-Friendly Enablement of 4G/5G LAN-Like ...

引用

International Conference on Smart Communications and Networking (SmartNets)

作者： Dirk Henrici Andreas Boose Dept. of Computer Science and Mathematics University of Applied Sciences Munich Germany B2B Technology Solutions Telefonica Germany Munich Germany

ISBN: (数字)9798350385328

ISBN: (纸本)9798350385335

Seamlessly connecting mobile devices to existing non-public, local networks like home networks or company networks appears to be an interesting opportunity for monetizing cellular 4G/5G networks by mobile operators. We present the results of a market study with SOHO (Small Office / Home Office) customers in Germany. Based on the findings, we developed a prototype for a plug&play device for virtually connecting mobile devices to an existing home network. This user-friendly approach was tested in a friendly user trial with promising results. Large-scale productization requires network equipment vendors to provide support in mobile packet core in a scalable and cost-effective manner.

关键词： Home automation Prototypes Companies Mobile handsets Virtual private networks

来源：评论

学校读者我要写书评

暂无评论

An Efficient Framework for classifying Cancer diseases using Ensemble machine learning over Cancer Gene Expression and Sequence Based Protein Interactions.

An Efficient Framework for classifying Cancer diseases using...

引用

Innovation in Technology (INOCON), IEEE International Conference for

作者： Prabhuraj Metipatil P Bhuvaneshwari Syed Muzamil Basha Siddanagouda S Patil School Of CSE REVA University Bengaluru India Dept. Of Computer Science & Applied Mathematics UAS Bengaluru Bengaluru India

In recent years, a significant number of deaths worldwide have been due to cancer diseases. Analysis of Microarray gene expressions and protein interaction data facilitates early cancer identification. The accurate prediction of information for thousands of genes is made possible by using DNA microarray technology. Protein-Protein Interactions (PPIs) are the crucial protein activities involved in the cell cycle that replicates the DNA and cellular signaling. Determining whether a pair of proteins interacts is crucial for diagnosing an illness in molecular biology is therefore important. In existing machine learning classifiers have two-class problem that is limited and only be used to solve binary class problems, additionally, they can be prone to overfitting, as the classification framework may also become too specialized to the training data and not generalized to the varied data. To overcome this problem, this paper proposes an ensemble machine learning technique; ensembling combines the strengths of both classifiers that allow more robust and accurate framework. The better combination of both Support Vector machine and Naïve Bayes ensemble provides better performance in terms of various performance parameters. The proposed SVM-NB Ensemble classifier outperforms the existing classifiers by 15-20% over various performance parameters like classification accuracy, time taken for classification, precision, recall, and F-measure. The results were drawn by comparing the proposed ensemble (SVM+NB) classifier with the existing most applied classifiers like Logistic Regression (LR), Support Vector Machine and Naive Bayes techniques.

关键词： Support vector machines Proteins Technological innovation DNA Training data Support vector machine classification Machine learning

来源：评论

学校读者我要写书评

暂无评论

Educational Matlab Tool for Evaluating Liver Tumors Using Ultrasound 16

Educational Matlab Tool for Evaluating Liver Tumors Using Ul...

引用

16th International Symposium on Electronics and Telecommunications, ISETC 2024

作者： Veres, Denisa Alexandra Bonciog, Dumitru Daniel Olei, Damian Cirnatu, Daniela Goje, Daniel Iacob Comaniciu, David Iovan, Liliana Matiu Lascu, Mihaela Ruxandra Ordodi, Valentin Laurentiu University Politehnica Timişoara Dept. of Computer and Information Technology Timişoara Romania University Politehnica Timişoara Dept. of Measurements and Optical Electronics Timişoara Romania National Institute of Public Health Timisoara Regional Centre Timişoara Romania 'Victor Babes' University of Medicine and Phanmacy Timişoara Dept.of Internal medicine I. Timişoara Romania University of Medicine and Phanmacy Timişoara Dept. of Biochemistry.'Victor Babes' Timişoara Romania University Politehnica Timişoara Engineering of Organic and Natural Compounds Dept. of Applied Chemistry Timisoara Romania 'Victor Babes' University of Medicine and Phanmacy Timişoara Dept. of Functional Science Timisoara Romania

ISBN: (纸本)9798350390865

This paper aims to develop a Matlab application for computer-assisted learning in the evaluation of liver tumors using ultrasound. Diagnosing liver tumors detected during standard ultrasound exams is challenging for doctors, often requiring detailed investigations for accurate diagnosis and effective treatment. computerized analysis of ultrasound images can significantly aid doctors in interpreting these images and formulating accurate diagnoses. The Matlab programming environment provides various tools for processing biomedical images. In this study, Histogram and Entropy functions were utilized to accurately differentiate between malignant and benign liver tumors. Additionally, Matlab App Designer was used to create a user-friendly interface for doctors. This application is also beneficial for students, offering a valuable tool for self-testing and skill development in clinical ultrasound courses and practical sessions. The proposed Matlab application employs classic image processing tools and serves as useful software for students learning liver ultrasound techniques. © 2024 IEEE.

关键词： MATLAB

来源：评论

学校读者我要写书评

暂无评论

Prediction Of Typhoid Using Machine Learning and ANN Prior To Clinical Test

Prediction Of Typhoid Using Machine Learning and ANN Prior T...

引用

International Conference on computer Communication and Informatics (ICCCI)

作者： Md. Atik Bhuiyan Sharaf Shahariare Rad Fatema Tuj Johora Abdullah Islam Md Ismail Hossain Aliza Ahmed Khan Dept. of Computer Science and Engineering Daffodil International University Dhaka Bangladesh Dept. of Communication and Environment Rhine-Waal University of Applied Sciences Kamp-Lintfort Germany

One of the most prevalent illnesses, typhoid causes a large number of fatalities each year, primarily in Africa. A quick and accurate diagnosis is essential in the medical sector. Self-medication, delayed diagnosis, a lack of medical expertise, and inadequate healthcare facilities all contribute to the high incidence of typhoid fever mortality. Machine learning as well as deep learning has worked wonders for extrapolative analysis in the health industry, and as a result, more health industries are utilizing machine learning techniques. This is the earliest evaluation where a typhoid fever prediction model is being developed which predicts prior to a clinical trial. In this paper, deep learning and machine learning have been employed to develop the model. Ten algorithms have been utilized here, and the XGBoost classifier is the best performer with 97.87% accuracy.

关键词： Deep learning Industries Machine learning algorithms Computational modeling Medical services Predictive models Prediction algorithms

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Application for Detection of Malaria

Deep Learning Application for Detection of Malaria

引用

IEEE International Conference on Electro-Information Technology

作者： Md Saifur Rahman Nafiz Rifat Mostofa Ahsan Sabrina Islam Md. Chowdhury Rahul Gomes Dept. of Math Science and Technology University of Minnesota Crookston Crookston MN USA Dept. of Computer Science North Dakota State University Fargo ND USA Dept. of Biological Sciences Florida Atlantic University Boca Raton FL USA Dept. of Computer Science East Stroudsburg University East Stroudsburg PA USA Dept. of Computer Science University of Wisconsin-Eau Claire Eau Claire WI USA

Malaria continues to be a significant burden on global health with 247 million clinical episodes and 619,000 deaths. Along with biomedical science, technology, and informatics have begun participating in the quest against malaria. Microscopy techniques are frequently used to detect malaria parasites in infected red blood cells. Giemsa stain has been used to stain blood parasites for over a century. The stain is applied after fixing blood smears in methyl alcohol for 25 to 30 minutes [1]. When stained slides are examined under a microscope, the parasites are easily discernible based on morphology and color. We observed that automating the detection of these slides using deep learning is possible with high accuracy. A comparison between deep learning models reveals ResNets provide better performance.

关键词：

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：