检索结果-内蒙古大学图书馆

2024 ASEE Annual Conference and Exposition

作者： Manasil, Kristi Heileman, Gregory L. Hemaraja, Rohit Sharma, Bhavya Slim, Ahmad Sharifi, Melika Sharifi, Roxana Yusuf, Husain Al The School of Information The University of Arizona United States Department of Electrical & Computer Engineering The University of Arizona United States Department of Computer Science The University of Arizona United States

engineering programs are typically among the most tightly prescribed programs within the academic landscape on any university campus. The strict nature of these programs often results in students taking more credits than stipulated, thereby leaving them struggling to graduate in a timely manner. The ability to identify potential blockers or challenges in an engineering program's curriculum is vital to student success and the promotion of on-time graduation. This paper provides a comprehensive examination of patterns and trends observed by a newly developed cohort tracking analytics platform. This platform provides analyses over a cohort of students which uncovers insights that are not easily identified when only looking at data at the individual student level. The analysis pinpoints courses that many students within the cohort have taken that are not applicable to the degree, along with the reasons why these courses are not applicable. It also identifies trends in courses that must be repeated by a significant portion of the cohort. It examines the courses constituting a program's degree requirements that have yielded both the best and worst grade value outcomes. In addition, an exploration of a cohort's efficiency of credit hour production is provided for both the home institution units and transfer units, which shows where credits are not aligning with degree requirements and therefore not counting towards degree completion. Finally, a comparative analysis of programs within the engineering field is performed as well as a comparison of engineering programs to non-engineering programs. This type of analysis demonstrates the differences in how students in engineering programs make progress towards their degree completion. The statistical analyses furnished by this platform provide administrators with an evidence-based foundation to support programmatic modifications and enhancements. This allows administrators to depart from the past practices of having to rely on a

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Efficient Framework with Sparse Acquisition in CMOS Image Sensors for Low-Power Edge Devices 10

Efficient Framework with Sparse Acquisition in CMOS Image Se...

引用

10th IEEE International Symposium on Smart Electronic Systems, iSES 2024

作者： Kisku, Wilfred Kaur, Amandeep Mishra, Deepak Indian Institute of Technology Department of Electrical Engineering Jodhpur India Indian Institute of Technology Department of Computer Science and Engineering Jodhpur India

ISBN: (纸本)9798331533229

Memory bandwidth and power consumption is of utmost importance in the design of low power edge devices. This makes it essential to conserve power both at the sensor node and the computational unit. Our paper proposes a system design methodology to save power by acquiring compressed data at the CMOS Image Sensor (CIS) and making the Convolutional Neural Network (CNN) models lightweight for edge applications. A detailed study highlights the effect of sensor directed sparsity at the CIS, and its effect on CNN models such as VGG-16, MobileNet, ResNet50, EfficientNet, and ResNext models. It provides empirical evidence that the performance is mostly maintained even after pruning and quantization of considered models. The imagenette dataset is used to curate the data with the proposed acquisition pipeline that introduces data sparsity, and the experimental results provide evidence that the proposed pipeline performs at par with conventional CIS acquisition mode. For example, the analysis of VGG-16 model for a sparsity threshold level of 0.2 shows compression of × 7.6, × 16, and × 61 using only pruning (P), only quantization (Q), and with both pruning+quantization (P+Q), respectively. It is also observed that the proposed acquisition pipeline can achieve significant parameter reduction through pruning for model compression, comparable to convectional pipeline. © 2024 IEEE.

关键词： CMOS integrated circuits

来源：评论

学校读者我要写书评

暂无评论

Application of 3D Magnetic Resonance Imaging in the Characterization of Brain Tumors 2

Application of 3D Magnetic Resonance Imaging in the Characte...

引用

2nd International Conference on Disruptive Technologies, ICDT 2024

作者： Singh, Samender Kishore, Lalit Badhoutiya, Arti Sagar, Pramod Kumar Saraswat, Laxmi Ajay Kumar Garg Engineering College Department of Computer Science & Engineering UP Ghaziabad India SRMIST Delhi Ncr Department of Computer Application Modinagar UP Ghaziabad India Gla University Department of Electrical Engineering UP Mathura India Raj Kumar Goel Institute of Technology Srmist Department of Computer Science & Engineering UP Ghaziabad India Abes Engineering College Department of Computer Science UP Ghaziabad India

ISBN: (纸本)9798350371055

MRI has revolutionized the analysis and remedy of disorder by using imparting precise and accurate images of soft tissue. This technology has been similarly superior by means of the improvement of 3-D MRI, which permits improved detection and characterization of tumors and improves accuracy of diagnosis and remedy. 3D MRI has been found to be in particular beneficial within the characterization of brain tumors as it offers a greater detailed assessment of tumor size, shape and composition, in comparison to traditional imaging techniques including X-ray or CT. This information can help in identifying the nature of the tumor that may impact its remedy and analysis. Furthermore, 3D MRI is useful for figuring out the chance of malignant invasion, which can provide important information for affected person control. Additionally, the advanced decision of 3-D MRI permits better delineation of areas of healthy and diseased tissue that can assist clinicians in making plans complex surgeries. (3-D) MRI has emerged as an increasingly more critical tool inside the characterization of brain tumors. The three-D imaging affords distinct anatomical visualization of the tumor and it's courting to surrounding systems, allowing for extra effective remedy making plans. Furthermore, the potential to visualize tumor tissue parameters together with edema, necrosis, and perfusion may be used to decide whether a tumor is malignant or no longer. 3D MRI has additionally enabled the evaluation of tumor vascularization, which serves as a vital prognostic indicator. The statistics received from 3-D imaging can then be used to be expecting the recurrence of a tumor following treatment, supporting to manual decisions on comply with-up remedies. The reduction in test time, decrease value, and better resolution of 3-D MRI in contrast to standard x-ray imaging also makes the technology more and more attractive for brain tumor management. © 2024 IEEE.

关键词： Attractive Characterization Superior Traditional

来源：评论

学校读者我要写书评

暂无评论

Providing and evaluating a comprehensive model for detecting fraudulent electronic payment card transactions with a two-level filter based on flow processing in big data

引用

International Journal of Information Technology (Singapore) 2023年第8期15卷 4161-4166页

作者： Banirostam, Hamid Banirostam, Touraj Pedram, Mir Mohsen Rahmani, Amir Masoud Department of Computer Engineering Central Tehran Branch Islamic Azad University Tehran Iran Department of Electrical and Computer Engineering Faculty of Engineering Kharazmi University Tehran Iran Department of Computer Engineering Science and Research Branch Islamic Azad University Tehran Iran

Previous research on fraud detection modeling is often based on a single algorithm, optimizing categories and clusters to find fraudulent patterns that they have provided unsupervised or supervised methods alone and within the framework of Hadoop. The proposed model, a model based on big data analysis extracts important features of user behavior patterns such as time, device type, values, and type of transaction, and their behavioral modeling. By creating different profiles for users, threshold values will be set for each of them. The proposed model for real-time fraud detection of electronic cash payment cards includes two fast and explicit filters. Fast filtering is the combination of the Hidden Markov Model of the first order and Self Organizing Maps (SOM) for fast transaction processing and the Baum-Welch algorithm to find the local Maximum Likelihood. Explicit filter model training is a combination of multilayer Perceptron Neural Network algorithms and Logistic Regression to create a cardholder profile and measure the amount of new transaction deviation from the created profile with the reduction mapping approach or parallel execution of the model. Also, transactions performed for aggregation filters and prediction of the output results of the Schaffer Demister function with the parameter θ in order to detect the degree of deviation of the transaction behavior from the normal state according to the profile of the customer and the maximum difference between two consecutive sequences of observations is used. Finally, the fraudulent or non-fraudulent label on the transaction will be applied and added to the relevant database for storage for later use. According to the model simulation results, the proposed Accuracy, Precision, Recall, and F1-Score criteria are 0.999, 0.9834, 0.7906, and 0.9214, respectively. Simulation results show that the proposed model will perform better in each of the compared criteria against each of the single algorithms. © 2023, The Author

关键词： Accuracy Electronic card Fraud Supervised algorithm Transaction Unsupervised

来源：评论

学校读者我要写书评

暂无评论

Emotion Recognition in Noisy Environments: Performance Analysis using Convolutional Neural Networks 8

Emotion Recognition in Noisy Environments: Performance Analy...

引用

8th International Conference of Yearly Reports on Informatics, Mathematics, and engineering, ICYRIME 2023

作者： Borzì, Maria Grazia Beritelli, Ludovica Puglisi, Valerio Francesco Avanzato, Roberta Beritelli, Francesco Department of Electrical Electronic and Computer Engineering University of Catania Catania Italy Department of Computer Science University of Catania Catania Italy

Recognition of emotional state through the sound of the voice is a crucial element in human interactions. This process, known as emotional prosody, allows an individual’s emotions to be interpreted without the need to see his or her face or body language. The ability to identify emotions through voice is critical in areas such as psychology, health care, marketing and human-computer interaction technology. Moreover, in artificial intelligence systems and voice interfaces, the analysis of emotional prosody can improve the user experience and make interactions more personalized and intuitive. This study explores the robustness of a neural network in recognizing emotions within speech signals, considering the presence of environmental noise at different SNR levels. The study involves using public datasets containing different classes of emotions, training a CNN network to classify these emotional states, and then testing to evaluate the obtained performance. The results show that emotion recognition in speech is more accurate when the recording is clean and free of other noises. However, misclassification depends on the type of noise present and the type of emotion: indeed, it turns out that some emotions are more frequently misclassified than others. In fact, it is observed that accuracy is higher when using RAW data rather than MFCCs, with an accuracy of 75% compared to 60%. In addition, adding noise to the recordings results in a decrease in model performance. © 2023 Copyright for this paper by its authors.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Compliance Design Options for Offline CBDCs: Balancing Privacy and AML/CFT 6

Compliance Design Options for Offline CBDCs: Balancing Priva...

引用

6th IEEE International Conference on Blockchain and Cryptocurrency, ICBC 2024

作者： Michalopoulos, Panagiotis Olowookere, Odunayo Pocher, Nadia Sedlmeir, Johannes Veneris, Andreas Puri, Poonam York University Osgoode Hall Law School Canada Luxembourg University of Toronto Department of Electrical and Computer Engineering Canada University of Toronto Department of Computer Science Canada

ISBN: (纸本)9798350316742

Many central banks are researching and piloting digital versions of fiat money, specifically retail Central Bank Digital Currencies (CBDCs). Core to these systems' design is the ability to perform transactions even without network connectivity. Due to the lack of direct involvement of third parties in these offline transfers, various regulatory requirements that are key in the financial space need to be accommodated. This paper deploys a compliance-by-design approach to evaluate technologies that can balance privacy with anti-money laundering and counterterrorism financing (AML/CFT) measures. It classifies privacy design options and corresponding technical building blocks for offline CBDCs, along with their impact on AML/CFT measures, and outlines commonalities and differences between offline and online solutions. As such, it provides a conceptual framework for further techno-legal assessments and implementations. © 2024 IEEE.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

Dual Operating Modes of In-Context Learning 41

Dual Operating Modes of In-Context Learning

引用

41st International Conference on Machine Learning, ICML 2024

作者： Lin, Ziqian Lee, Kangwook Department of Computer Science University of Wisconsin-Madison MadisonWI United States Department of Electrical & Computer Engineering University of Wisconsin-Madison MadisonWI United States

In-context learning (ICL) exhibits dual operating modes: task learning, i.e. acquiring a new skill from in-context samples, and task retrieval, i.e., locating and activating a relevant pretrained skill. Recent theoretical work proposes various mathematical models to analyze ICL, but they cannot fully explain the duality. In this work, we analyze a generalized probabilistic model for pretraining data, obtaining a quantitative understanding of the two operating modes of ICL. Leveraging our analysis, we provide the first explanation of an unexplained phenomenon observed with real-world large language models (LLMs). Under some settings, the ICL risk initially increases and then decreases with more in-context examples. Our analysis offers a plausible explanation for this "early ascent" phenomenon: a limited number of in-context samples may lead to the retrieval of an incorrect skill, thereby increasing the risk, which will eventually diminish as task learning takes effect with more in-context samples. We also analyze ICL with biased labels, e.g., zero-shot ICL, where in-context examples are assigned random labels, and predict the bounded efficacy of such approaches. We corroborate our analysis and predictions with extensive experiments with Transformers and LLMs. The code is available at: https://***/UW-Madison-Lee-Lab/Dual_Operating_Modes_of_ICL. Copyright 2024 by the author(s)

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Wireless Semantic Communication Based on Probability Distribution: An Initial Work

Wireless Semantic Communication Based on Probability Distrib...

引用

2025 IEEE Wireless Communications and Networking Conference, WCNC 2025

作者： Luo, Qingxiang Guo, Yashuang Zheng, Aoran Ni, Zhitong Yu, F. Richard Leung, Victor C.M. School of Computer Science and Technology Beijing Jiaotong University China Big Data Technologies Centre University of Technology Sydney Australia Carleton University Department of Systems and Computer Engineering Canada The University of British Columbia Department of Electrical and Computer Engineering Canada

ISBN: (纸本)9798350368369

In the paper, we consider the general semantic transmission in wireless networks based on probability distribution. Firstly, we extract a multidimensional semantic probability distribution function, independent of any a specific wireless channel model, by using the variational inference technique. Secondly, we propose a new semantic similarity metric for measuring the difference between the received semantics and the expected semantics based on Kullback-Leibler divergence. Then, we formulate the semantic transmission problem as an optimization problem of transmission symbol adjustment with the aim to maximize the semantic similarity. Finally, we develop an optimal semantic transformation and transmission (STT) algorithm to obtain the optimal transmission symbol adjustment decision. This decision makes the closed-form expression of semantic transmission symbol available, which can realize lossless semantic transmission with energy constraint. Simulation results verify the effectiveness and robustness of the proposed STT algorithm. © 2025 IEEE.

关键词： Normal distribution

来源：评论

学校读者我要写书评

暂无评论

Score-Based Unsupervised Anomaly Detection using Graph Clustering and the Activity and Event Network Model 7

Score-Based Unsupervised Anomaly Detection using Graph Clust...

引用

7th Cyber Security in Networking Conference, CSNet 2023

作者： Bagha, Amir Mohammadi Woungang, Isaac Traore, Issa Toronto Metropolitan University Department of Computer Science Toronto Canada University of Victoria Department of Electrical and Computer Engineering Victoria Canada

ISBN: (纸本)9798350342871

Developing a real-time unsupervised anomaly detection system that is able to detect a broad range of attacks, including insider threats, distributed denial-of-service attacks, and Advanced Persistent Threats, among others, in real-time, is still a challenge. In this paper, a new anomaly detection scheme is proposed, which is based on the Activity and Event Network (AEN) framework, a new knowledge graph model that allows capturing the dynamicity and uncertainty inherent in network activities while providing a foundation for expressing various threat detection approaches. The considered unsupervised learning-based approach takes advantage of node clustering in the network to establish a normal behaviour for each cluster and detect the anomalies accordingly. Our proposed scheme is evaluated using the CIC 2018 IDS dataset, yielding a false positive rate (FPR) of 0.83% and a detection rate (DR) of 79%. © 2023 IEEE.

关键词： Denial-of-service attack

来源：评论

学校读者我要写书评

暂无评论

Novel Face Mask Detection Technique using Machine Learning to control COVID’19 pandemic

引用

Materials Today: Proceedings 2023年 80卷 3714-3718页

作者： Gupta, Sandeep Sreenivasu, S.V.N. Chouhan, Kuldeep Shrivastava, Anurag Sahu, Bharti Manohar Potdar, Ravindra Department of Electrical Engineering JECRC University Jaipur India Department of Computer Science and Engineering Narasaraopeta Engineering College Andhra Pradesh Narasaraopet India Department of Computer Science and Engineering I.T.S Engineering College Greater Noida India Lakshmi Narain College of Technology and Science Indore India Department of Computer Science and Engineering Dr. D.Y. Patil Institute of Technology Pimpri Pune India Department of Electronics and Telecommunication Engineering Bhilai Institute of Technology Chhattisgarh Durg India

The COVID-19 pandemic has been scattering speedily around the world since 2019. Due to this pandemic, human life is becoming increasingly involutes and complex. Many people have died because of this virus. The lack of antiviral drugs is one of the reasons for the spreading of COVID-19 virus. This disease is spreading continuously and easily due to some common mistakes by people, like breathing, coughing and sneezing by infected persons. The main symptom is the normal flu. Therefore, in the present condition, the best precaution for this disease is the face mask, which covers both areas of mouth & nose. According to the government and the World Health Organization, everyone should wear a face mask in busy places like hospitals and marketplaces. In today's environment, it's difficult to tell if someone is wearing a mask or not, and physical inspection is impractical since it adds to labour costs. In this research, we present a mask detector that uses a machine learning facial categorization system to determine whether a person is wearing a mask or not, so that it may be connected to a CCTV system to verify that only persons wearing masks are allowed in. © 2021

关键词： COVID-19

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：