检索结果-内蒙古大学图书馆

International Conference on Advanced Cloud and Big Data (CBD)

作者： Jingjing Wan Ling Wang Wenjie Zhu Bolun Chen Department of Gastroenterology The Second People's Hospital of Huai'an Huaian China Faculty of Computer and Software Engineering Huaiyin Institute of Technology Huaian China

ISBN: (数字)9798331511074

ISBN: (纸本)9798331511081

Early detection of colorectal polyps is crucial for preventing colorectal cancer. Although endoscopy is the current standard diagnostic method, it still faces challenges in terms of accuracy, efficiency, and patient comfort. To address these issues, this paper proposes a colorectal polyp detection model named DM-Net. This model utilizes the Dual Feature Integration Block (DFIB) to integrate channel and spatial features, enhancing feature extraction efficiency. Additionally, the model incorporates the Multi-Layer Path Aggregation Network (MLPAN) to handle the multi-scale variations of polyps. Experimental results demon-strate that DM-Net significantly outperforms existing methods in terms of accuracy and efficiency, offering high clinical applica-bility.

关键词： YOLO Accuracy Endoscopes Colorectal cancer Big Data Feature extraction Real-time systems Standards Faces

来源：评论

学校读者我要写书评

暂无评论

Accuracy-Aware Low-Complexity Deep Learning Models for Automatic Modulation Recognition

Accuracy-Aware Low-Complexity Deep Learning Models for Autom...

引用

Computing, Internet of Things and Microwave Systems (ICCIMS), International Conference on

作者： Mobin Vaziri Shervin Vakili J.M. Pierre Langlois Department of Computer and Software Engineering Polytechnique Montréal Canada Institut national de la recherche scientifique (INRS) Montréal Canada

ISBN: (数字)9798350351736

ISBN: (纸本)9798350351743

This paper explores the design of Deep Learning (DL) models for Automatic Modulation Recognition (AMR) in wireless communications. The primary goal is to enhance the efficiency and hardware compatibility of convolutional neural networks (CNNs) for AMR through hyperparameter tuning and model compression. The paper first examines the effectiveness of applying quantization and pruning on the accuracy and compu-tational cost of two prominent CNN models from the literature. It then introduces a new CNN model that achieves superior accu-racy with lower computational complexity compared to previous work. The design flow integrates TensorFlow Lite for pruning and quantization, and NVIDIA TensorRT for benchmarking on GPUs specialized for machine learning computing. Experimental results show significant reductions in model size and computational complexity while maintaining accuracy, rendering the proposed DL models suitable for real-time applications on edge devices.

关键词： Deep learning Wireless communication Quantization (signal) Accuracy Computational modeling Modulation Rendering (computer graphics) Convolutional neural networks Computational complexity Tuning

来源：评论

学校读者我要写书评

暂无评论

VRGrip: Developing a Grip Strength Training Platform Integrating a Wireless E-Textile Forearm Band with an Adaptive 3D VR Game Environment

VRGrip: Developing a Grip Strength Training Platform Integra...

引用

2023 IEEE MIT Undergraduate Research Technology Conference, URTC 2023

作者： Chatragadda, Lohith Fletcher, Aiden Delmonico, Matthew J. Mankodiya, Kunal Solanki, Dhaval University of Rhode Island Interdisciplinary Neuroscience Program KingstonRI United States University of Rhode Island Department of Kinesiology KingstonRI United States University of Rhode Island Department of Electrical Computer and Biomedical Engineering KingstonRI United States

ISBN: (纸本)9798350308600

Neurodegenerative disease is a growing global problem. Many of these diseases such as Parkinson's disease can cause grip strength weakness. In this work, we focused on developing an e-Textile based EMG acquisition system interfaced with an adaptive VR game engine for detecting hand squeeze actions. We prototyped and performed preliminary tests on a grip detection system consisting of: 1) an e-Textile forearm band (E-band), 2) a portable data acquisition module (DAM), and 3) an adaptive 3D VR game engine. Digital health solutions such as our VRGrip system widen the scope of in-home rehabilitation and assessment, and we aimed to make the first step in creating a system for grip strength rehabilitation. To validate our concept, we performed preliminary tests. These tests revealed that the Bluetooth communication protocol between DAM and the game engine was reliable, that the adaptive game engine successfully tailors the game's difficulty level to the user's performance, and that the system was reliable at detecting hand squeezes (F=0.96). © 2023 IEEE.

关键词： Virtual reality

来源：评论

学校读者我要写书评

暂无评论

Genealogical Relationship Extraction from Unstructured Text Using Fine-Tuned Transformer Models

Genealogical Relationship Extraction from Unstructured Text ...

引用

International Conference on Semantic Computing

作者： Carloangello Parrolivelli Lubomir Stanchev Computer Science and Software Engineering Department California Polytechnic State University San Luis Obispo CA USA

The paper tackles the task of extracting genealogical relationships, such as "sibling-of", "parent-of", "child-of", and "spouse-of", from unstructured, free-form text. In order ...

The paper tackles the task of extracting genealogical relationships, such as "sibling-of", "parent-of", "child-of", and "spouse-of", from unstructured, free-form text. In order to solve the problem, we propose a three-stage pipeline consisting of Named Entity Recognition (NER), Coreference Resolution (CR), and Relationship Classification (RC). NER identifies tokens in the text that refer to people, such as proper nouns or nicknames, using the SpaCy software. CR maps multiple tokens representing pronouns to their antecedents. For example, CR could map "She", "His sister", and "Maria" to the antecedent "Maria Johnson". CR allows us to transform a genealogical relationship between two tokens, such as the sibling relationship between "him" and "his sister", to a relationship between the corresponding antecedents, for example, "Bob Johnson" and "Maria Johnson". Our novel algorithm for coreference resolution is based on the AllenNLP software. The last step is the RC, which classifies the relationship between two sets of tokens given adjacent context. We use the LUKE transformer deep-learning model to extract the genealogical relationships. The end-to-end pipeline can extract and correctly classify genealogical relationships from our hand-labeled testing set of Wikipedia documents with macro precision, recall, and F1 scores of 0.794, 0.616, and 0.676, respectively.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Approaching Test Time Augmentation in the Context of Uncertainty Calibration for Deep Neural Networks

arXiv

引用

arXiv 2023年

作者： Conde, Pedro Barros, Tiago Lopes, Rui L. Premebida, Cristiano Nunes, Urbano J. The Institute of Systems and Robotics Department of Electrical and Computer Engineering University of Coimbra Portugal Critical Software South Africa

With the rise of Deep Neural Networks, machine learning systems are nowadays ubiquitous in a number of real-world applications, which bears the need for highly reliable models. This requires a thorough look not only at the accuracy of such systems, but also at their predictive uncertainty. Hence, we propose a novel technique (with two different variations, named M-ATTA and V-ATTA) based on test time augmentation, to improve the uncertainty calibration of deep models for image classification. By leveraging an adaptive weighting system, M/V-ATTA improves uncertainty calibration without affecting the model’s accuracy. The performance of these techniques is evaluated by considering diverse metrics related to uncertainty calibration, demonstrating their robustness. Empirical results, obtained on CIFAR-10, CIFAR-100, Aerial Image Dataset, as well as in two different scenarios under distribution-shift, indicate that the proposed methods outperform several state-of-the-art post-hoc calibration techniques. Furthermore, the methods proposed also show improvements in terms of predictive entropy on out-of-distribution samples. Code for M/V-ATTA available at: https://***/pedrormconde/MV-ATTA Copyright © 2023, The Authors. All rights reserved.

关键词： Calibration

来源：评论

学校读者我要写书评

暂无评论

Malaria Disease Prediction Using Frequency-Based Machine Learning Algorithms and Ensembles Algorithms

Malaria Disease Prediction Using Frequency-Based Machine Lea...

引用

Multidisciplinary engineering and Applied Science (ICMEAS), International Conference on

作者： Hamisu Ismail Ahmad Rajesh Prasad Birendra Kumar Sharma Aisha Yakubu Madaki Aliyu Rabiu Shuaibu Department of Computer Science Nile University of Nigeria Abuja Nigeria Department of Computer Science and Engineering Ajay Kumar Garg Engineering College Ghaziabad UP India Department of MCA Ajay Kumar Garg Engineering College Ghaziabad UP India Department of Software Engineering Nile University of Nigeria Abuja Nigeria

Malaria is one of the most serious and threatening diseases in Sub-Saharan Africa. Its cases increase drastically during the rainy season. It can spread throughout an infected person's entire body in less than an hour. The diagnosis process is time-consuming, and its accuracy is negatively affected due to the lack of technical tools and infrastructure in many laboratories in malaria-prone countries. Hospitals, however, keep patient data during the diagnosis period. Due to the vast amount of diagnostic data gathered and stored on a daily basis in various mediums such as files, it has become increasingly crucial to develop powerful techniques for analysing and interpreting the data. This analysis aims to extract meaningful knowledge and insights that can significantly contribute to malaria diagnosis and decision-making processes. This paper selected four frequency-based algorithms, namely Naïve Bayes, (J48) Decision Tree, ZeroR, and the OneR algorithm, to develop a hybrid model for frequency-based classification algorithms using the available dataset of malaria diagnoses collected from the Federal Medical Centre in Yola, Adamawa state. The research results indicate that the highest accuracy, 88.4%, was achieved by the (J48) Decision Tree model. Additionally, this research employed ensemble methods to enhance the performance of each classification model. The results demonstrated that the accuracy of the Decision Tree and ZeroR models remained the same at 88.4% and 60.9% before and after boosting, respectively. In contrast, the accuracy of the Naïve Bayes and OneR models increased from 79.9% to 87.0% and from 79.8% to 87.6% before and after boosting, respectively. In conclusion, among the frequency-based classification models, the Decision Tree model consistently exhibited the highest accuracy both before and after boosting.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Towards Integrating Emerging AI Applications in SE Education

Towards Integrating Emerging AI Applications in SE Education

引用

Conference on software engineering Education & Training (CSEE&T)

作者： Michael Vierhauser Iris Groher Tobias Antensteiner Clemens Sauerwein Department of Computer Science University of Innsbruck Innsbruck Austria Institute of Business Informatics - Software Engineering Johannes Kepler University Linz Linz Austria

ISBN: (数字)9798350378979

ISBN: (纸本)9798350378986

Artificial Intelligence (AI) approaches have been incorporated into modern learning environments and software engineering (SE) courses and curricula for several years. However, with the significant rise in popularity of large language models (LLMs) in general, and OpenAI's LLM-powered chatbot ChatGPT in particular in the last year, educators are faced with rapidly changing classroom environments and disrupted teaching principles. Examples range from programming assignment solutions that are fully generated via ChatGPT, to various forms of cheating during exams. However, despite these negative aspects and emerging challenges, AI tools in general, and LLM applications in particular, can also provide significant opportunities in a wide variety of SE courses, supporting both students and educators in meaningful ways. In this early research paper, we present preliminary results of a systematic analysis of current trends in the area of AI, and how they can be integrated into university-level SE curricula, guidelines, and approaches to support both instructors and learners. We collected both teaching and research papers and analyzed their potential usage in SE education, using the ACM computer Science Curriculum Guidelines CS2023. As an initial outcome, we discuss a series of opportunities for AI applications and further research areas.

关键词： Systematics Large language models Learning (artificial intelligence) Chatbots Market research Artificial intelligence engineering education

来源：评论

学校读者我要写书评

暂无评论

Frequency-based View Selection in Gaussian Splatting Reconstruction

arXiv

引用

arXiv 2024年

作者： Li, Monica M.Q. Lajoie, Pierre-Yves Beltrame, Giovanni The Department of Computer and Software Engineering Polytechnique Montréal MontrealQCH3T 1J4 Canada

Three-dimensional reconstruction is a fundamental problem in robotics perception. We examine the problem of active view selection to perform 3D Gaussian Splatting reconstructions with as few input images as possible. Although 3D Gaussian Splatting has made significant progress in image rendering and 3D reconstruction, the quality of the reconstruction is strongly impacted by the selection of 2D images and the estimation of camera poses through Structure-from-Motion (SfM) algorithms. Current methods to select views that rely on uncertainties from occlusions, depth ambiguities, or neural network predictions directly are insufficient to handle the issue and struggle to generalize to new scenes. By ranking the potential views in the frequency domain, we are able to effectively estimate the potential information gain of new viewpoints without ground truth data. By overcoming current constraints on model architecture and efficacy, our method achieves state-of-the-art results in view selection, demonstrating its potential for efficient image-based 3D reconstruction. Copyright © 2024, The Authors. All rights reserved.

关键词： 3D reconstruction

来源：评论

学校读者我要写书评

暂无评论

Air Particulate Matters Auto-Rule-Based Labeling to Support Long-Distance Run Environment Data Classification 13

Air Particulate Matters Auto-Rule-Based Labeling to Support ...

引用

13th IEEE Symposium on computer Applications and Industrial Electronics, ISCAIE 2023

作者： Wandy, Wandy Adi, Kusworo Ayu, Media Anugerah Diponegoro University Doctoral Program of Information Systems School of Postgraduate Studies Semarang Indonesia Diponegoro University Faculty of Science and Mathematics Department of Physics Semarang Indonesia Sampoerna University Faculty of Engineering and Technology Department of Computer Science South Jakarta Indonesia

ISBN: (纸本)9798350347319

Sports Science is an interdisciplinary and multidisciplinary science that strives to increase athletic performance and endurance. Sport Science recognizes and prevents injuries. Sensors and statistics formalize Sports Science. Runners need coaches and teams to support them before, during, and after the run race. Coaches generate running training plans to boost performance. Running race performances may be impacted by air pollution exposure while training, so coaches should consider limiting air pollution exposure when training. One of the external factors is Particulate Matter (PM2.5 and PM10). Sensors connecting to the Internet can record external factors and produce csv data. The foundation of supervised machine learning is the labeling process. Labeling a set of data is one of the laborious and time-consuming phases in every machine-learning application because it requires verifying the accuracy of the labels and making any necessary revisions. This research aimed to find a solution to automatically label numerous air particulate matter raw data using a rule based on parameters to reduce manual work, human errors and faster processes. This labeled data will later be used for supervised machine learning classification to support the coach in generating training programs for the runners in a Sports Information System. Based on Indonesia Air Quality index rule-based approach, labeled text data in csv has been generated and tested with PM2.5 and PM10 parameters in three scenarios with a 100% success rate. It was possible to automate the labeling process, and it explained how automation results in fast and accurate results. © 2023 IEEE.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Design of a Secured and Trusted E-Voting System Using Blockchain Technology

Design of a Secured and Trusted E-Voting System Using Blockc...

引用

SmartBlock4Africa, IEEE

作者： Elegba Jeremiah. O. Agbelusi O. Olutayo Victor A. Akintola Kolawole G. Department of Software Engineering School of Computing The Federal University of Technology Akure Nigeria Department of Computer Science School of Computing The Federal University of Technology Akure Nigeria

ISBN: (数字)9798350376838

ISBN: (纸本)9798350376845

The state of election in Nigeria is worrisome. Electoral malpractices have been a major challenge to the Nigerian government in recent times. Government has made frantic efforts to tackle these electoral challenges but the recent event in year 2023 presidential election is worrisome suggests that the solution is not near. Recently, the percentage of eligible voters who vote on election days is declining. Voters turnout went up from 52.3% in 1999 - the first general election since 1993 - to 69% in 2003. But, it has been on the decline nearly ever since − 57.5% in 2007, 53.7% in 2011, 43.7% in 2015 and 34.8% in 2019. In 2023, it is 28.63%. With the way things are going, if urgent action is not taken, this could signal the end of democracy in Nigeria. This research presents the design and implementation of a trusted and secured e-voting based on blockchain technology that will increase voters confidence and trust. The system utilizes *** for the frontend and Solidity on the Polygon blockchain for the backend, ensuring transparency, tamper-resistance, and heightened security. This innovative approach addresses trust and reliability concerns in traditional voting systems, offering a modernized and trustworthy electoral process.

关键词： Smart contracts Authentication Blockchains Reliability Security Electronic voting

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：