检索结果-内蒙古大学图书馆

arXiv 2021年

作者： Chung, Youjin Cho, Jihoon Park, Jinah Computer Graphics and Visualizations lab School of Computing Korea Advanced Institute of Science and Technology Daejeon Korea Republic of

We propose a new training scheme for domain generalization in mitotic figure detection. Mitotic figures show different characteristics for each scanner. We consider each scanner as a’domain’ and the image distribution specified for each domain as’style’. The goal is to train our network to be robust on scanner types by using various’style’ images. To expand the style variance, we transfer a style of the training image into arbitrary styles, by defining a module based on StarGAN. Our model with the proposed training scheme shows positive performance on MIDOG Preliminary Test-Set containing scanners never seen before. Copyright © 2021, The Authors. All rights reserved.

关键词： Scanning

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Models for Multiple Face Mask Detection under a Complex Big Data Environment

引用

Procedia computer science 2022年 215卷 706-712页

作者： V Rekha J Samuel Manoharan R Hemalatha D Saravanan Assistant Professor Dept. of Computer Science Rajalakshmi Engineering College Thandalam Chennai. Professor/Dept. of ECE Sir Isaac Newton College of Engineering & Technology Nagapattinam South India. Research Scholar Depart. of Information Technology Annamalai University Chidhambaram.

The Covid-19 (coronavirus) pandemic creates a worldwide health crisis. According to the WHO, the effective protection system is wearing a face mask in public places. Many studies proved that carrying a face mask is also one of the precautions to decrease the possibility of viral transmission. Strict monitoring of face mask being worn by people is now enforced in many countries. Manual observation and monitoring is quite tedious. Hence, automated systems have been researched using well-kwown face mask detection methods. However, this research paper, deals with some deep learning models which can be effectively used to detect multiple face masks in a crowded environment when the amount of incoming data from sensors is huge or in otherwise stated to a Big data problem. Hence, standalone face detection models are not quite suited. Deep learning models are required in such Big data scenario which forms the essence of this study.

关键词： Big Data Complex Data analytics Deep Learning models Face Mask Detection Multi – Sensor Data Acquisition

来源：评论

学校读者我要写书评

暂无评论

Preservice teachers’ mobile learning experience: An exploratory study of iPad-enhanced collaborative learning

引用

Journal of Digital Learning in Teacher Education 2020年第2期36卷 111-123页

作者： Kuo, Yu-Chun Kuo, Yu-Tung Department of Science Technology Engineering Arts and Math Education Rowan University Glassboro NJ United States Computer Graphics Technology Purdue University West Lafayette IN United States

This study investigated preservice teachers’ perceived learning experience in iPad-enhanced, collaborative learning environments. Three variables, including sense of community, perceived collaborative learning, and iPad self-efficacy, and their correlations with perceived learning experience were examined. The participants were 67 preservice teachers from a northeastern university in the United States. Data were collected using an online survey. The results indicated that there were significantly positive correlations between perceived learning and the sense of community, perceived collaborative learning, and iPad self-efficacy. Sense of community was the stronger predictor of preservice teachers’ perceived learning. iPad self-efficacy was a key factor in preservice teachers’ learning experience in the iPad-enhanced collaborative work. iPad ownership had a potential influence on preservice teachers’ iPad self-efficacy. © 2020, © 2020 ISTE.

关键词： collaborative learning iPad self-efficacy mobile learning preservice teachers sense of community

来源：评论

学校读者我要写书评

暂无评论

AIR-DA: Adversarial Image Reconstruction for Unsupervised Domain Adaptive Object Detection

arXiv

引用

arXiv 2023年

作者： Sun, Kunyang Lin, Wei Shi, Haoqin Zhang, Zhengming Huang, Yongming Bischof, Horst The School of Information Science and Engineering The National Mobile Communications Research Laboratory Southeast University Nanjing210096 China The Pervasive Communications Center Purple Mountain Laboratories Nanjing211111 China The Institute for Computer Graphics and Vision Graz University of Technology Graz8010 Austria

Unsupervised domain adaptive object detection is a challenging vision task where object detectors are adapted from a label-rich source domain to an unlabeled target domain. Recent advances prove the efficacy of the adversarial based domain alignment where the adversarial training between the feature extractor and domain discriminator results in domain-invariance in the feature space. However, due to the domain shift, domain discrimination, especially on low-level features, is an easy task. This results in an imbalance of the adversarial training between the domain discriminator and the feature extractor. In this work, we achieve a better domain alignment by introducing an auxiliary regularization task to improve the training balance. Specifically, we propose Adversarial Image Reconstruction (AIR) as the regularizer to facilitate the adversarial training of the feature extractor. We further design a multi-level feature alignment module to enhance the adaptation performance. Our evaluations across several datasets of challenging domain shifts demonstrate that the proposed method outperforms all previous methods, of both one- and two-stage, in most settings. Copyright © 2023, The Authors. All rights reserved.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

A Deep Learning Approach to Automated Sleep Stages Classification Using Multi-Modal Signals

A Deep Learning Approach to Automated Sleep Stages Classific...

引用

2022 International Conference on Machine Learning and Data Engineering, ICMLDE 2022

作者： Satapathy, Santosh Kumar Kondaveeti, Hari Kishan Sreeja, S.R. Madhani, Hiral Rajput, Nitinsingh Swain, Debabrata Information and Communication Technology Pandit Deendayal Energy University Gujarat Gandhi Nagar382007 India VIT-AP University Assistant Professor CSE Group IIIT Sri City Andhra Pradesh Amaravati522237 India Computer Science and Engineering Pandit Deendayal Energy University Gujarat Gandhi Nagar382007 India

Sleep is one of the elements most vital to human life. However, the modern lifestyle continues to push people to neglect this critical requirement. With a vast majority of people falling victim to various sleep disorders, it has become increasingly essential to have a robust system for diagnosing and treating such ailments. Sleep stage classification is one of the primary steps for identifying sleep-related anomalies. Sleep stages are classified according to the frequency and nature of signals received during a polysomnography test. Since the early days, this has been performed manually with the help of trained technicians. However, manual scoring is often prone to error and subjectivity and requires tremendous time and effort. It is, therefore, essential to automatize this process. Several challenges from the correct selection of features remain to be faced in the machine learning-based sleep stage classification system. As an alternative, Deep Learning, capable of automatic feature extraction, proves far more reliable for this task. This experimental study analyses both techniques to compare and decide on a better approach. Three popular Machine Learning classifiers, namely Random Forest (RF), K-Nearest Neighbours (KNN), and Support Vector Machines (SVM), and a neural network comprising CNN and LSTM, have been trained on a vast base of diverse data. The proposed model reported an accuracy of 87.4% with CNN+LSTM. © 2022 Elsevier B.V.. All rights reserved.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

+TrajVis: a visual clinical decision support system to translate artificial intelligence trajectory models in the precision management of chronic kidney disease

arXiv

引用

arXiv 2024年

作者： Li, Zuotian Liu, Xiang Tang, Ziyang Zhang, Pengyue Jin, Nanxin Eadon, Michael T. Song, Qianqian Chen, Yingjie Su, Jing Department of Biostatistics and Health Data Science Indiana University School of Medicine Indiana United States Department of Computer Graphics Technology Purdue University Indiana United States Department of Computer and Information Technology Purdue University Indiana United States Department of Medicine Indiana University School of Medicine Indiana United States Department of Health Outcomes and Biomedical Informatics College of Medicine University of Florida Florida United States Gerontology and Geriatric Medicine Wake Forest School of Medicine North Carolina United States

Objective: Our objective is to develop and validate TrajVis, an interactive tool that assists clinicians in using artificial intelligence (AI) models to leverage patients' longitudinal electronic medical records (EMR) for personalized precision management of chronic disease progression. Methods: We first perform requirement analysis with clinicians and data scientists to determine the visual analytics tasks of the TrajVis system as well as its design and functionalities. A graph AI model for chronic kidney disease (CKD) trajectory inference named DEPOT is used for system development and demonstration. TrajVis is implemented as a full-stack web application with synthetic EMR data derived from the Atrium Health Wake Forest Baptist Translational Data Warehouse and the Indiana Network for Patient Care research database. A case study with a nephrologist and a user experience survey of clinicians and data scientists are conducted to evaluate the TrajVis system. Results: The TrajVis clinical information system is composed of four panels: the Patient View for demographic and clinical information, the Trajectory View to visualize the DEPOT-derived CKD trajectories in latent space, the Clinical Indicator View to elucidate longitudinal patterns of clinical features and interpret DEPOT predictions, and the Analysis View to demonstrate personal CKD progression trajectories. System evaluations suggest that TrajVis supports clinicians in summarizing clinical data, identifying individualized risk predictors, and visualizing patient disease progression trajectories, overcoming the barriers of AI implementation in healthcare. Conclusion: TrajVis bridges the gap between the fast-growing AI/ML modeling and the clinical use of such models for personalized and precision management of chronic diseases. © 2024, CC BY-NC-ND.

关键词： Diseases

来源：评论

学校读者我要写书评

暂无评论

Prediction of the Impact of Covid-19 Vaccine on Public Health Using Twitter

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer science 2022年第1期49卷 19-29页

作者： Syed, Ayesha Ayub Gaol, Ford Lumban Suparta, Wayan Abdurachman, Edi Trisetyarso, Agung Matsuo, Tukoro Ph.D. student at the Department of Doctor of Computer Science Bina Nusantara University Jakarta Indonesia Associate Professor at Informatics Engineering and Information System at Bina Nusantara University Jakarta Indonesia Malaysia Associate Professor at the Department of Doctor of Research in Management at Bina Nusantara University Jakarta Indonesia Assistant Professor at the Department of Doctor of Computer Science Bina Nusantara University Jakarta Indonesia Professor at the Graduate School of Industrial Technology Advanced Institute of Industrial Technology Japan and Department of M-Commerce and Multimedia Applications Asia University Taiwan

Social media is a source of big data. Media like Twitter and Facebook has been used for collecting and analyzing user data for different purposes. The data can be used to analyze people opinions towards certain topics and incidents by applying sentiment analysis and then certain useful insights can be drawn from the analyzed data. During the current time of Covid-19, people have been sharing information regarding Covid-19 statistics, vaccines, and discussing the effects of the vaccine concerning public health. The purpose of this study is to analyze tweet data regarding the Covid-19 vaccine by applying sentiment analysis and predicting the impact of the vaccine on public health. Also, the tweets are analyzed for hidden topics by applying Topic Modelling using Latent Dirichlet Allocation (LDA) and Latent Semantic Analysis (LSA). The source of data for this study is Twitter API. The coding and data analysis is done using Python programming language in the Spyder (Scientific Python Development Environment) that is an integrated development environment for scientific programming, testing, and data analysis. The results of the study indicate a greater positive sentiment reflecting a healthy public discussion about the Covid-19 vaccine, information, awareness, and public acceptance. With these results, a positive impact of the Covid-19 vaccine on public health is predicted. The results of topic modeling discovered 10 hidden topics from the tweet dataset. © 2022. All Rights Reserved.

关键词： Vaccines

来源：评论

学校读者我要写书评

暂无评论

Automatic Story Generation Based on Graph Model Using Godot Engine 19th

Automatic Story Generation Based on Graph Model Using Godot ...

引用

19th International Conference on Artificial Intelligence and Soft Computing, ICAISC 2020

作者： Grabska-Gradzińska, Iwona Nowak, Leszek Grabska, Ewa Department of Games Technology Faculty of Physics Astronomy and Applied Computer Science Jagiellonian University Krakow Poland Department of Information Technologies Faculty of Physics Astronomy and Applied Computer Science Jagiellonian University Krakow Poland Department of Design and Computer Graphics Faculty of Physics Astronomy and Applied Computer Science Jagiellonian University Krakow Poland

ISBN: (纸本)9783030614003

This paper deals with storytelling aspects of computer games. A new approach to automatic story generation based on graph modelling is proposed. Godot Engine for visualization of output stories is used. In our research the plot is represented by a specific graph called the layered graph composed of story building blocks, namely layer of locations, layer of characters, layer of items, layer of narration and plot dependencies. An appropriate set of graph rules (productions) stored in JSON format is a generative tool for the building blocks. Graph rules are internal representation of actions which can be taken by the hero of the story. A sequence of the rules from start to end can describe the whole gameplay of one player or can be generated in automatic way as a possible gameplay. All such sequences of the rules create the space of all potential gameplays. Every rule corresponds to a certain animation made in Godot Engine. It allows to visualize every possible story, represented by sequence of rules, by seamless animation. Given a rule, animation is based on graphic assets corresponding to the rule nodes of the layered graph. © 2020, Springer Nature Switzerland AG.

关键词： Animation

来源：评论

学校读者我要写书评

暂无评论

Ultra-Low Energy 5G IoT Devices with Updatable AI Models: A Dynamic AI-IoT Perspectives

Ultra-Low Energy 5G IoT Devices with Updatable AI Models: A ...

引用

Communications and Information Technologies (GCCIT), Global Conference on

作者： Dharmesh Dhabliya K. Yuvaraj Rahul Chauhan Dikshit Sharma Kala Priyadarshini G Shankar Prasad S Department of Information Technology Vishwakarma Institute of Technology Pune India Department of Computer Science Karpagam Academy of Higher Education Coimbatore Computer Science and Engineering Graphic Era Hill University Dehradun Adjunct Professor Graphic Era Deemed to be University Dehradun Uttarakhand India Centre of Research Impact and Outcome Chitkara University Rajpura Punjab India Department of Electrical and Electronics Engineering Prince Shri Venkateshwara Padmavathy Engineering College Chennai School of Management - UG Management JAIN (Deemed to be University) Bangalore Karnataka

ISBN: (数字)9798350388916

ISBN: (纸本)9798350388923

Ultra-Low-Power (ULP) Internet of Things (IoT) devices are becoming more important in the fast developing 5G and possible 6G networks, and this study tackles the problem of incorporating dynamic AI capabilities into these devices. We present the Dynamic AI-IoT design, a new framework that aims to do away with the tedium of firmware upgrades. This design makes use of Narrowband IoT (NB-IoT) to make cloud interactions run smoothly and includes specialized firmware modifications to make dynamic interactions with TinyML models possible. Many sectors rely on deep learning that is based on massive amounts of high-quality data. The information accumulation of the framework relies on real-time acquisitions, which makes it difficult to directly implement deep learning in a real-time system. But such systems' analytical activities can't be done by collecting data for a lengthy period since they need to be executed in real time. This research presents a novel progressive deep-learning architecture and performs experiments on picture identification to address the challenges of high-quality data collection, high-timeliness data processing, and the difficulties of directly integrating deep-learning algorithms in real-time systems. The experimental findings demonstrate the efficacy, performance, and ability to get a conclusion comparable to the deep-learning architecture using large-scale data. An advanced method for managing memory is presented for the effective handling of AI tasks; it is based on memory alignment and the dynamic resolution of AI activities. Through the use of ULP IoT gadgets on a 5G testbed, empirical investigations show that the implementation of a Dynamic AI-IoT network is feasible. With an average inference duration of around 46 ms and model updates averaging less than one second, the findings are impressive.

关键词： Deep learning 5G mobile communication Heuristic algorithms Tiny machine learning Data collection Real-time systems Internet of Things Artificial intelligence Microprogramming Narrowband

来源：评论

学校读者我要写书评

暂无评论

A Method for the Identification of Multiple Labels in Social Media Text using CNN Approach

A Method for the Identification of Multiple Labels in Social...

引用

Communications and Information Technologies (GCCIT), Global Conference on

作者： Dharmesh Dhabliya Vikrant Sharma N. Thangarasu Prakriti Kapoor V. Mahalakshmi Sampangi Rama Reddy B.R Department of Information Technology Vishwakarma Institute of Technology Pune India Computer Science and Engineering Graphic Era Hill University Dehradun Adjunct Professor Graphic Era Deemed to be University Dehradun Uttarakhand India Department of Computer Science Karpagam Academy of Higher Education Coimbatore Centre of Research Impact and Outcome Chitkara University Rajpura Punjab India Department of Electrical and Electronics Engineering Prince Shri Venkateshwara Padmavathy Engineering College Chennai School of Sciences Physics & Electronics JAIN (Deemed to be University) Bangalore Karnataka

ISBN: (数字)9798350388916

ISBN: (纸本)9798350388923

People are able to communicate their thoughts and perspectives with one another through various online settings, including social networks, news websites, and blogs. Because so many individuals instantaneously post their thoughts on the internet, the number of submissions that have been submitted has increased dramatically. It is critical for establishments and organizations to do an effective and speedy analysis of the massive amounts of data at their disposal in order to provide summaries that discuss the perspectives or sentiments of individuals. In this article, we offer a scalable framework that makes sentiment classification by assessing the compound probability scores of the most generally used methods in sentiment analysis using a fuzzy inference mechanism in a CNN-style way. This evaluation is carried out in order to make the framework applicable to a broad range of situations. The valance of an individual's attitude may be estimated by the designed fuzzy inference system by calculating the compound scores of the valance aware dictionary, word embedding, and count vectorization procedures. The suggested technique differs from the convolutional neural networks (CNNs) in that it permits weighting of base learners and combines the strengths of each algorithm using fuzzy rules. Another distinction is that the proposed method is more general. The process of estimating sentiment based on textual data can be handled as either a two-class (positive and negative) or a three-class (positive, neutral, and negative) problem. The former has less challenges than the latter. We carried out the experimental study on four different tagged social network data sets, and discovered that the suggested strategy delivers gains in accuracy for both 2-class and 3-class classifications.

关键词： Fuzzy logic Training Sentiment analysis Social networking (online) Standards organizations Organizations Media Information and communication technology Convolutional neural networks Compounds

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：