检索结果-内蒙古大学图书馆

10th IEEE Smart World Congress, SWC 2024

作者： Li, Danlei Nair, Nirmal Wang, Kevin I-Kai Sakurai, Kouichi University of Auckland Department of Electrical Computer and Software Engineering Auckland New Zealand Kyushu University Department of Informatics Faculty of Information Science and Electrical Engineering Fukuoka Japan

ISBN: (纸本)9798331520861

Anomaly detection is the identification of instances that substantially deviate from the majority of the data and do not conform to a well-defined normal behavior. Investigating time series anomalies has become increasingly popular in recent years due to widely deployed IoT devices and sensors. Although most existing methods apply image anomaly detection algorithms directly on time series and commonly assume the data is IID (independent and identically distributed). However, in most realworld applications, the underlying data distribution alters over time. Changes in data over time are referred to as concept drift and can significantly degrade the efficacy of an anomaly detection model. Existing research also indicates that the majority of anomaly detection frameworks ignored concept drift, resulting in poor long-term model performance. This paper presents an ensemble-based, unsupervised anomaly detection framework that is adaptive to the concept drift in time series data. Both distribution and performance-based techniques are used to detect concept drift and update the model to ensure the framework's longterm performance reliability. The proposed method is evaluated using a real-world univariate time series data that captures sea surface temperature data over a 2 -year period. By comparing our proposed framework to other state-of-the-art algorithms, we demonstrate our proposed method has obvious benefits in accuracy and long-term stability. © 2024 IEEE.

关键词： Anomaly detection Concept drift adaptation Ensemble method Time series Unsupervised method

来源：评论

学校读者我要写书评

暂无评论

引用

14th International Conference on Information, Intelligence, Systems and Applications, IISA 2023

作者： Gkillas, Athanasios Simos, Michael Angelos Makris, Christos University of Patras Department of Electrical and Computer Engineering Patras Greece University of Patras Department of Computer Engineering and Informatics Patras Greece

ISBN: (纸本)9798350318067

YouTube is a popular online video-sharing and social media platform, hosting diverse content for all ages. The platform is gradually gaining ground against traditional media, with major turnover increasingly turning there due to videos and creators' influence, especially in younger audiences. Some of the videos may attain a large number of views often accompanied by many comments. Comments, often focus not only on the content of the video but also on several other aspects and topics. In this paper, we focus on the sentiment analysis of video comments, using a variety of methods we explore the correlation between user comments' sentiment polarity score and the popularity of the relevant video. Some adversities of machine learning or neural network model approach to the task include linguistic irregularities, typing errors, niche vocabulary, sarcastic and ambiguous content, or unrelated events influencing the commented themes. Unlike previous works [1], [2], our focus is centered on leveraging comment sentiment polarity for deriving predictions of the popularity and success of the respective videos. Our methods depend on lexicon-based sentiment analysis approaches like VADER and TextBlob but also on neural network architectures including BERT and RoBERTa. Our experimental analysis resulted in promising results, setting a baseline for this application theme. © 2023 IEEE.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Occupant-Centric Proactive Control Methodology for IEQ 18

Occupant-Centric Proactive Control Methodology for IEQ

引用

18th International Conference on Indoor Air Quality and Climate, INDOOR AIR 2024

作者： Kakoulli, Christina Kyriacou, Alexis Michaelides, Michalis Department of Electrical Engineering Computer Engineering and Informatics Cyprus University of Technology Limassol Cyprus Lelantus Innovations Ltd Nicosia Cyprus

ISBN: (纸本)9798331306816

Indoor Environmental Quality (IEQ) is undeniably a crucial factor that affects the occupants’ overall health and well-being. However, there is a significant lack of available actionable and measurable real-time metrics for the impact of IEQ on occupants, restricting potential control solutions and the validation of their effectiveness. Even more, occupant-centric and energy concerned automatic control solutions with direct aim to decrease or even prevent any negative effects of poor IEQ on occupants are required for creating sustainable built environments. In this work, we propose a novel framework for an occupant-centric proactive air-purification and ventilation control methodology (PPV). The method utilizes state-space models along with what-if scenarios to generate the potential solution space. A multi-objective optimization extracts the optimal solution in terms of air-purification and ventilation operation points, while minimizing various computed IEQ impact metrics in terms of the occupants’ health, comfort and virus transmissibility risk, and also considering the energy expenditure. © 2024 18th Conference of the International Society of Indoor Air Quality and Climate, INDOOR AIR 2024 - Conference Program and Proceedings. All rights reserved.

关键词： Indoor air pollution

来源：评论

学校读者我要写书评

暂无评论

Detection and Counting Bacilli Tuberculosis Images using Classical Image Processing and Deep Learning 24

Detection and Counting Bacilli Tuberculosis Images using Cla...

引用

2nd International Conference on Communications, Computing and Artificial Intelligence, CCCAI 2024

作者： Yuniarti, Heny Fatichah, Chastine Sigit, Riyanto Department of Informatics Faculty of Intelligent Electrical and Informatics Technology Institut Teknologi Sepuluh Nopember Indonesia Department of Computer Engineering Politeknik Elektronika Negeri Surabaya Indonesia

ISBN: (纸本)9798400716898

Tuberculosis (TB) has been known as the top two murdering disease worldwide after HIV/AIDS. Indonesia is the second country with the highest number of tuberculosis. One of the reasons for the high number of cases is that the microscopic sputum smear image analysis process takes a long time. This study compares the detection and counting of bacilli images using classical image processing and deep learning to ascertain the performance of both methods. Steps in classical image processing consist of RGB to HSV converting, HSV thresholding, morphological operation and bacilli counting. YOLOv5 is used as deep learning method in this research to detect and count bacilli. While classical image processing accuracy is 59.59%, YOLOv5 can achieve 86.09% which is better than the previous research. By comparing classical image processing methods and deep learning, it is evident that deep learning techniques can detect and count overlapping bacilli. This result can be further improved by increasing the learning dataset especially the severe TB patient. © 2024 ACM.

关键词： Image analysis

来源：评论

学校读者我要写书评

暂无评论

Generating animatable 3D cartoon faces from single portraits

引用

虚拟现实与智能硬件（中英文） 2024年第4期6卷 292-307页

作者： Chuanyu PAN Guowei YANG Taijiang MU Yu-Kun LAI Department of Electrical Engineering and Computer Sciences University of California BerkeleyBerkeley CA 94704USA Department of Computer Science and Technology Tsinghua UniversityBeijing 100084China School of Computer Science and Informatics Cardiff UniversityCardiff CF244AGUK

Background With the development of virtual reality(VR)technology,there is a growing need for customized 3D ***,traditional methods for 3D avatar modeling are either time-consuming or fail to retain the similarity to the person being *** study presents a novel framework for generating animatable 3D cartoon faces from a single portrait *** First,we transferred an input real-world portrait to a stylized cartoon image using *** then proposed a two-stage reconstruction method to recover a 3D cartoon face with detailed *** two-stage strategy initially performs coarse estimation based on template models and subsequently refines the model by nonrigid deformation under landmark ***,we proposed a semantic-preserving face-rigging method based on manually created templates and deformation *** Compared with prior arts,the qualitative and quantitative results show that our method achieves better accuracy,aesthetics,and similarity ***,we demonstrated the capability of the proposed 3D model for real-time facial animation.

关键词： 3D reconstruction Cartoon face reconstruction Face rigging Stylized reconstruction Virtual reality

来源：评论

学校读者我要写书评

暂无评论

Impact of Fast Frequency Response on Renewable Energy Source Curtailment and Load Shedding

Impact of Fast Frequency Response on Renewable Energy Source...

引用

CIRED 2024 Vienna Workshop

作者： Panagi, Savvas Lazari, Antonis Koutsoloukas, Vrahimis Aristidou, Petros Department of Electrical Engineering Computer Engineering and Informatics Cyprus University of Technology Limassol Cyprus Transmission System Operator of Cyprus Cyprus

ISBN: (纸本)9781837242672

System operators in low-inertia power systems often have to curtail renewable energy sources (RES) and employ strict under-frequency load shedding (UFLS) schemes to ensure frequency security after an event leading to loss of generation. This approach limits the maximum RES penetration in a system and results to the loss of load. To tackle these problems, fast frequency response (FFR) schemes can be used to limit the frequency Nadir after a disturbance and decrease the need for RES curtailments and UFLS. This article provides insights on the interaction between the Kinetic Energy (KE), Frequency Containment Reserves (FCR), and Nadir after a disturbance, which are the driving mechanisms leading to RES curtailment. Then, it analyses the impact of FFR on the Nadir and its ability to alleviate RES curtailment problems. The low-inertia, islanded, Cyprus dynamic model is used to quantify the results and showcase the impact on a real system. © The Institution of engineering & Technology 2024.

关键词： Electric load shedding

来源：评论

学校读者我要写书评

暂无评论

A machine learning approach for skin lesion classification on iOS: implementing and optimizing a convolutional transfer learning model with Create ML

引用

International Journal of computers and Applications 2024年第8期46卷 666-685页

作者： Benedek Szabo, Aron Katona, Jozsef Department of Software Development and Application Institute of Computer Engineering University of Dunaujvaros Dunaujvaros Hungary Department of Instrumentation and Automation Institute of Electronics and Communication Systems Kandó Kálmán Faculty of Electrical Engineering Obuda University Budapest Hungary Department of Informatics GAMF Faculty of Engineering and Computer Science Kecskemét Hungary

The integration of machine learning (ML) into mobile applications presents unique challenges, particularly in resource-constrained environments such as iOS devices. Skin lesion classification is a critical task in dermatology, where accurate and efficient diagnostic tools can significantly aid in early detection of malignant lesions. This study aims to implement a machine learning-based iOS mobile application and develop a binary classification model for skin lesion images to determine whether a lesion is malignant. The research utilized Create ML to develop a convolutional neural network optimized for iOS, employing a transfer learning approach. A logistic regression model was cascaded with the convolutional neural network N to enhance classification accuracy. The model's performance was assessed through various validation metrics, ensuring its robustness and efficiency within the constraints of mobile hardware. A curated dataset from the International Skin Imaging Collaboration archive was used for training and testing. The model achieved an accuracy of 92%, a precision of 90%, a recall of 93%, and an F1-score of 91% in classifying skin lesions. These metrics validate the model's efficacy in identifying malignant lesions. Data curation involved collecting, labeling, and preparing a dataset from publicly available sources, ensuring the inclusion of diagnostically relevant features. The final model was integrated into an iOS application using Core ML and Vision frameworks. The developed application demonstrates reliable performance in classifying skin lesions with high accuracy on iOS devices. The inclusion of comprehensive performance metrics justifies the efficacy of the proposed approach. Future work will explore enhancements in model architecture and object detection capabilities to further improve diagnostic precision and application usability. © 2024 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Image recognition

来源：评论

学校读者我要写书评

暂无评论

Sensitive Content Detection in Social Networks Using Deep Learning Models and Explainability Techniques 9

Sensitive Content Detection in Social Networks Using Deep Le...

引用

9th IEEE/ACIS International Conference on Big Data, Cloud Computing, and Data Science, BCD 2024

作者： Perikos, Isidoros University of Patras Computer Engineering and Informatics Department Patras Greece University of Peloponnese Computer Technology Institute and Press 'Diophantus' Electrical and Computer Engineering Dept. Patras Greece

ISBN: (纸本)9798350394191

In the dynamic landscape of online social networks, recognizing sensitive content is essential for safeguarding user privacy, fostering inclusivity, and enhancing diversity awareness. Building on prior research, this study explores new dimensions and methodologies for detecting sensitive content. We examine the temporal evolution of sensitive content, revealing how patterns shift over time, and address cross-linguistic challenges, emphasizing cultural and contextual nuances in detection. We employ advanced machine learning techniques, including deep learning models and BERT that improve the accuracy and robustness of the detection procedure. In the experimental study, BERT transformer reported the best performance in detecting sensitive content in text. Additionally, we incorporate explainability techniques such as LIME and SHAP to provide deeper insights into the model's decision-making processes, ensuring predictions are interpretable and reliable. Our work enhances the theoretical framework of sensitive content detection in social networks and provide methods that are accurate and scalable and can facilitate the creation of user-centric interaction that prioritize privacy and user experience. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Increasing the RES Hosting Capacity of the Cyprus Distribution System

Increasing the RES Hosting Capacity of the Cyprus Distributi...

引用

CIRED 2024 Vienna Workshop

作者： Therapontos, Phivos Konstantinou, Rafail Charalambous, Charalambos A. Aristidou, Petros Electricity Authority of Cyprus Nicosia Cyprus University of Cyprus Department of Electrical and Computer Engineering Nicosia Cyprus Dept. of Electrical & Computer Engineering & Informatics Cyprus University of Technology Limassol Cyprus

ISBN: (纸本)9781837242672

The Renewable Energy Sources (RES) penetration in the power system of Cyprus has dramatically increased over the last years. As a result, the system is already facing significant challenges limiting the RES hosting capacity of the Distribution Network (DN). The major limitation factors are network congestion and voltage security. In this paper, alternative solutions for increasing the RES hosting capacity in DNs are reviewed, and a methodology is introduced to evaluate their effectiveness. More specifically, different inverter settings and an advanced centralised voltage control from power transformers are used to mitigate voltage-related issues, while network reinforcements and upgrading the operating voltage are considered for further increase in hosting capacity. The solutions and evaluation methodology are tested using a real MV network of the Cyprus distribution system and the strategic plan of the Cyprus Distribution System Operator (DSO) for maximizing RES hosting capacity is outlined. © The Institution of engineering & Technology 2024.

关键词： Distribution transformers

来源：评论

学校读者我要写书评

暂无评论

Vpit: real-time embedded single object 3D tracking using voxel pseudo images

引用

Neural Computing and Applications 2024年第32期36卷 20341-20354页

作者： Oleksiienko, Illia Nousi, Paraskevi Passalis, Nikolaos Tefas, Anastasios Iosifidis, Alexandros Department of Electrical and Computer Engineering Aarhus University Finlandsgade Aarhus8200 Denmark Department of Informatics Aristotle University of Thessaloniki University Campus Thessaloniki54124 Greece Department of Chemical Engineering Aristotle University of Thessaloniki University Campus Thessaloniki54124 Greece

In this paper, we propose a novel voxel-based 3D single object tracking (3D SOT) method called Voxel Pseudo Image Tracking (VPIT). VPIT is the first method that uses voxel pseudo images for 3D SOT. The input point cloud is structured by pillar-based voxelization, and the resulting pseudo image is used as an input to a 2D-like Siamese SOT method. The pseudo image is created in the Bird’s-eye View (BEV) coordinates;and therefore, the objects in it have constant size. Thus, only the object rotation can change in the new coordinate system and not the object scale. For this reason, we replace multi-scale search with a multi-rotation search, where differently rotated search regions are compared against a single target representation to predict both position and rotation of the object. Experiments on KITTI [1] Tracking dataset show that VPIT is the fastest 3D SOT method and maintains competitive Success and Precision values. Application of a SOT method in a real-world scenario meets with limitations such as lower computational capabilities of embedded devices and a latency-unforgiving environment, where the method is forced to skip certain data frames if the inference speed is not high enough. We implement a real-time evaluation protocol and show that other methods lose most of their performance on embedded devices;while, VPIT maintains its ability to track the object. © The Author(s) 2024.

关键词： 3D tracking Embedded deep learning (DL) Pillars Pseudo images Real-time neural networks Single object tracking Voxels

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：