检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Sartinas, Evangelos G. Psarakis, Emmanouil Z. Kosmopoulos, Dimitrios I. Department of Computer Engineering and Informatics University of Patras Patras26504 Greece

An interesting problem in many video-based applications is the generation of short synopses by selecting the most informative frames, a procedure which is known as video summarization. For sign language videos the benefits of using the t-parameterized counterpart of the curvature of the 2-D signer’s wrist trajectory to identify keyframes, have been recently reported in the literature. In this paper we extend these ideas by modeling the 3-D hand motion that is extracted from each frame of the video. To this end we propose a new informative function based on the t-parameterized curvature and torsion of the 3-D trajectory. The method to characterize video frames as keyframes depends on whether the motion occurs in 2-D or 3-D space. Specifically, in the case of 3-D motion we look for the maxima of the harmonic mean of the curvature and torsion of the target’s trajectory;in the planar motion case we seek for the maxima of the trajectory’s curvature. The proposed 3-D feature is experimentally evaluated in applications of sign language videos on (1) objective measures using ground-truth keyframe annotations, (2) human-based evaluation of understanding, and (3) gloss classification and the results obtained are *** Codes 68T45, 68U10 © 2023, CC BY.

关键词： Torsional stress

来源：评论

学校读者我要写书评

暂无评论

Improving Indonesian Informal to Formal Style Transfer via Pre-Training Unlabelled Augmented Data

Improving Indonesian Informal to Formal Style Transfer via P...

引用

International Conference of computer and informatics engineering (IC2IE)

作者： Bassamtiano Renaufalgi Irnawan Rizky Adi Department of Computer and Informatics Engineering Politeknik Negeri Jakarta Depok Indonesia

Neural Machine Translation is already used for Indonesian informal to formal style transfer. It works by translating the input language to the target language. In Indonesian informal-to-formal style transfer task, informal sentence work as an input language, and formal sentence is the target the model needs to translate to. Currently, the STIF parallel dataset is the only manually labelled informal-to-formal dataset. We need sufficient data to achieve a good model for style transfer performance. In contrast, the current Indonesian informal to formal dataset is insufficient. We adopted the pre-train augmentation architecture introduced by work done in GEC tasks to elevate the Low-Resource data. We create the augmented dataset with a simpler word replacement approach. We benchmark several transformer-based pre-trained model architectures, including BART, GPT2, and BERT Encoder Decoder. We train the augmented dataset to all models as a pre-trained model and fine-tune it with the STIF dataset. We perform the sacreBLEU benchmarking techniques to find which approach with better style transfer quality. The result is the BART model that was pre-trained with an augmented dataset and fine-tuned with the STIF dataset with a score of sacreBLEU 53,19.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Personalized Learner Assistance Through Dynamic Adaptation of Chatbot Using Fuzzy Logic Knowledge Modeling

Personalized Learner Assistance Through Dynamic Adaptation o...

引用

International Workshop on Semantic Media Adaptation and Personalization (SMAP)

作者： Christos Troussas Akrivi Krouska Phivos Mylonas Cleo Sgouropoulou Department of Informatics and Computer Engineering University of West Attica Egaleo Greece

Personalized approaches and tailored support have become increasingly significant in the field of online education, aiming to enhance the overall learning experiences of learners. This paper introduces a novel approach for addressing challenges in providing tailored support by utilizing chatbot technology and the flexibility of fuzzy logic. The chatbot is responsible for delivering precise and tailored responses to learners, considering their input, typically in text form. This is accomplished through the utilization of a rule-based system that is capable of generating accurate answers according to predefined criteria. To augment this support, fuzzy logic is employed for modeling the learners' knowledge, thereby enhancing the chatbot's proficiency in accurately evaluating and responding to inquiries. Consequently, the provision of assistance can be tailored to the specific knowledge level of learners, aiding them in achieving their educational goals. This methodology is incorporated in an intelligent tutoring system designed to provide tutoring for the programming language Java. The evaluation findings demonstrated the effectiveness of our approach in delivering personalized assistance through a chatbot. The results indicated that the chatbot's responses were highly rated in terms of clarity, relevance, and usefulness. Additionally, the system was found to effectively address learners' needs with quality and adequacy.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A New Segmentation Framework for Arabic Handwritten Text Using Machine Learning Techniques

引用

computers, Materials & Continua 2021年第8期68卷 2727-2754页

作者： Saleem Ibraheem Saleem Adnan Mohsin Abdulazeez Zeynep Orman Department of Information Technology Technical Informatics College of AkreDuhok Polytechnic UniversityDuhok42004Kurdistan RegionIraq Department of Computer Engineering Faculty of EngineeringIstanbul University-CerrahpasaIstanbul34320Turkey

The writer identification(WI)of handwritten Arabic text is now of great concern to intelligence agencies following the recent attacks perpetrated by known Middle East terrorist *** is also a useful instrument for the digitalization and attribution of old text to other authors of historic studies,including old national and religious *** this study,we proposed a new affective segmentation model by modifying an artificial neural network model and making it suitable for the binarization stage based on *** modified method is combined with a new effective rotation model to achieve an accurate segmentation through the analysis of the histogram of binary ***,propose a new framework for correct text rotation that will help us to establish a segmentation method that can facilitate the extraction of text from its *** projections and the radon transform are used and improved using machine learning based on a co-occurrence matrix to produce binary *** training stage involves taking a number of images for model *** images are selected randomly with different angles to generate four classes(0–90,90–180,180–270,and 270–360).The proposed segmentation approach achieves a high accuracy of 98.18%.The study ultimately provides two major contributions that are ranked from top to bottom according to the degree of *** proposed method can be further developed as a new application and used in the recognition of handwritten Arabic text from small documents regardless of logical combinations and sentence construction.

关键词： Writer identification handwritten Arabic biometric systems artificial neural network segmentation skew detection model

来源：评论

学校读者我要写书评

暂无评论

Optimizing reinforcement learning for large action spaces via generative models: Battery pattern selection

引用

Pattern Recognition 2025年 160卷

作者： Hu, Jingwei Li, Xinjie Li, Xiaodong Hou, Zhensong Zhang, Zhihong School of Informatics Xiamen University Xiamen China School of Electrical and electronic engineering North China Electric Power University Beijing China Department of Computer Science The University of Hong Kong Hong Kong China

Intrinsic and environmental factors contribute to variability in the performance of cells within a battery pack, affecting the lifespan and safety of battery systems. To solve this problem, active and passive equalization methods are proposed. However, existing passive equalization methods suffer from energy loss and low efficiency among batteries, while existing active equalization methods necessitate complex expert knowledge and control algorithms. We propose an active equalization model that leverages a generative model (GM) to assist in pattern selection for a reinforcement learning (RL) scheme, tailored for Dynamic Reconfigurable Battery (DRB) systems. The proposed model overcomes the pattern selection challenge in large-scale discrete action spaces by employing a Variational Autoencoder (VAE) for dimensionality reduction and latent space mapping, actively balancing DRB systems. Moreover, the use of pattern subgraphs diminishes dependence on expert knowledge, enabling the model to recognize structural information and adjust the system's stability. The experimental setup adheres to the laws of physics and tests the model's functionality on a simulation system. Results show that the proposed Generative Model-based Reinforcement Learning (GMRL) approach effectively addresses decision-making challenges in large-scale spaces. It can learn the structured features of the battery network, thus balancing the energy storage system and maximizing discharge efficiency gains. © 2024

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Green Machine Learning Approach for QoS Improvement in Cellular Communications 2

Green Machine Learning Approach for QoS Improvement in Cellu...

引用

2nd IEEE International Maghreb Meeting of the Conference on Sciences and Techniques of Automatic Control and computer engineering, MI-STA 2022

作者： Saeed, Mamoon M. Saeed, Rashid A. Azim, Mohammad Abdul Ali, Elmustafa Sayed Mokhtar, Rania A. Khalifa, Othman Faculty of Engineering Communications and Electronics Engineering Department Yemen Taif University College of Computers and Information Technology Department of Computer Engineering Taif21944 Saudi Arabia Asian University for Women Department of Computer Science Bangladesh Department of Electrical Electronic Engineering Port Sudan Sudan College of Electronics Engineering Khartoum Sudan International Islamic University Malaysia Faculty of Engineering Dept of Electrical and Electronics Engineering Malaysia

ISBN: (数字)9781665479189

ISBN: (纸本)9781665479189

Green cellular communications are becoming an important approach due to large-scale and complex radio networks. Due to the dynamic cellular network behaviors related to interference distribution, traffic bottlenecks, congestion points, and hotspots, there is a need to evaluate the dynamic processes in cellular systems in addition to ensuring spectrum availability. The delay, loss rate, and SNR are the most issues that may affect cellular communication performance. Artificial intelligent algorithms such as machine learning (ML) enable to detection of the dynamics in cellular networks by analyzing the complex cellular network processes and evaluating the spectrum and links qualities. It enables the extraction of spectrum knowledge from the network autonomously. The extracted information helps to know about every dynamic change in wireless parameters, related to frequency, modulation, route selection, etc. This paper provides details about the use of ML in green cellular networks to efficiently upgrade the communications and enhances different related approaches including quality of services (QoS), signal traffic load, and energy efficiency, which are critical issues of green cellular communication paradigms. The paper also presents the technical concept of green ML approaches to solve significant problems in cellular communications, in addition to future aspects and considerations for energy consumption minimization using the green ML approach in cellular radio communications. © 2022 IEEE.

关键词： Energy efficiency

来源：评论

学校读者我要写书评

暂无评论

Enhancing Robotic Arm Activity Recognition with Vision Transformers and Wavelet-Transformed Channel State Information 35

Enhancing Robotic Arm Activity Recognition with Vision Trans...

引用

35th IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, PIMRC 2024

作者： Zandi, Rojin Behzad, Kian Motamedi, Elaheh Salehinejad, Hojjat Siami, Milad Northeastern University Department of Electrical and Computer Engineering BostonMA United States Mayo Clinic Kern Center for the Science of Healthcare Delivery RochesterMN United States Mayo Clinic Department of Artificial Intelligence and Informatics RochesterMN United States

ISBN: (纸本)9798350362244

Vision-based methods are commonly used in robotic arm activity recognition. These approaches typically rely on line-of-sight (LoS) and raise privacy concerns, particularly in smart home applications. Passive Wi-Fi sensing represents a new paradigm for recognizing human and robotic arm activities, utilizing channel state information (CSI) measurements to identify activities in indoor environments. In this paper, a novel machine learning approach based on discrete wavelet transform and vision transformers for robotic arm activity recognition from CSI measurements in indoor settings is proposed. This method outperforms convolutional neural network (CNN) and long short-term memory (LSTM) models in robotic arm activity recognition, particularly when LoS is obstructed by barriers, without relying on external or internal sensors or visual aids. Experiments are conducted using four different data collection scenarios and four different robotic arm activities. Performance results demonstrate that wavelet transform can significantly enhance the accuracy of visual transformer networks in robotic arms activity recognition. © 2024 IEEE.

关键词： Robotic arms

来源：评论

学校读者我要写书评

暂无评论

Performance Evaluation of Multi-Contrast Dixon MRI and CT for Abdominal Fat and Muscle Segmentation Using a UNet CNN

Performance Evaluation of Multi-Contrast Dixon MRI and CT fo...

引用

2024 IEEE International Conference on Big Data, BigData 2024

作者： Heller, Andrew Liu, Andrew Chang, Lin-Ching Cohen, Gregg Jones, Elizabeth C. Hsu, Li-Yueh National Institutes of Health Clinical Center BethesdaMD United States Catholic University of America Department of Electrical Engineering and Computer Science WashingtonDC United States New Jersey Institute of Technology Department of Informatics NewarkNJ United States

ISBN: (纸本)9798350362480

We evaluate the performance of a deep learning framework for segmenting abdominal fat and muscle using multi-contrast Dixon magnetic resonance (MR) and computed tomography (CT) images. We aim to compare MR image segmentation by testing Dixon fat-only, water-only, and combination of both types of images and comparing the results with CT images. Nineteen subjects underwent abdominal CT and Dixon MR imaging on the same day. For each participant, three pairs of matched axial images from both CT and MR were selected at the intervertebral levels of L2-L3, L3-L4, and L4-L5 for analysis. References labels were generated through semi-automated segmentation of subcutaneous adipose tissue, visceral adipose tissue, and muscle areas. They were then used to train and evaluate a U-Net based Convolutional Neural Network (CNN) framework with a 3-fold cross-validation to compare the segmentation performance across CT, Dixon fat-only and water-only MR images. Combining the fat-only and water-only MR image inputs produced superior results in all labels. Our study demonstrates that CNN-based segmentation performance for abdominal fat and muscle improves with the inclusion of additional input channels, such as combining Dixon fat-only and water-only MR images. While CT results represent the gold standard in abdominal image segmentation, increasing the number of input image channels used in MR segmentation can approach, and even match, the results of CT. © 2024 IEEE.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Machine Learning Techniques for Brain Tumor Classification: A CNN-SVM Approach

Machine Learning Techniques for Brain Tumor Classification: ...

引用

International Conference on Innovations in Science, engineering and Technology (ICISET)

作者： Tanvir Hayat Md Saiful Islam Mosrur Hossain Nowshad Hasan Masud Parvez Md. Jiabul Hoque Department of Electronic and Telecommunication Engineering Chittagong University of Engineering and Technology Chittagong Bangladesh Department of Computer and Communication Engineering International Islamic University Chittagong Chittagong Bangladesh

ISBN: (数字)9798350355499

ISBN: (纸本)9798350355505

Brain tumors are the tenth most common type of tumor affecting people of all ages and are a leading cause of death in humans. However, early detection significantly enhances treatability. Classification of brain tumors typically relies on biopsy, a procedure often deferred until definitive brain surgery. Developing image classification techniques for tumor disorders is crucial to minimize errors in manual diagnoses by radiologists and to accelerate treatment. Advancements in machine learning (ML) offer a promising approach for assisting radiologists in diagnosing tumors using non-invasive magnetic resonance imaging (MRI). This research introduces a network based on feature extraction, where features are derived using a Convolutional Neural Network (CNN) and subsequently classified with a Support Vector Machine (SVM). The process involves transforming the multidimensional feature maps into a 2 D array where each row represents the features of a single image. The proposed method was evaluated using MRI brain images of three types of tumors: pituitary, meningioma, and glioma. The CNN-SVM method attained an accuracy of 98.75%, surpassing the CNN-RF (Random Forest) method, which achieved an accuracy of 96.25%. This result surpasses the performance of many other models. The effectiveness of the proposed approaches is analysed based on different metrics and outcomes compared to various method.

关键词： Measurement Support vector machines Accuracy Magnetic resonance imaging Brain tumors Feature extraction Brain modeling Convolutional neural networks Random forests Image classification

来源：评论

学校读者我要写书评

暂无评论

Analysis of Virtual Reality Based on the Internet of Things on Human Psychology ‘Internet of Thoughts’ (IoThs) for Rich Content Extraction Applied Natural Language Processing and Deep Learning 20th

Analysis of Virtual Reality Based on the Internet of Things ...

引用

20th International Conference on Smart Living and Public Health, ICOST 2023

作者： Mah, Pascal Muam Skalna, Iwona Pelech-Pilichowski, Tomasz Derlecki, Tomasz Nasr, Mahmoud Munyeshuri, Eric Amuzang, Gilly Njoh Itoe, Micheal Blake Somaah Tah, Ning Frida Department of Technical Informatics and Telecommunications AGH University of Science and Technology Krakow Poland Department of Business Informatics and Management Engineering AGH University of Science and Technology Krakow Poland Institute of Computer Science AGH University of Science and Technology Krakow Poland Department of Economics and Management Industry AGH University of Science and Technology Krakow Poland Department of Biomedical Engineering AGH University of Science and Technology Krakow Poland Department of Engineering Mathematics and Physics Faculty of Engineering and Technology Future University Cairo Egypt Department of Production Engineering and Logistics Enterprise Silesian University of Technology Gliwice Poland Department of Logistics Management Silesian University of Technology Gliwice Poland

ISBN: (纸本)9783031439490

In the past decade, a lot of challenges to access, assess, and to acquire the needed technological opportunities to teach computers what naturally comes from the human brain and to understand how we naturally react when we rely on technology. The ability to document human thoughts, reactions and behavior to computers has led to the coming of NLP, AI, Dl, & ML. Aim to understand the influence of IoT on humans with the use of DL to achieve content correctness and accuracy with virtual technology. Studies show that the way we think, react, and do the things we think "Internet of thoughts" reflect our personality. The way we think determines the way we react and the way we do things are based on how we think. Technology advancement has reinforced a lot of changes in humans which makes humans vulnerable to personal content exposure misappropriation due to the continuously changing nature of humanity and language. The study uses NLP, DL and behavior-oriented drive and influential function and results show that IoT based on VR influences human psychology "Internet of Thoughts". © 2023, The Author(s).

关键词： Internet of things

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：