检索结果-内蒙古大学图书馆

Joint 30th International Conference on Computational Linguistics and 14th International Conference on Language Resources and Evaluation, LREC-COLING 2024

作者： Lu, Bo-Han Lin, Yi-Hsuan Lee, En-Shiun Annie Tsai, Richard Tzong-Han Department of Computer Science and Information Engineering National Central University Taiwan Department of Computer Science Faculty of Arts and Science University of Toronto Canada Computer Science Program Faculty of Science Ontario Tech University Canada Center for GIS Research Center for Humanities and Social Sciences Academia Sinica Taiwan

ISBN: (纸本)9782493814104

Machine translation focuses mainly on high-resource languages (HRLs), while low-resource languages (LRLs) like Taiwanese Hokkien are relatively under-explored. The study aims to address this gap by developing a dual translation model between Taiwanese Hokkien and both Traditional Mandarin Chinese and English. We employ a pre-trained LLaMA 2-7B model specialized in Traditional Mandarin Chinese to leverage the orthographic similarities between Taiwanese Hokkien Han and Traditional Mandarin Chinese. Our comprehensive experiments involve translation tasks across various writing systems of Taiwanese Hokkien as well as between Taiwanese Hokkien and other HRLs. We find that the use of a limited monolingual corpus still further improves the model's Taiwanese Hokkien capabilities. We then utilize our translation model to standardize all Taiwanese Hokkien writing systems into Hokkien Han, resulting in further performance improvements. Additionally, we introduce an evaluation method incorporating back-translation and GPT-4 to ensure reliable translation quality assessment even for LRLs. The study contributes to narrowing the resource gap for Taiwanese Hokkien and empirically investigates the advantages and limitations of pre-training and fine-tuning based on LLaMA 2. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Neural machine translation

来源：评论

学校读者我要写书评

暂无评论

Survey of Online Exam Proctoring Model to Detect Cheating Behavior based on Face Recognition 8

Survey of Online Exam Proctoring Model to Detect Cheating Be...

引用

8th International Conference on Systems and Informatics, ICSAI 2022

作者： Rabiha, Sucianna Ghadati Kartowisastro, Iman Herwidiana Setiawan, Reina Budiharto, Widodo Bina Nusantara University Binus Graduate Program - Computer Science. Computer Science Department Jakarta11480 Indonesia Bina Nusantara University Faculty of Engineering Computer Science Department Binus Graduate Program - Computer Science Computer Engineering Department Jakarta11480 Indonesia Bina Nusantara University School of Computer Science Computer Science Department Binus Graduate Program - Computer Science Computer Science Department Jakarta11480 Indonesia

ISBN: (纸本)9781665474481

Exams are an important component of any educational program, including online education. In any test, there is a possibility of cheating, so its detection and prevention is important. This study aims to conduct an in-depth study of the online exam monitoring model approach based on facial recognition used to detect cheating. Based on the inclusion and exclusion criteria designed, 13 selected studies were obtained. From these studies, we conducted further analysis regarding the Face Detection Method, Face Recognition Method, Initial Feature, Behavior Analysis and Evaluation Metrics used in each study so as to provide answers to research questions. the most frequently used Face detection method was Viola-Jones with a presentation of 20%, then CNN and MTCNN with a total presentation of 21%. The most widely used face recognition method in selected studies is CNN and metrics Accuracy is one of the most frequently used evaluations with a percentage of 33%. While the features that are usually used to detect cheating during online exams include facial motion and head pose which occupies the first position. The second is eye movement, then multiple faces gaze estimation and facial expression is in third place. Other features that also play a role in analyzing cheating behavior are mouth detection, facial vector, landmark location, gesture and posture. © 2022 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Retinal Blood Vessel Tracking and Diameter Estimation via Gaussian Process With Rider Optimization Algorithm

引用

IEEE Journal of Biomedical and Health Informatics 2024年第3期28卷 1173-1184页

作者： Ahmad, Nehal Lai, Kuan-Ting Tanveer, M. Indian Institute of Technology Indore Simrol Department of Mathematics Indore453552 India National Taipei University of Technology Department of International Program of Electrical Engineering and Computer Science Taipei106 Taiwan National Taipei University of Technology Department of Electronic Engineering Taipei106 Taiwan

Retinal blood vessels structure analysis is an important step in the detection of ocular diseases such as diabetic retinopathy and retinopathy of prematurity. Accurate tracking and estimation of retinal blood vessels in terms of their diameter remains a major challenge in retinal structure analysis. In this research, we develop a rider-based Gaussian approach for accurate tracking and diameter estimation of retinal blood vessels. The diameter and curvature of the blood vessel are assumed as the Gaussian processes. The features are determined for training the Gaussian process using Radon transform. The kernel hyperparameter of Gaussian processes is optimized using Rider Optimization Algorithm for evaluating the direction of the vessel. Multiple Gaussian processes are used for detecting the bifurcations and the difference in the prediction direction is quantified. The performance of the proposed Rider-based Gaussian process is evaluated with mean and standard deviation. Our method achieved high performance with the standard deviation of 0.2499 and mean average of 0.0147, which outperformed the state-of-the-art method by 6.32%. Although the proposed model outperformed the state-of-the-art method in normal blood vessels, in future research, one can include tortuous blood vessels of different retinopathy patients, which would be more challenging due to large angle variations. We used Rider-based Gaussian process for tracking blood vessels to obtain the diameter of retinal blood vessels, and the method performed well on the 'STrutred Analysis of the REtina (STARE) Database' accessed on Oct. 2020 (https://***/~ahoover/stare/). To the best of our knowledge, this experiment is one of the most recent analysis using this type of algorithm. © 2013 IEEE.

关键词： Blood vessels

来源：评论

学校读者我要写书评

暂无评论

Collaborative Service Provisioning for UAV-Assisted Mobile Edge Computing

引用

Chinese Journal of Electronics 2024年第6期33卷 1504-1514页

作者： Yuben QU Zhenhua WEI Zhen QIN Tao WU Jinghao MA Haipeng DAI Chao DONG The Key Laboratory of Dynamic Cognitive System of Electromagnetic Spectrum Space Ministry of Industry and Information Technology Colllege of Electronic and Information Engineering Nanjing University of Aeronautics and Astronautics Xi'an Research Institute of High Technology Department of Information and Communication Noncommissioned Officer Academy of PAP National University of Defense Technology Department of Computer Science and Technology Nanjing University

Unmanned aerial vehicle(UAV)-assisted mobile edge computing(MEC), as a way of coping with delaysensitive and computing-intensive tasks, is considered to be a key technology to solving the challenges of terrestrial MEC networks. In this work, we study the problem of collaborative service provisioning(CSP) for UAV-assisted MEC. Specifically, taking into account the task latency and other resource constraints, this paper investigates how to minimize the total energy consumption of all terrestrial user equipments, by jointly optimizing computing resource allocation, task offloading, UAV trajectory, and service placement. The CSP problem is a non-convex mixed integer nonlinear programming problem, owing to the complex coupling of mixed integral variables and non-convexity of CSP. To address the CSP problem, this paper proposes an alternating optimization-based solution with the convergence guarantee as follows. We iteratively deal with the joint service placement and task offloading subproblem, and UAV movement trajectory subproblem, by branch and bound and successive convex approximation, respectively,while the closed form of the optimal computation resource allocation can be efficiently obtained. Extensive simulations validate the effectiveness of the proposed algorithm compared to three baselines.

关键词： Energy consumption Multi-access edge computing Collaboration programming Autonomous aerial vehicles Minimization Trajectory Resource management Optimization Convergence

来源：评论

学校读者我要写书评

暂无评论

Semantic Segmentation and YOLO Detector over Aerial Vehicle Images

引用

computers, Materials & Continua 2024年第8期80卷 3315-3332页

作者： Asifa Mehmood Qureshi Abdul Haleem Butt Abdulwahab Alazeb Naif Al Mudawi Mohammad Alonazi Nouf Abdullah Almujally Ahmad Jalal Hui Liu Faculty of Computing and AI Air UniversityIslamabad44000Pakistan Department of Computer Science College of Computer Science and Information SystemNajran UniversityNajran55461Saudi Arabia Department of Information Systems College of Computer Engineering and SciencesPrince Sattam bin Abdulaziz UniversityAl-Kharj16273Saudi Arabia Department of Information Technology College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Cognitive Systems Lab University of BremenBremen28359Germany

Intelligent vehicle tracking and detection are crucial tasks in the realm of highway ***,vehicles come in a range of sizes,which is challenging to detect,affecting the traffic monitoring system’s overall *** learning is considered to be an efficient method for object detection in vision-based *** this paper,we proposed a vision-based vehicle detection and tracking system based on a You Look Only Once version 5(YOLOv5)detector combined with a segmentation *** model consists of six *** the first step,all the extracted traffic sequence images are subjected to pre-processing to remove noise and enhance the contrast level of the *** pre-processed images are segmented by labelling each pixel to extract the uniform regions to aid the detection phase.A single-stage detector YOLOv5 is used to detect and locate vehicles in *** detection was exposed to Speeded Up Robust Feature(SURF)feature extraction to track multiple *** on this,a unique number is assigned to each vehicle to easily locate them in the succeeding image frames by extracting them using the feature-matching ***,we implemented a Kalman filter to track multiple *** the end,the vehicle path is estimated by using the centroid points of the rectangular bounding box predicted by the tracking *** experimental results and comparison reveal that our proposed vehicle detection and tracking system outperformed other state-of-the-art *** proposed implemented system provided 94.1%detection precision for Roundabout and 96.1%detection precision for Vehicle Aerial Imaging from Drone(VAID)datasets,respectively.

关键词： Semantic segmentation YOLOv5 vehicle detection and tracking Kalman filter SURF

来源：评论

学校读者我要写书评

暂无评论

Exploring Children’s Strategies in Response to Robot’s Advice During a Group Task with iCub and Nao 16th

Exploring Children’s Strategies in Response to Robot’s A...

引用

16th International Conference on Social Robotics, ICSR + AI 2024

作者： Pusceddu, Giulia Sangineto, Mariapia Cocchella, Francesca Bogliolo, Michela Belgiovine, Giulia Lastrico, Linda Casadio, Maura Rea, Francesco Gena, Cristina Sciutti, Alessandra Robotics Brain and Cognitive Science Department Italian Institute of Technology Genoa Italy Department of Informatics Bioengineering Robotics and Systems Engineering University of Genoa Genoa Italy Department of Computer Science University of Turin Turin Italy Cognitive Architecture for Collaborative Technologies Unit Italian Institute of Technology Genoa Italy Scuola di Robotica Genoa Italy

ISBN: (纸本)9789819635184

This study investigates children’s trust in two humanoid robots, Nao and iCub, through a cooperative game designed to elicit spontaneous behaviors and group dynamics. We investigate whether participants change their choices after hearing advice from a robot, considering rounds where the advice is correct or incorrect, with the robot apologizing for its mistakes. Results reveal that, in most cases, participants prefer not to change their moves after the robot’s advice. Additionally, after the robots made mistakes, participants are less likely to actively follow their advice and more likely to reject it. No significant differences in strategies among the iCub and Nao samples are evidenced. Furthermore, we examine the influence of group dynamics, identifying "Proactive Players" who consistently initiated moves. A significant association between the strategies of this player and their teammates is found, suggesting potential influence. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Enhanced Diagnosis of Schizophrenia Using Integrated Node Embedding and Causal Discovery on Imbalanced fMRI Data 10

Enhanced Diagnosis of Schizophrenia Using Integrated Node Em...

引用

10th International Conference on Signal Processing and Intelligent Systems, ICSPIS 2024

作者： Sabzevar, Sina Masoudnia, Saeed Araabi, Babak N Nazemzadeh, MohammadReza Tavakoli, Hosna University of Tehran Department of Electrical and Computer Engineering Tehran Iran Monash University Department of Neuroscience Melbourne Australia Institute of Cognitive Science Studies Computational and Artificial Intelligence Department Tehran Iran

ISBN: (纸本)9798331532543

This study introduces a novel diagnostic method for schizophrenia using causal discovery and node embedding techniques on resting-state fMRI data. Data from 148 subjects (27 schizophrenia patients, 121 healthy controls) were analyzed across 116 brain regions, applying several causal discovery techniques to assess directional connectivity and uncover schizophrenia-related patterns. By combining advanced causal discovery with node embedding techniques and a rigorous data-balancing pipeline, our method improves sensitivity to subtle network variations critical for schizophrenia diagnosis. Our approach addresses key challenges in the literature: the difficulty of extracting effective neurobiological markers from noisy and complex fMRI data, the limited interpretability of connectivity networks, and the imbalance in datasets that often skews classification performance. Experimental results demonstrate the pipeline's effectiveness in overcoming imbalanced data challenges and extracting features from noisy and complex data, with Regression Dynamic Causal Modeling (RDCM) achieving the highest classification metrics, including 90.06% accuracy and 89.09% sensitivity. This method identified 15 brain regions with strong diagnostic potential, offering improved sensitivity and specificity for schizophrenia diagnosis Compared to contemporary domain techniques. © 2024 IEEE.

关键词： Diseases

来源：评论

学校读者我要写书评

暂无评论

Development of Character Recognition Model Inspired by Visual Explanations

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第3期5卷 1362-1372页

作者： Ralekar, Chetan Choudhary, Shubham Gandhi, Tapan Kumar Chaudhury, Santanu Department of Brain and Cognitive Sciences Massachusetts Institute of Technology CambridgeMA02139 United States Harvard John A. Paulson School of Engineering and Applied Sciences CambridgeMA02138 United States Department of Electrical Engineering Indian Institute of Technology New Delhi110016 India Department of Computer Science and Engineering Indian Institute of Technology Jodhpur342030 India

Deep neural networks (DNNs) currently constitute the best-performing artificial vision systems. However, humans are still better at recognizing many characters, especially distorted, ornamental, or calligraphic characters compared with the highly sophisticated recognitionmodels. Understanding themechanism of character recognition by humans may give some cues for building better recognition models. However, the appropriate methodological approach to using these cues has not been much explored for developing character recognition models. Therefore, this paper tries to understand the process of character recognition by humans and DNNs by generating visual explanations for their respective decisions. We have used eye tracking to assay the spatial distribution of information hotspots for humans via fixation maps. We have proposed a gradient-based method for visualizing the reasoning behind the model's decision through visualization maps and have proved that our method is better than the other class activation mapping methods. Qualitative comparison between visualization maps and fixation maps reveals that both model and humans focus on similar regions in character in the case of correctly classified characters. However, when the focused regions are different for humans and model, the characters are typically misclassified by the latter. Hence, we propose to use the fixation maps as a supervisory input to train the model that ultimately results in improved recognition performance and better generalization. As the proposedmodel gives some insights about the reasoning behind its decision, it can find applications in fields, such as surveillance and medical applications, where explainability helps to determine system fidelity. Impact Statement-Humans and DNNs rely on selective information uptake while classifying a character. This information selection strategy can be understood by visualizing the important, informative character regions that ultimately govern the decision o

关键词： Visualization

来源：评论

学校读者我要写书评

暂无评论

The Effect of Viewpoint Change Strategies on Multi-View Video and Audio Transmission QoE Over ICN/CCN 38

The Effect of Viewpoint Change Strategies on Multi-View Vide...

引用

38th IEEE Region 10 Conference, TENCON 2023

作者： Nunome, Toshiro Takada, Yoshiyuki Computer Science Program Graduate School of Engineering Nagoya Institute of Technology Department of Engineering Nagoya466-8555 Japan

ISBN: (纸本)9798350302196

In this paper, we exploit caches on intermediate nodes for QoE enhancement of multi-view video and audio transmission over ICN/CCN by controlling the content request start timing of consumers. We assume the selected single viewpoint transmission method;a consumer receives video and audio streams of a requested viewpoint. We perform a simple experiment with two consumers. When the consumers play video and audio with the time difference, we assess the effect of cached content by the former consumer's request on the output quality of the latter consumer. We deal with two types of viewpoint change strategies for the former consumer, which affect the efficiency of cache utilization. From the assessment results, we see that cache utilization has an important factor in enhancing QoE. © 2023 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

The Effect of Bitrate Control Methods of MPEG-DASH Streaming on Multi-View Video and Audio QoE

The Effect of Bitrate Control Methods of MPEG-DASH Streaming...

引用

2023 International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2023

作者： Nunome, Toshiro Wakayama, Yamato Nagoya Institute of Technology Computer Science Program Department of Engineering Graduate School of Engineering Nagoya466-8555 Japan

ISBN: (纸本)9798350324174

This paper investigates the effect of bitrate control methods on QoE of multi-view video and audio streaming with MPEG-DASH. We adopt three bitrate control methods for conventional single-view video streaming to the MVV-A system with MPEG-DASH. We conduct a subjective experiment changing available network bandwidth and investigate the effect of the methods on QoE. © 2023 IEEE.

关键词： Video streaming

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：