检索结果-内蒙古大学图书馆

IAENG International Journal of computer science 2025年第3期52卷 533-545页

作者： Amalia, Amalia Lydia, Maya Silvi Muchtar, Muhammad Anggia Manik, Fuzy Yustika Sinu Gunawan, Dani Department of Computer Science Universitas Sumatera Utara Medan Indonesia Department of Information Technology Universitas Sumatera Utara Medan Indonesia

Automating the grading of short answers in Indonesian presents unique challenges, primarily due to the inherent variability in student responses and the limited linguistic resources available for fine-tuning models. This study aims to develop an automated grading system for short answers by employing a modified BERT model tailored explicitly for the Indonesian language. Our methodology involves translating the Stanford Question Answering Dataset (SQuAD 2.0) and augmenting it with domain-specific content from lecture materials, resulting in an additional 1,000 question-answer pairs. Furthermore, we optimize the model using the Optuna library for hyperparameter tuning. Experimental results show a minimum loss of 1.81 for the model, with optimized hyperparameters including a learning rate of 3.36 × 10-5, a training batch size of 8 per device, and two training epochs. The model achieves an F1-score of 69%, demonstrating a satisfactory level of accuracy comparable to or slightly exceeding typical performances on SQuAD 2.0 in English, which averages around 66%. Evaluations were conducted across various scenarios, including short-answer, long-context, and long-answer assessments. The results revealed that short answers achieved the highest cosine similarity and a perfect QWK score of 1. In contrast, long-context and long-answer scenarios yielded a QWK of 0.5, indicating a need for further improvement;however, the model shows promise for long answers with additional enhancements. This study demonstrates BERT’s potential to enhance equity and precision in grading Indonesian short answers, despite limitations in response completeness. © (2025), (International Association of Engineers). All rights reserved.

关键词： Linguistics

来源：评论

学校读者我要写书评

暂无评论

Explainable ensemble deep learning-based model for brain tumor detection and classification

引用

Neural Computing and Applications 2025年第3期37卷 1289-1306页

作者： Hosny, Khalid M. Mohammed, Mahmoud A. Salama, Rania A. Elshewey, Ahmed M. Department of Information Technology Faculty of Computers and Information Zagazig University P.O.BOX 44519 Zagazig Egypt Department of Computer Science Faculty of Computers and Information Suez University P.O.BOX 43221 Suez Egypt Department of Information Technology Faculty of Computers and Information Suez University P.O.BOX 43221 Suez Egypt

Brain tumors are very dangerous as they cause death. A lot of people die every year because of brain tumors. Therefore, accurate classification and detection in the early stages can help in recovery. Various deep learning techniques have achieved good results in brain tumor classification. The traditional deep learning methods and training the neural network from scratch are time-consuming and can last for weeks of training. Therefore, in this work, we proposed an ensemble approach depending on transfer learning that utilizes pre-trained models of DenseNet121 and InceptionV3 to detect three forms of brain tumors: meningioma, glioma, and pituitary. While developing the ensemble model, some changes were made to the architecture of pre-trained models by replacing their classifiers (fully connected and SoftMax layers) with a new classifier to adopt the recent task. In addition, gradient-weighted class activation maps (Grad-CAM) are an explainable model to verify results and achieve high confidence. The suggested model was validated using a publicly available dataset and achieved 99.02% accuracy, 98.75% precision, 98.98% recall, and a 98.86% F1 score. The suggested approach outperformed others in detecting and classifying brain tumor MRI data, and verifying results using the explainable model achieved a high degree of trust. © The Author(s) 2024.

关键词： Transfer learning

来源：评论

学校读者我要写书评

暂无评论

Graph-Segmenter:graph transformer with boundary-aware attention for semantic segmentation

引用

Frontiers of computer science 2024年第5期18卷 97-108页

作者： Zizhang WU Yuanzhu GAN Tianhao XU Fan WANG Computer Vision Perception Department of ZongMu Technology Shanghai 201203China Faculty of Electrical Engineering Information TechnologyPhysicsTechnical University of BraunschweigBraunschweig 38106Germany

Thetransformer-based semantic segmentation approaches,which divide the image into different regions by sliding windows and model the relation inside each window,have achieved outstanding ***,since the relation modeling between windows was not the primary emphasis of previous work,it was not fully *** address this issue,we propose a Graph-Segmenter,including a graph transformer and a boundary-aware attention module,which is an effective network for simultaneously modeling the more profound relation between windows in a global view and various pixels inside each window as a local one,and for substantial low-cost boundary ***,we treat every window and pixel inside the window as nodes to construct graphs for both views and devise the graph *** introduced boundary-awareattentionmoduleoptimizes theedge information of the target objects by modeling the relationship between the pixel on the object's *** experiments on three widely used semantic segmentation datasets(Cityscapes,ADE-20k and PASCAL Context)demonstrate that our proposed network,a Graph Transformer with Boundary-aware Attention,can achieve state-of-the-art segmentation performance.

关键词： graph transformer graph relation network boundary-aware attention semantic segmentation

来源：评论

学校读者我要写书评

暂无评论

Classifying Psychiatric Patients Using Machine Learning 25

Classifying Psychiatric Patients Using Machine Learning

引用

25th International Arab Conference on information technology, ACIT 2024

作者： Alrjoob, Hmmam M. Alazaidah, Raed Batyha, Radwan Khafajeh, Hayel Elsoud, Esraa Abu Saeb Al-Sherideh, Ala'a Elrashidi, Ali Alqaraleh, Muhyeeddin Amer, Waleed Zarqa University Faculty of Information Technology Department of Computer Science Zarqa Jordan Zarqa University Faculty of Information Technology Department of Data Science and Ai Zarqa Jordan Applied Science Private University Faculty of Information Technology Department of Computer Science Amman Jordan Zarqa University Faculty of Information Technology Department Computer Science Zarqa Jordan Zarqa University Faculty of Information Technology Department of Cyper Security Zarqa Jordan University of Business and Technology Department of Electrical Engineering Jeddah21432 Saudi Arabia Zarqa University Faculty of Information Technology Department of Software Engineering Zarqa Jordan Zarqa University Faculty on Information Technology Department of Cyber Security Zarqa Jordan

ISBN: (纸本)9798331540012

Psychological disorders are considered chronic illnesses that affect a wide range of populations. Some studies in the United States indicate that one in every eight individuals is affected by a psychological disorder. Additionally, individuals with a psychological disorder often have an impact on the nature of the lives of those around them. It is beneficial to detect the illness in its early stages to increase the likelihood of recovery. In this research, thirteen classification algorithms are used. Each algorithm has an interesting approach. It can be used to analyze several datasets where three datasets specific to Psychiatric patients were selected. The datasets have been chosen from UCI Machine Learning Repository. The datasets were re-analyzed after only 50% of the characteristics were approved after being evaluated by the several feature selection techniques. The main objective of this research is to find the best classifier and feature selection strategy in analyzing datasets of Psychiatric patients. After conducting numerous comparisons, it became evident that, the best classifier for the Autism-Adult-Data dataset are Bagging and Random Forest. As for the best learning strategy, it was found to be Functions and Trees, the best feature selection was obtained using Symmetrical Uncourt Attribute Evaluation. Both IBK and KStar classifier proved to be the best classifier for the Parkinson's dataset, Lazy demonstrated to be the most effective learning strategy, Correlation Attribute Evaluation technique yielded the best feature selection. Logistic emerged as the best classifier for planning-relax dataset, the best learning strategy was Bayes, the best feature selection was found to be Symmetrical Uncert Feature Selection. © 2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

SLOG: smartphone-based semantic lifelogging framework for digital prosthetic memory development

引用

Personal and Ubiquitous Computing 2023年第6期27卷 2071-2102页

作者： Ali, Shaukat Khusro, Shah Khan, Akif Khan, Inayat Alam, Iftikhar Solehria, Salman Faiz Department of Computer Science University of Peshawar Peshawar25120 Pakistan Department of Computer Science Sarhad University of Information Technology Peshawar25000 Pakistan Department of Computer Science University of Engineering and Technology Mardan Mardan23200 Pakistan

The advancements in sensors, processing, storage and networking technologies have turned smartphone into de facto lifelogging device. Realizing the lifelogging potential of smartphone, researchers have postulated with smartphone-based lifelogging systems with varying capabilities of capturing and storing information about peoples’ daily life activities, behaviors, interactions, locations, environments, etc., into their lifelog archives. However, the systems are lacking with several capturing, storing, retrieving and design issues, importantly not organizing the lifelog information with the same semantics as they exist in the real world and encoded in human episodic memory. The semantic organization of lifelog information will develop a true digital prosthetic memory to solve problems in different real-world domains including memory augmentation. In this paper, we have presented the architecture and implementation of Semantic LifelOGging (SLOG) framework for smartphone-based semantic lifelogging using Semantic Web technologies. The SLOG framework incorporates smartphone sensors and processing capabilities, and Semantic Web technologies to develop a semantically enriched digital prosthetic memory on smartphone. The semantic lifelog data model is developed by extending SmartOntoSensor ontology with relevant concepts and properties to represent, store and relate lifelog and contextual information. The semantic lifelog data model will provide contextual semantics to organize, annotate and relate lifelog information like human episodic memory and provide enough contextual cues to recall lifelog information like associative recall in human memory. The proof-of-concept applications are developed to show practicality of the SLOG framework. The quantitative and empirical evaluations have shown the SLOG framework and system has improved features and is acceptable in usability and effectiveness. © 2022, The Author(s), under exclusive licence to Springer-Verlag London Ltd., part

关键词： Smartphones

来源：评论

学校读者我要写书评

暂无评论

Vehicle Detection in Challenging Scenes Using CenterNet Based Approach

引用

computers, Materials & Continua 2023年第2期74卷 3647-3661页

作者： Ayesha Muhammad Javed Iqbal Iftikhar Ahmad Madini OAlassafi Ahmed SAlfakeeh Ahmed Alhomoud Department of Computer Science University of Engineering and Technology Taxila47050Pakistan Faculty of Computing and Information Technology King Abdulaziz University Jeddah21589Saudi Arabia Faculty of Computing and Information Technology Northern Border UniversityRafhaSaudi Arabia

Contemporarily numerous analysts labored in the field of Vehicle detection which improves Intelligent Transport system(ITS)and reduces road *** major obstacles in automatic detection of tiny vehicles are due to occlusion,environmental conditions,illumination,view angles and variation in size of *** research centers on tiny and partially occluded vehicle detection and identification in challenging scene specifically in crowed *** this paper we present comprehensive methodology of tiny vehicle detection using Deep Neural Networks(DNN)namely *** DNN disregards objects that are small in size 5 pixels and more false positives likely to happen in crowded *** there are two categories of deep learning models single-step and two-step.A single forward pass model is the one in which detection is performed directly to possible location over dense sampling,wherein two-step models incorporated by Region proposals followed by object *** in this research scrutinize one-step State of the art(SOTA)model CenteNet as proposed recently with three different feature extractor ResNet-50,HourGlass-104 and ResNet-101 one by *** train our model on challenging KITTI dataset which outperforms in comparison with SOTA single-step technique MSSD300∗which depicts performance improvement by 20.2%mAPandSMOKEby with 13.2%mAP *** of CenterNet can be justified through the huge improved *** performance of our model is evaluated on KITTI(Karlsruhe Institute of technology and Toyota Technological Institute)benchmark dataset with different backbones such as ResNet-50 gives 62.3%mAP ResNet-10182.5%mAP,last but not the least HourGlass-104 outperforms with 98.2%mAP CenterNet-HourGlass-104 achieved high mAP among above mentioned feature *** also compare our model with other SOTA techniques.

关键词： CenteNet SOTA object detection deep learning MSSD SMOKE DNN KITTI

来源：评论

学校读者我要写书评

暂无评论

Enhancing License Plate Recognition in Foggy Condition Base on Yolov8 and Dark Channel Prior

Enhancing License Plate Recognition in Foggy Condition Base ...

引用

2024 International of Seminar on Application for technology of information and Communication, iSemantic 2024

作者： Arafat Andono, Pulung Nurtantio Syukur, Abdul Affandy, Affandy Soeleman, Moch Arief Department of Computer Science Semarang Indonesia Banjarmasin Faculty of Information Technology Semarang Indonesia

ISBN: (纸本)9798350366648

This paper proposes a novel approach that combines the YOLOv8 object detection model with the Dark Channel Prior (DCP) algorithm to improve car license plate detection in foggy weather conditions. The proposed approach consists of two stages: the first stage is data preprocessing with the DCP approach, and the second stage is image preprocessing. The DCP model is trained to detect the license plates in the enhanced images, providing the final output of the proposed method. The experimental findings show that the suggested strategy significantly boosts detection accuracy and improves license plate visibility compared to conventional methods. Moreover, the integration of DCP with Yolov8 resulted in robust performance against fog, achieving an average precision of 0.975. © 2024 IEEE.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

A Service Level Agreement Aware Online Algorithm for Virtual Machine Migration

引用

computers, Materials & Continua 2023年第1期74卷 279-291页

作者： Iftikhar Ahmad Ambreen Shahnaz Muhammad Asfand-e-Yar Wajeeha Khalil Yasmin Bano Department of Computer Science and Information Technology University of Engineering and TechnologyPeshawar25120Pakistan Department of Computer Science Shaheed Benazir Bhutto Women UniversityPeshawar25000Pakistan Center of Excellence in Artificial Intelligence Department of Computer ScienceBahria UniversityIslamabadPakistan

The demand for cloud computing has increased manifold in the recent *** specifically,on-demand computing has seen a rapid rise as organizations rely mostly on cloud service providers for their day-to-day computing *** cloud service provider fulfills different user requirements using virtualization-where a single physical machine can host multiple *** virtualmachine potentially represents a different user environment such as operating system,programming environment,and ***,these cloud services use a large amount of electrical energy and produce greenhouse *** reduce the electricity cost and greenhouse gases,energy efficient algorithms must be *** specific area where energy efficient algorithms are required is virtual machine *** virtualmachine consolidation,the objective is to utilize the minimumpossible number of hosts to accommodate the required virtual machines,keeping in mind the service level agreement *** research work formulates the virtual machine migration as an online problem and develops optimal offline and online algorithms for the single host virtual machine migration problem under a service level agreement constraint for an over-utilized *** online algorithm is analyzed using a competitive analysis *** addition,an experimental analysis of the proposed algorithm on real-world data is conducted to showcase the improved performance of the proposed algorithm against the benchmark *** proposed online algorithm consumed 25%less energy and performed 43%fewer migrations than the benchmark algorithms.

关键词： Cloud computing green computing online algorithms virtual machine migration

来源：评论

学校读者我要写书评

暂无评论

Exploring Knowledge Landscapes: Clustering and Topic Modeling of Sebha University Scientific Publications 4

Exploring Knowledge Landscapes: Clustering and Topic Modelin...

引用

4th IEEE International Maghreb Meeting of the Conference on sciences and Techniques of Automatic Control and computer Engineering, MI-STA 2024

作者： Abdelhamed, Roqyia Essgaer, Mansour Sebha University Faculty of Science Department of Computer Sciences Sebha Libya Sebha University Faculty of Information Technology Department of Information Technology Sebha Libya

ISBN: (纸本)9798350372632

The continuous and exponential growth in scientific publications poses a challenge in manually sorting and searching for specific topics within a university's publications. Recent advancements in artificial intelligence, natural language processing, and clustering methods, particularly the K-Means, have shown satisfactory performance in handling large volumes of textual data. Furthermore, topic modeling methods, notably praised for their effectiveness in organizing textual documents. This study focuses on evaluating the performance of the K-Means and Latent Dirichlet Allocation in clustering Sebha University scientific publications. Different term weighting is employed namely term frequency inverse document frequency, and Word2Vec to represent the tokens. K-Means demonstrated notable performance when coupled with Word2vec, as indicated by metrics such as the Davies-Bouldin Index, Silhouette Score, Calinski-Harabasz Index, and Inertia. The best performance was observed when using term frequency inverse document frequency with the Entropy metric. The evaluation metrics for K-Means did not unanimously agree on the optimal number of clusters. In contrast, Latent Dirichlet Allocation consistently performed well, with all measurements agreeing on the best results with the smallest number of clusters. Among the various combinations tested, Latent Dirichlet Allocation paired with Word2vec produced the best results across metrics, demonstrating superior cohesion strength through the topic coherence metric and minimizing noise aggregates based on the entropy metric. © 2024 IEEE.

关键词： Statistics

来源：评论

学校读者我要写书评

暂无评论

Concept Drift Analysis and Malware Attack Detection system Using Secure Adaptive Windowing

引用

computers, Materials & Continua 2023年第5期75卷 3743-3759页

作者： Emad Alsuwat Suhare Solaiman Hatim Alsuwat Department of Computer Science College of Computers and Information TechnologyTaif UniversityTaif26571Saudi Arabia Department of Computer Science College of Computer and Information SystemsUmm Al-Qura UniversityMakkah24382Saudi Arabia

Concept drift is a main security issue that has to be resolved since it presents a significant barrier to the deployment of machine learning(ML)*** to attackers’(and/or benign equivalents’)dynamic behavior changes,testing data distribution frequently diverges from original training data over time,resulting in substantial model *** to their dispersed and dynamic nature,distributed denial-of-service attacks pose a danger to cybersecurity,resulting in attacks with serious consequences for users and *** paper proposes a novel design for concept drift analysis and detection of malware attacks like Distributed Denial of Service(DDOS)in the *** goal of this architecture combination is to accurately represent data and create an effective cyber security prediction *** intrusion detection system and concept drift of the network has been analyzed using secure adaptive windowing with website data authentication protocol(SAW_WDA).The network has been analyzed by authentication protocol to avoid malware *** data of network users will be collected and classified using multilayer perceptron gradient decision tree(MLPGDT)*** on the classification output,the decision for the detection of attackers and authorized users will be *** experimental results show output based on intrusion detection and concept drift analysis systems in terms of throughput,end-end delay,network security,network concept drift,and results based on classification with regard to accuracy,memory,and precision and F-1 score.

关键词： Concept drift machine learning DDOS cyber security SAW_WDA MLPGDT

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：