检索结果-内蒙古大学图书馆

29th International computer Conference, computer Society of Iran, CSICC 2025

作者： Mohammadvand, Reza Mozayani, Naser Dept. Computer Engineering Iran University of Science and Technology Tehran Iran

ISBN: (纸本)9798331523114

Given the rapid advancements in technologies such as the metaverse and digital twins, the need for effective integration of communication, computation, and storage in complex systems and edge computing environments is becoming increasingly apparent. This paper proposes an architecture to support an industrial metaverse based on digital twins that optimizes resources by leveraging mobile edge computing and ultra-reliable low-latency communication. This architecture reduces latency by employing task offloading and storage methods on nearby edge servers, meeting the precise requirements of future metaverses in terms of reliability and latency minimization. The proposed model, utilizing reinforcement learning algorithms such as Deep Q-Network (DQN), Double Deep Q-Network (DDQN), and Dueling Double Deep Q-Network (Dueling DDQN), is capable of achieving adaptability in dynamic conditions and making intelligent decisions. The simulation results demonstrate that the proposed method reduces by more than 10% on average compared to the best method available in the literature and improves resource utilization efficiency. © 2025 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Detecting adversarial examples using constraint-based learning

Detecting adversarial examples using constraint-based learni...

引用

2024 International Conference on Smart Devices, ICSD 2024

作者： Kumar, Vivek Kumar, Kamal Singh, Maheep National Institute of Technology Dept. of Computer Science -and Engineering Uttarakhand Pauri India IGDTUW Department of Information Technology New Delhi India Doon University Dept. of Computer Science and Engineering Dehradun India

ISBN: (纸本)9798350394412

Deep learning models have been applied in various fields and continue producing exceptional results. However, these models are vulnerable to adversarial attacks, which are modified data samples maliciously crafted to cause misclassification by a target model. Fields like Network intrusion detection systems (NIDS) are security-critical;therefore, adversarial attacks pose a high risk in such applications. The attacker not only aims for evasion but also tries to launch an attack once the adversarial example has evaded detection by a NIDS. This paper presents a novel technique for detecting adversarial examples targeted towards a NIDS by learning the dependencies between features known as domain constraints. Our proposed method detects adversarial examples that violate these constraints learned during training. Compared with prevalent detection techniques, our method does not require generating adversarial examples for training, which reduces its computational requirement. We assume a grey-box threat model where the attacker can access the training data. We perform experiments on the CICIDS2017 dataset against five state-of-the-art adversarial attacks: FGSM, BIM, PGD, Deep fool and BIM. The results show that our method can detect up to 100% adversarial attacks, outperforming the contemporary schemes. © 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

Enhancing Programming Language Models for C++ Code Clone Detection 10

Enhancing Programming Language Models for C++ Code Clone Det...

引用

10th International Conference on Platform Technology and Service, PlatCon 2024

作者： Kang, Hyonjun Chun, Haeun Kim, Mucheol School of Computer Sceince and Engineering Chung-Ang University Seoul Korea Republic of Dept. of Computer Science and Engineering Chung-Ang University Seoul Korea Republic of

ISBN: (纸本)9798350367874

This paper examines the fine-tuning methods of two Code Language Models, namely Code-Reviewer and GraphCodeBERT, with the objective of enhancing code clone detection in C++ code. Two experiments are conducted: sentence embeddings and sentence classification with the BM25L algorithm for enhanced pair selection. The Euclidean distance method demonstrated superior performance among the five similarity methods in the sentence embedding comparison. The combination of sentence classification with BM25L as a sentence pairing algorithm achieves superior performance, indicating a promising avenue for further development. The results show that GraphCodeBERT outperforms Code-Reviewer in every task, and sentence classification outperforms sentence embedding in a large margin. This paper highlights the potential of these models for clone detection and emphasizes the benefits of methodological enhancements for improved accuracy, providing valuable insights for software developers and future research. © 2024 IEEE.

关键词： C++ (programming language)

来源：评论

学校读者我要写书评

暂无评论

Breast Cancer Diagnosis Using a Machine Learning Model and Swarm Intelligence Approach 1

Breast Cancer Diagnosis Using a Machine Learning Model and S...

引用

1st International Conference in Advanced Innovation on Smart City, ICAISC 2023

作者： Gad, Ibrahim Elmezain, Mahmoud Alwateer, Majed M. Almaliki, Malik Elmarhomy, Ghada Atlam, Elsayed Tanta University Computer Science Division Faculty of Science Tanta31527 Egypt Taibah University College of Computer Science and Engineering Dept. of Computer Science Yanbu966144 Saudi Arabia

ISBN: (纸本)9781665472753

The features selection for machine learning models requires careful consideration. A good selection of features can enable machine learning models to better identify patterns in data and make more accurate predictions. Also, relevant features in the data have an impact on the accuracy of the model and lengthen the training process. In this study, the suggested feature selection strategy is implemented using pigeon inspired optimizer (PIO). The PIO is a continuous swarm intelligent algorithm. The machine learning models were trained and tested using the proposed PIO optimizer in the context of medical data. Both the training and testing steps use the Wisconsin breast cancer dataset. The best results are generated by the Random Forest model, which has accuracy, F-score, recall, and precision values of 97.2%, 97.3%, 97.3%, and 97.3%, respectively. It was concluded that the selected features perform better for classification than the original high-dimensional features, both in terms of accuracy and the F-score. As a consequence of this, the proposed approach can be utilized to better categorize breast cancer data. © 2023 IEEE.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

EBV: Electronic Bee-Veterinarian for Principled Mining and Forecasting of Honeybee Time Series

EBV: Electronic Bee-Veterinarian for Principled Mining and F...

引用

2024 SIAM International Conference on Data Mining, SDM 2024

作者： Hossain, Shamima Faloutsos, Christos Baer, Boris Kim, Hyoseung Tsotras, Vassilis J. Dept. of Computer Science and Engineering University of California Riverside United States School of Computer Science Carnegie Mellon University United States Dept. of Entomology University of California Riverside United States Dept. of Electrical and Computer Engineering University of California Riverside United States

ISBN: (纸本)9781611978032

Honeybees are vital for pollination and food production. Among many factors, extreme temperature (e.g., due to climate change) is particularly dangerous for bee health. Anticipating such extremities would allow beekeepers to take early preventive action. Thus, given sensor (temperature) time series data from beehives, how can we find patterns and do forecasting? Forecasting is crucial as it helps spot unexpected behavior and thus issue warnings to the beekeepers. In that case, what are the right models for forecasting? ARIMA, RNNs, or something else? We propose the EBV (Electronic Bee-Veterinarian) method, which has the following desirable properties: (i) principled: it is based on a) diffusion equations from physics and b) control theory for feedback-loop controllers;(ii) effective: it works well on multiple, real-world time sequences, (iii) explainable: it needs only a handful of parameters (e.g., bee strength) that beekeepers can easily understand and trust, and (iv) scalable: it performs linearly in time. We applied our method to multiple real-world time sequences, and found that it yields accurate forecasting (up to 49% improvement in RMSE compared to baselines), and segmentation. Specifically, discontinuities detected by EBV mostly coincide with domain expert's opinions, showcasing our approach's potential and practical feasibility. Moreover, EBV is scalable and fast, taking about 20 minutes on a stock laptop for reconstructing two months of sensor data. Copyright © 2024 by SIAM.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Algorithms for Pneumonia: A Comparative Approach to Classification and Segmentation 6

Deep Learning Algorithms for Pneumonia: A Comparative Approa...

引用

6th International Conference on Cybernetics, Cognition and Machine Learning Applications, ICCCMLA 2024

作者： Mishra, Vishnu Kumar Mishra, Megha Tiwari, Rajesh Diwan, Tarun Dhar Soni, Rama Patra, Raj Kumar Malla Reddy Engineering college for Women Hyderabad Jntuh Dept. of Computer Science and Engineering Hyderabad India Shri Shankaracharya Technical Campus Csvtu Department of Computer Science and Engineering Bhilai India Atal Bihari Vajpayee University Dept. of Computer Science and CoE Chhattisgarh Bilaspur India Dr. C. V. Raman University Chhattisgarh Bilaspur India CMR Engineering College Department of Computer Science and Engineering Hyderabad India

ISBN: (纸本)9798331505790

The study compares the outcomes of numerous modern CNN architectures, such as ResNet-50, VGG-16, and DenseNet-121, with conventional machine learning classifiers in order to assess each architecture's performance. Recent developments in deep learning have shown that medical image analysis, particularly the diagnosis of pneumonia, can be done with great accuracy. This work investigates different deep learning techniques for lung X-ray-based pneumonia classification. The study emphasizes the use of convolutional neural networks (CNNs) to analyze radiographic images and distinguish between pneumonia-affected lung tissue and normal lung tissue.. This underscores the significant potential of deep learning algorithms to enhance the accuracy and efficiency of pneumonia diagnosis, aiding medical professionals in making informed patient care decisions. The results indicate that deep learning methods surpass traditional machine learning techniques in classifying pneumonia, achieving approximately 9 5 % accuracy across all examined CNN models. These deep learning models are trained on an extensive dataset of annotated chest X-rays, which allows them to recognize intricate patterns and characteristics linked to pneumonia. Considering the substantial influence pneumonia has on the world's health, prompt and correct diagnosis is essential to successful management and treatment. © 2024 IEEE.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

Towards Developing a Transformer-Based Bangla Typing Error Correction Model: A Deep Learning-Based Approach 2

Towards Developing a Transformer-Based Bangla Typing Error C...

引用

2nd International Conference on Information and Communication Technology for Sustainable Development, ICICT4SD 2023

作者： Salim, Shahidul Islam, Taqiul Zannat, Rowzatul Mia, Naeem Fuad, Mohtasim Murad, Hasan Khulna University of Engineering Technology Dept. of Computer Science and Engineering Khulna9203 Bangladesh Chattogram University of Engineering Technology Dept. of Computer Science and Engineering Chattogram4349 Bangladesh

ISBN: (纸本)9798350358667

Despite being the sixth most widely spoken language globally, the Bangla language faces a lack of effective and efficient automated methods to correct keyboard errors in written text. Such errors can result in communication misunderstandings and misinterpretations, particularly in the digital domain, where written text plays a crucial role in communication and is heavily relied upon. However, no keyboard error model for the Bangla language has been found due to the lack of available datasets. In this research work, we have proposed a probabilistic approach to generate a synthetic Bangla keyboard typing error correction (BKTEC) dataset. By utilizing the novel BKTEC dataset, we have fine-tuned a transformer-based model for Keyboard error correction. After training it to convergence, the proposed model has been tested on a test dataset, and we have found a word level accuracy of 88.56%. © 2023 IEEE.

关键词： Error correction

来源：评论

学校读者我要写书评

暂无评论

Optimal Trajectory Planning of an Omni-directional Mobile Manipulator for Obstacle Avoidance

引用

Transactions of the Korean Institute of Electrical Engineers 2023年第8期72卷 952-958页

作者： Kim, Minsu Sung, Young Whee School of Electrronic Engineering Kumoh National Institute of Technolgy Korea Republic of Dept. of Electrical Engineering and Computer Science DGIST Korea Republic of

The ultimate purpose of a mobile robot is to do some tasks in addition to moving to a desired point, so it is very natural for a mobile robot to be equipped with a manipulator. Among many types of mobile robots, omni-directional mobile robot has been very popular because it has advantages in trajectory planning and control. In this paper, we deal with an omni-directional mobile manipulator that consists of a four Mecanum wheeled mobile robot platform and a manipulator mounted on it. In general, it is very difficult to solve inverse kinematics for a mobile manipulator because it has infinite solutions. Omni-directional mobile robot platform can move in any direction and can rotate on a pivot in a plane, so we propose it to be modeled as a manipulator with two prismatic joints and one revolute joint. Then the overall omni-directional mobile manipulator can be considered as a manipulator with an (n+3) DOF manipulator if the mounted manipulator has n DOF. We propose a weighted pseudo-inverse based method to solve the inverse kinematics problem at velocity level for the omni-directional mobile manipulator. In our method, in addition to a primary task of following end-effector trajectory, a secondary task of obstacle avoidance is accomplished by maximizing a performance index that represents a distance between a link of the manipulator and an obstacle. Simulation results showed the feasibility of the proposed method. © 2023 Korean Institute of Electrical Engineers. All rights reserved.

关键词： Manipulators

来源：评论

学校读者我要写书评

暂无评论

Classification of Depression, Anxiety and Stress from Handwriting and Drawing with Stacking Models 8

Classification of Depression, Anxiety and Stress from Handwr...

引用

8th International Symposium on Innovative Approaches in Smart Technologies, ISAS 2024

作者： Bayrak, Semra Golgiyaz, Sedat Aykut, Murat Karadeniz Technical University Graduate School of Natural and Applied Science Dept. of Computer Engineering Trabzon61080 Turkey Bingöl University Dept. of Computer Engineering Bingöl Turkey Karadeniz Technical University Dept. of Computer Engineering Trabzon Turkey

ISBN: (纸本)9798331540104

Recently, emotion analysis from online handwriting and drawings has become an important research topic. This study aims to determine the optimal model combination using dynamic and statistical features obtained from handwriting and drawings for detecting emotional states. The feature vectors are derived from physical (kinematic), statistical and signal processing (spectral, cepstral and frequency domain) analyses. To address the class imbalance problem, ADASYN (Adaptive Synthetic Sampling) and Tomek Links methods were applied together. In this way, the number of samples of the minority class has been increased, while over-sampled classes have been reduced. Dimensionality reduction was realized using PCA, and Gradient Boosting Classifier (GBC) feature selection. After determining the fewest and most effective features, the samples were classified using the two-level stacking ensemble method. eXtreme Gradient Boosting (XGBoost), Light Gradient-Boosting Machine (LightGBM), GBC, Random Forest (RF) methods are selected for base and meta models. Model and hyperparameter optimization was carried out by Optuna frameork. The performance of the models is evaluated through experiments in publicly available EMOTion recognition from HAndWriting and draWing (EMOTHAW) database. The aim was to determine the emotional states of depression, anxiety and stress with the data obtained from 7 different drawing/writing tasks of 129 people. The results obtained from accuracy, f1-score, precision and recall metrics show that the model provides remarkable performance. © 2024 IEEE.

关键词： Dimensionality reduction

来源：评论

学校读者我要写书评

暂无评论

Detection and Analysis of Single Blackhole Node with TCP Connection in MANETs using Machine Learning Algorithms 6

Detection and Analysis of Single Blackhole Node with TCP Con...

引用

6th International Conference on Inventive Computation Technologies, ICICT 2023

作者： Gotti, Rajeswari Polagani, Abhiram Posina, Gagana Sai Lakshmi Veerapaneni, Srihari Prasanth, Talluri Dept. of Computer Science and Engineering Seshadri Rao Gudlavalleru Engineering College Gudlavalleru India Dept. of Computer Science and Engineering Seshadri Rao Gudlavalleru Engineering College Machilipatnam India

ISBN: (纸本)9798350398496

A MANET facilitates the creation of a wireless connection between two devices without relying on any fixed infrastructure or centralized administration. Security is the main drawback in MANET's. As a result, intruder can easily connect to the network. To overcome this problem an efficient Intrusion Detection System should be built, in such a way the IDS can identify the intrusion. To detect the intruder like blackhole attack, this research study has proposed a novel technique in which a network consisting of 25 nodes is simulated and connected with TCP and UDP connection and collect different characteristics from the trace file. Now these characteristics are analyzed by using various machine learning algorithms. © 2023 IEEE.

关键词： Intrusion detection

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：