检索结果-内蒙古大学图书馆

ABL-SMOTE: A Novel Resampling Method by Handling Noisy and Borderline Challenge for Imbalanced Dataset for Software Defect Prediction

引用

IEEE Access 2025年 13卷 78781-78794页

作者： Bashir, Kamal Abdelwahab Ghorashi, Sara Ahmed, Ali Khader, Abdolraheem Karary University College of Computer Science and Information Technology Department of Information Technology Omdurman12304 Sudan Princess Nourah bint Abdulrahman University College of Computer and Information Sciences Department of Computer Sciences P.O. Box 84428 Riyadh11671 Saudi Arabia King Abdulaziz University Faculty of Computing and Information Technology Jeddah21589 Saudi Arabia Nanjing University of Science and Technology School of Computer Science and Engineering Nanjing210094 China

Machine learning algorithms face important implementation difficulties due to imbalanced learning since the Synthetic Minority Oversampling Technique (SMOTE) helps improve performance through the creation of new minority class examples in feature space before preprocessing. The underlying problem leading to performance deterioration emerges from noise and boundary instances found in the minority class rather than an excessive imbalance ratio. Noise and marginal samples become problematic during oversampling operations since they reduce the performance of classification models. This paper proposes an advanced version of Borderline-SMOTE (BL-SMOTE), called ABL-SMOT approach that utilizes INFFC for noise filtering to optimize data quality in the dataset. This improvement method for minority class sample placement and noise filtering ensures higher classifier performance through ABL-SMOTE. The proposed technique receives assessment through the usage of software defect datasets along with multiple testing classifiers against conventional data sampling procedures. The experimental outcomes show that ABL-SMOTE performs better than alternative approaches in most tested datasets because it is an effective preprocessing tool for imbalanced classification problems. © 2013 IEEE.

关键词： Software testing

来源：评论

学校读者我要写书评

暂无评论

Formula 1 Race Winner Prediction Using Random Forest and SHAP Analysis

Formula 1 Race Winner Prediction Using Random Forest and SHA...

引用

2025 International Conference on Control, Automation, and Instrumentation, IC2AI 2025

作者： El Haber, Elias Sawaya, Elie Attieh, Maroun Tannous, Aldo Ghazaly, Weam Owayjan, Michel American University of Science & Technology Faculty of Engineering and Computer Science Beriut Lebanon Nantes University Saint-Nazaire France

ISBN: (纸本)9798331529604

Predicting the outcomes of Formula 1 (F1) races presents a significant challenge due to the complex interplay of numerous factors, including driver skill, vehicle performance, team strategy, and unpredictable race-day conditions. This paper investigates the application of the Random Forest algorithm enhanced with SHAP (Shapley Additive Explanations) analysis to forecast race winners while providing interpretability to the model's predictions. By leveraging extensive historical data encompassing driver metrics and race conditions, we aim to build a model that achieves high predictive accuracy and offers valuable insights into the most influential factors affecting race outcomes. Our approach demonstrates how integrating machine learning with explainable AI techniques can assist F1 teams in making data-driven decisions, ultimately optimizing their strategies in the highly competitive environment of professional racing. © 2025 IEEE.

关键词： Random forests

来源：评论

学校读者我要写书评

暂无评论

Attention-Driven Emotion Recognition in EEG: A Transformer-Based Approach With Cross-Dataset Fine-Tuning

引用

IEEE Access 2025年 13卷 69369-69394页

作者： Ghous, Ghulam Najam, Shaheryar Alshehri, Mohammed Alshahrani, Abdulmonem AlQahtani, Yahya Jalal, Ahmad Liu, Hui Riphah International University Department of Electrical and Computer Engineering Islamabad44000 Pakistan Bahria University Department of Electrical Engineering Islamabad44000 Pakistan King Khalid University Department of Computer Science Abha61421 Saudi Arabia King Khalid University Department of Informatics and Computer Systems Abha61421 Saudi Arabia Air University Faculty of Computer Science and AI Islamabad44000 Pakistan Korea University College of Informatics Department of Computer Science and Engineering Seoul02841 Korea Republic of University of Bremen Cognitive Systems Laboratory Bremen28359 Germany Guodian Nanjing Automation Company Ltd. Nanjing210032 China

Emotion recognition from EEG (electroencephalogram) signals is crucial in mental health diagnostics and human-computer interaction but is often hindered by high dimensionality, noise, and complex temporal dependencies in the data. This paper presents a novel approach that integrates transformer models, attention mechanisms, and transfer learning to enhance emotion recognition accuracy from EEG signals. The proposed methodology consists of two phases: Attention Enhanced Base Model Development (AE-BMD) and Cross-Dataset Fine Tuning Adaptation (CD-FTA). In the AE-BMD phase, the base model is developed and trained on the SEED-IV dataset (15 participants, 62 EEG channels), achieving an accuracy of 84%, with an average precision of 84.75%, recall of 84% and F1-score of 84%. This phase employs optimized feature extraction from key EEG frequency bands (Delta, Theta, Alpha, Beta, Gamma) using techniques such as MFCC, GFCC, power spectral density, and Hjorth parameters. A transformer encoder with integrated spectral and temporal attention mechanisms captures intricate patterns and long-range dependencies within the EEG signals. In the CD-FTA phase, the model undergoes fine-tuning on the SEED-V dataset (20 participants, 62 channels) leading to an improved accuracy of 90%, with an average precision of 90.6%, recall of 90.6%, and F1-score of 90.6%. The model’s generalization is further validated on the MPED dataset (23 participants, 62 channels, seven emotion classes), achieving 79%, with an average precision of 79.3%, recall of 79.3% and F1-score of 79.1% across diverse emotional states. This cross-dataset adaptation leverages transfer learning to enhance the model’s generalization across different emotional states and EEG datasets. Experimental results show that the proposed approach outperforms traditional methods, achieving superior accuracy and robustness in emotion recognition tasks. This work advances emotion recognition systems by addressing challenges in EEG signal proc

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

Real-Time Drone System for Detecting, Tracking, and Following of a Mobile Robot in ROS/Gazebo

Real-Time Drone System for Detecting, Tracking, and Followin...

引用

2025 International Conference on Control, Automation, and Instrumentation, IC2AI 2025

作者： Salman, Ali Dika, Mohammad Diab, Hassan Owayjan, Michel Achkar, Roger American University of Science & Technology Faculty of Engineering and Computer Science Beriut Lebanon Nantes University Saint-Nazaire France

ISBN: (纸本)9798331529604

Surveillance drones equipped with video transmission capabilities play a crucial role in modern security systems, with the integration of OpenCV for object detection marking a significant advancement. This study evaluates the detecting accuracy by varying the distance between the drone and the target object, addressing gaps in current research through a detailed practical insight. This research explores key aspects of drone-based video surveillance, including object detecting, tracking and real-time following, aiming to enhance understanding and methodology in computer Vision (CV) applications with Unmanned Aerial Vehicles (UAVs). The Simulation of a surveillance drone system in Gazebo simulator within the ROS2 framework. The drone is programmed to dynamically follow and track a mobile robot, enabling systematic analysis of the object detection algorithm's performance. The findings contribute to advancing the reliability and effectiveness of drone-based video surveillance systems for future innovations in security and monitoring applications. © 2025 IEEE.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

Enhancing Machine Learning and Deep Learning Models For Depression Detection: A Focus on SMOTE, RoBERTa, and CNN-LSTM

Informatica (Slovenia)

引用

Informatica (Slovenia) 2025年第14期49卷 1-18页

作者： Taoussi, Chaimae Lyaqini, Soufiane Metrane, Abdelmoutalib Hafidi, Imad Laboratory of Process Engineering Computer Science and Mathematics University Sultan Moulay Slimane Beni Mellal Morocco Faculty of Sciences and Technology Cadi Ayyad University Marrakech Morocco

Depression is a major public health concern, affecting millions worldwide, and necessitates early, accurate detection for timely intervention. This study focuses on enhancing machine learning (ML) and deep learning (DL) models for improved accuracy in depression detection using the Counsel Chat Dataset. To address the challenges of class imbalance, we employed advanced preprocessing techniques, including the Synthetic Minority Oversampling Technique (SMOTE), alongside model fine-tuning and architectural optimizations. Our results demonstrated significant performance improvements, particularly with transformerbased models and hybrid architectures. RoBERTa, a transformer-based model, achieved an accuracy of 91.55%, an F1-score of 0.91, and a recall of 92.10%, outperforming state-of-the-art approaches. Similarly, CNN-LSTM attained an accuracy of 91.67% with a 95% CI of (0.8987, 0.9312), while XGBoost achieved the highest accuracy among ML models at 93.06%, with a 95% CI of (0.921, 0.941). Statistical tests validated the superiority of these models, with p-values of 5.48e-13 for RoBERTa and 3.41e-16 for XGBoost. These findings underscore the pivotal role of data augmentation and preprocessing in creating balanced datasets and enhancing the predictive capabilities of AI models for depression detection. © 2025 Slovene Society Informatika. All rights reserved.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

CAMeL: Cross-Modality Adaptive Meta-Learning for Text-Based Person Retrieval

引用

IEEE Transactions on Information Forensics and Security 2025年 20卷 4651-4663页

作者： Yu, Hang Wen, Jiahao Zheng, Zhedong Shanghai University School of Computer Engineering and Science Shanghai200444 China University of Macau Faculty of Science and Technology Institute of Collaborative Innovation China

Text-based person retrieval aims to identify specific individuals within an image database using textual descriptions. Due to the high cost of annotation and privacy protection, researchers resort to synthesized data for the paradigm of pretraining and fine-tuning. However, these generated data often exhibit domain biases in both images and textual annotations, which largely compromise the scalability of the pre-trained model. Therefore, we introduce a domain-agnostic pretraining framework based on Cross-modality Adaptive Meta-Learning (CAMeL) to enhance the model generalization capability during pretraining to facilitate the subsequent downstream tasks. In particular, we develop a series of tasks that reflect the diversity and complexity of real-world scenarios, and introduce a dynamic error sample memory unit to memorize the history for errors encountered within multiple tasks. To further ensure multitask adaptation, we also adopt an adaptive dual-speed update strategy, balancing fast adaptation to new tasks and slow weight updates for historical tasks. Albeit simple, our proposed model not only surpasses existing state-of-the-art methods on realworld benchmarks, including CUHK-PEDES, ICFG-PEDES, and RSTPReid, but also showcases robustness and scalability in handling biased synthetic images and noisy text annotations. © 2005-2012 IEEE.

关键词： Adaptation models Metalearning Training Data models Lighting Image color analysis Visualization Annotations Synthetic data Optimization

来源：评论

学校读者我要写书评

暂无评论

Resource Scheduling and Delay Optimization of IoT Devices in UAV-Assisted Multi-Access Edge Computing

引用

IEEE Internet of Things Journal 2025年第11期12卷 16998-17011页

作者： Qu, Long Wang, Jiming Assi, Chadi Faculty of Electrical Engineering and Computer Science of Ningbo University Ningbo315211 China MontrealQCH3G 2W1 Canada

Multi-access edge computing (MEC) plays a crucial role in providing low-latency and high-data transmission services to Internet of Things (IoT) devices. However, in remote areas where deploying edge devices is challenging, optimizing delay remains a significant research focus. To address this issue, our research investigates a multi-unmanned aerial vehicle (UAV)-assisted IoT task offloading model. In this model, tasks generated by IoT devices equipped with energy harvesting (EH) capabilities are offloaded to MEC servers with the assistance of multiple UAVs. In order to monitor and manage the energy consumption of IoT devices and the task backlog of edge servers, the energy consumption and task update queues are established. We formulate a mixed integer non-linear programming (MINLP) problem, which aims to optimize the allocation of communication and computation resources to minimize the execution latency of IoT devices. To ensure the stability of each queue, we employ the weighted perturbation method within the Lyapunov optimization framework to decompose the original problem. And a low complexity Multi-UAV Assisted Offloading (MUAO) algorithm is designed. Simulation results show that MUAO consistently exhibits lower latency and energy consumption compared to the baseline scheme and other existing algorithms, while maintaining a low packet loss rate of only 5%. © 2025 IEEE.

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

Unveiling the Superiority of Unsupervised Learning on GPU Cryptojacking Detection: Practice on Magnetic Side Channel-Based Mechanism

引用

IEEE Transactions on Information Forensics and Security 2025年 20卷 4874-4889页

作者： Cao, Hangcheng Xu, Guowen He, Ziyang Shi, Shaoqing Xu, Shengmin Wu, Cong Ning, Jianting Hunan University College of Computer Science and Electronic Engineering Changsha410012 China City University of Hong Kong Department of Computer Science Hong Kong Hong Kong University of Electronic Science and Technology of China School of Computer Science and Engineering Chengdu610054 China Zhengzhou University School of Cyber Science and Engineering Zhengzhou450001 China Zhengzhou University International College Zhengzhou450001 China Fujian Provincial Key Laboratory of Network Security and Cryptology College of Computer and Cyber Security Fuzhou350007 China Nanyang Technological University College of Computing and Data Science Jurong West 639798 Singapore Wuhan University School of Cyber Science and Engineering Wuhan430072 China City University of Macau Faculty of Data Science Macau China

Ample profits of GPU cryptojacking attract hackers to recklessly invade victims’ devices, for completing specific cryptocurrency mining tasks. Such malicious invasion undoubtedly obstructs normal device usage and wastes computation resources. To resist the threat of GPU cryptojacking, existing works aim to timely detect and clear away it, by distinguishing the dissimilitude between it and legitimate applications. However, these detection mechanisms inappropriately rely on two conflict cornerstones, manifested in leveraging mutable samples of illegitimate cryptojacking to design supervision-based detection models requiring samples with stable patterns. This limitation compromises the practicability of existing detection mechanisms in the face of mutable cryptojacking samples. To fill the gap, we explore the superiority of unsupervised learning in handling this issue and further propose an unsupervised manner-enabled detection mechanism named MagInspector, only using legitimate applications’ magnetic signatures from GPU side channels for model construction. MagInspector innovates in training an unsupervised autoencoder network by an adversarial mode that well learns the stable signature patterns of legitimate applications, while incompatible with mutable cryptojacking ones. In the process of model training, we elaborately extract mutual energy cumulation distribution features to represent legitimate applications to overcome the impact of their inter-type differences. Meanwhile, a locality sensitive hashing-driven outlier removal algorithm is designed to enhance MagInspector’s robustness to the noise samples. Finally, extensive experiments are conducted on GPUs covering four generations of common NVIDIA architectures and two generations of AMD architectures;the results show that applying MagInspector to mutable cryptojacking signature detection achieves a significant average accuracy improvement of 25.5% and 17.8%, respectively. © 2005-2012 IEEE.

关键词： Unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

From 5G to 6G: The Role of AI, Machine Learning, and Deep Learning in Wireless Systems 4

From 5G to 6G: The Role of AI, Machine Learning, and Deep Le...

引用

4th International Conference on Sentiment Analysis and Deep Learning, ICSADL 2025

作者： Gote, Pradnyawant M. Kumar, Praveen Verma, Prateek Yesankar, Prajyot Pawar, Adesh Saratkar, Saniya Faculty of Engineering and Technology Department of Computer Science & Design Maharashtra Wardha442001 India Faculty of Engineering and Technology Department of Computer Science & Medical Engineering Maharashtra Wardha442001 India Faculty of Engineering and Technology Department of Artificial Intelligence & Machine Learning Maharashtra Wardha442001 India Faculty of Engineering and Technology Department of Artificial Intelligence & Data Science Maharashtra Wardha442001 India

ISBN: (纸本)9798331523923

The swift progression of wireless communication technologies-specifically from 5G to 6G is an approach that could be the most significant revolutionary leap towards changing connectivity and data transmission forever. Additionally, algorithms of machine learning (ML) and deep learning (DL) in 5G and 6G can offer promising advancements toward predictive maintenance, dynamic spectrum management, and enhanced security. The review of the potentially transformative nature of these technologies underlines the need for future directions of research efforts called for in circumventing challenges towards making full capability exploitation of 6G possible. © 2025 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

A Review on the Applications of Machine Learning and Deep Learning Techniques for Skin Cancer Detection 4

A Review on the Applications of Machine Learning and Deep Le...

引用

4th International Conference on Sentiment Analysis and Deep Learning, ICSADL 2025

作者： Wanjari, Ketan Verma, Prateek Faculty of Engineering and Technology Department of Computer Science and Engineering Maharashtra Wardha442001 India Faculty of Engineering and Technology Department of Artificial Intelligence and Data Science Maharashtra Wardha442001 India

ISBN: (纸本)9798331523923

Skin cancer is the most commonly reported type of cancer globally and one of the few cancers that can be effectively treated if detected in its early stages. Recent advancements in artificial intelligence (AI) have significantly improved skin cancer diagnosis through Machine Learning (ML) and Deep Learning (DL) models. This review explores ML and DL methodologies, including Random Forest, Convolutional Neural Networks (CNNs), Inception Networks, ResNets, and Support Vector Machines (SVMs), which have been widely used for automated diagnosis. However, existing AI-based systems face several challenges, including the need for balanced datasets, model interpretability issues, computational complexity, and difficulties in generalizing across diverse populations. Moreover, multimodal detection systems and Explainable AI (XAI) present additional challenges in making AI-driven diagnoses more reliable and transparent. This article discusses these challenges while also highlighting emerging techniques that integrate multiple data sources and individualized diagnostic tools to enhance precision. By addressing these challenges, this review aims to provide insights into the practical advancements and future directions of AI in skin cancer detection and treatment. © 2025 IEEE.

关键词： Adversarial machine learning

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：