检索结果-内蒙古大学图书馆

International Conference on Electrical, computer and Communication engineering (ECCE)

作者： Md. Zobayer Ibna Kabir Department of Computer Science and Engineering Ahsanullah University of Science & Technology Dhaka Bangladesh

ISBN: (数字)9798350357509

ISBN: (纸本)9798350357516

Prevention of any disease in the early stage is a blessing for any person. If it is related to cancer, then getting rid of it from the beginning itself is a blessing. Hence, early detection of this kind of disease is very much important for ensuring the effective treatment of breast cancer of the patient. The primary goal of this paper is to develop a novel deep learning approach to categorize breast cancer and make the decision-making process of the trained models more trustable and transparent. For this classification purpose, we collect a breast cancer related dataset. As the sourced dataset is unbalanced, we augment the data to improve the performance of the models. We perform the necessary preprocessing to refine the dataset to feed the model. For classification purposes, we use 4 pre-trained CNN models such as EfficientNetV2S, InceptionResNetV2, EfficientNetV2M, and XceptionNet. To enrich the interpretation of the trained models, we use explainable AI techniques including Faster ScoreCAM and LIME. Deploying AI-driven diagnostic technologies into clinical workflows requires greater transparency, and for this reason, we need to utilize Explainable AI to ensure the interpretability of trained models. Among all implemented models, EfficientNetV2S achieved the highest accuracy with 91.02%. However, the present results provide a beneficial observation to the performance and explainability of deep learning models in breast cancer classification. Source Code: https://***/ZobayerAkib/Breast-Cancer-Classification-ECCE2025-IEEE2025

关键词： Deep learning Training Accuracy Explainable AI Source coding Prevention and mitigation Decision making Breast cancer Data models Feeds

来源：评论

学校读者我要写书评

暂无评论

Gamma Sampling for Intrusion Detection with Imbalanced Data 10th

Gamma Sampling for Intrusion Detection with Imbalanced Dat...

引用

10th International Symposium on Security and Privacy in Social Networks and Big Data, SocialSec 2024

作者： Kamalov, Firuz Mitra, Rohan Sulieman, Hana Department of Electrical Engineering Canadian University Dubai Dubai United Arab Emirates Department of Computer Science and Engineering American University of Sharjah Sharjah United Arab Emirates Department of Mathematics and Statistics American University of Sharjah Sharjah United Arab Emirates

ISBN: (纸本)9789819637737

Recent advances in artificial intelligence have prompted the use of machine learning methods in network security. In this paper, we address the issue of imbalanced data that is often present in network security datasets used in machine learning. We propose an oversampling method based on the Gamma distribution to balance the data prior to training. The results on several imbalanced datasets show the potential of the proposed method as a viable tool to build intrusion detection systems based on artificial intelligence. The accompanying code for the study is available on Github. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Network intrusion

来源：评论

学校读者我要写书评

暂无评论

Bringing Machine Learning Classifiers Into Critical Cyber-Physical Systems: a Matter of Design

引用

IEEE Access 2025年

作者： Sayin, Burcu Zoppi, Tommaso Marchini, Nicolo Khokhar, Fahad Ahmed Passerini, Andrea University of Trento Department of Information Engineering - Computer Science Via Sommarive 9 Povo Italy University of Florence Department of Mathematics and Computer Science Viale Morgagni 65 Florence Italy

Domain experts are striving to solve classification problems through the application of Machine Learning (ML) models, aiming for the highest accuracy. However, classifiers are inherently prone to misclassifications, especially when encountering unknown input data. This complicates their deployment in critical Cyber-Physical Systems (CPSs), where misclassifications can have detrimental effects on people, the environment, or infrastructure. This paper argues that ML classifiers should not be designed or evaluated in isolation but rather conceptualized as Critical System Classifiers (CSCs), which can reject uncertain predictions and trigger mitigation strategies in accordance to the encompassing CPS. We present an high-level architecture for CSCs that supports black-box integration, preprocessing for unknown detection, post-hoc calibration, and cost-sensitive thresholding, emphasizing the importance of evaluating classifiers using cost-aware metrics that account for rejections. We validate our proposal through experiments on tabular datasets related to failure prediction, intrusion and error detection, which are common applications of classifiers within CPSs. Key findings are: i) cost-sensitive evaluation promotes the usage of different classifiers compared to standard performance metrics, ii) tree-based models outperform statistical ones, classification-wise, iii) prediction calibration and rejection strategies provide a robust notion of confidence, and iv) combining multiple uncertainty-based rejectors achieves a favorable balance between high accuracy, low rejection rate, and cost. The analyses above were conducted using our publicly available CINNABAR GitHub repository for experimentations with CSCs. Overall, our study offers a system-level perspective, software architecture and implementation on classifier deployment in critical CPSs domains, paving the way for a trustworthy ML integration into real-world infrastructures. © 2013 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Developing a Transformer-based Autoencoder Model for Sentence Embedding

Developing a Transformer-based Autoencoder Model for Sentenc...

引用

International Conference on Electrical, computer and Communication engineering (ECCE)

作者： Sanjida Nuri Pearl Mohammad Shamsul Arefin Department of Computer Science & Engineering Chittagong University of Engineering & Technology Chattogram Bangladesh

ISBN: (数字)9798350357509

ISBN: (纸本)9798350357516

In natural language processing (NLP), sentence embedding plays a key role in converting sentences into fixed-length vectors or numerical representations. These embeddings capture the relationships between words and the overall meaning of a sentence. It enhances machine understanding of language. Traditional sentence embedding methods often rely on feature-based techniques and demand substantial amounts of labeled data. These huge amount of data are expensive and challenging to obtain. To overcome this issue, this paper introduces a method that utilizes unsupervised techniques to generate sentence embeddings from unlabeled data. The proposed approach combines transformers and autoencoder to create more meaningful sentence representations. Autoencoders are used to compress data and map it into a latent space. On the other hand, transformers focus on capturing the contextual meaning of words within sentences. This hybrid model provides a more comprehensive way of representing sentence semantics.

关键词： Hands Computational modeling Autoencoders Semantics Transformers Information retrieval Feature extraction Natural language processing Data models Vectors

来源：评论

学校读者我要写书评

暂无评论

Cloud-Enabled Bio-Inspired Optimization: A Scalable Solution for Complex Real-World Problems

Cloud-Enabled Bio-Inspired Optimization: A Scalable Solution...

引用

Intelligent Data Communication Technologies and Internet of Things (IDCIoT), International Conference on

作者： Shreya Sakthi U R Sruthi S Shruti S G N. Rajganesh Department of Computer Science and Engineering Sri Venkateswara College of Engineering Sriperumbudur India

ISBN: (数字)9798331527549

ISBN: (纸本)9798331527556

Bio-inspired optimization algorithms, encompassing swarm intelligence, neural networks, fuzzy logic, and evolutionary algorithms, are powerful tools for addressing complex real-world optimization problems. However, their effectiveness is often constrained by the computational demands of training and execution, which surpass the capabilities of traditional computing platforms-particularly when dealing with massive datasets or intricate optimization challenges. These limitations restrict the practical applicability of such algorithms. This work introduces a framework that leverages cloud computing as a scalable and efficient platform to overcome these computational barriers. By integrating bio-inspired optimization techniques with the elastic and high-performance resources of cloud environments, this research demonstrates significant improvements in both the scope and efficiency of these algorithms. Experimental results validate the proposed approach, showcasing enhanced performance, reduced execution time, and the ability to solve larger and more complex real-world problems, thereby unlocking the full potential of bio-inspired optimization in practical applications.

关键词： Training Fuzzy logic Cloud computing Artificial neural networks Computational efficiency Internet of Things Data communication Particle swarm optimization Optimization Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Ambisync: An Internet of Things Approach to Room Energy Automation and Optimization

Ambisync: An Internet of Things Approach to Room Energy Auto...

引用

Emerging Smart Computing and Informatics (ESCI), Conference on

作者： Yashas Garg Simarjit Kaur Khangura Vasudhara Mahajan Avantika Raina Naman Jain Sushma Jain Computer Science and Engineering Department Thapar Institute of Engineering and Technology Patiala Punjab

ISBN: (数字)9798331515683

ISBN: (纸本)9798331515690

The primary objective of this paper is to design and implement an intelligent room appliance management system that seamlessly integrates hardware and software components to optimize energy consumption and enhance user comfort. The system provides both manual and automated control over key room appliances, including air conditioners and lighting, through a mobile application and an automated hardware unit equipped with sensors and infrared (IR) controllers. The system's ability to adapt to varying conditions, such as changing temperatures and occupancy levels, has significantly enhanced energy efficiency. The novelty of this system lies in its integration of advanced real-time occupancy detection and environmental sensing, enabling appliances to respond dynamically to user presence and room conditions. By automatically deactivating appliances in unoccupied rooms, the system not only minimizes energy wastage but also contributes to sustainable energy management, offering a smarter and more efficient solution for modern living spaces.

关键词： Home appliances Energy consumption Automation Heuristic algorithms Lighting Hardware Energy efficiency Mobile applications Time factors Internet of Things

来源：评论

学校读者我要写书评

暂无评论

Mental Stress Prediction Using Machine Learning and Facial Emotion Recognition

Mental Stress Prediction Using Machine Learning and Facial E...

引用

International Conference on Communication Systems and Network Technologies (CSNT)

作者： C. Swedheetha E. Nihil Sankar C.B. Ramkishan Department of Computer Science and Engineering Velammal College of Engineering and Technology Madurai India

ISBN: (数字)9798331531935

ISBN: (纸本)9798331531942

Stress and mental well-being have emerged as frontline investigation issues in health research. Chronic stress has been proven to affect people’s physical and mental well-being. This paper proposes an integrated model of ML models and CNN to predict stress from the detection of facial emotion. It is designed based on the capabilities of Random Forest and adaptive boosting to evaluate the real-time emotional states of the students, thus keeping in line with the intention of early identification of mental health issues. In particular, the project targets university students for categorizing their academic stress levels either as low, medium, or high through their behavioral and emotional data. This is a CNN model that takes facial expressions and produces emotion categories like happiness, sadness, anger, and surprise, and provides real-time insight to the stress levels. The most important features regarding stress levels are discussed at length, like GPA, social support, sleep patterns, and digital behaviors. The results suggest how ML and CNN-based emotion detection can monitor and improve student mental health by managing their stress levels by introducing intervention at an early stage, supportive environments, and personalized coping strategies.

关键词： Emotion recognition Adaptation models Focusing Mental health Predictive models Feature extraction Real-time systems Convolutional neural networks Monitoring Random forests

来源：评论

学校读者我要写书评

暂无评论

Swarm-driven fine-tuned hybrid transfer learningarchitecture on breast ultrasound scans for advanced prognostic

引用

International Journal of Information Technology (Singapore) 2025年第2期17卷 1031-1038页

作者： Khan, Yusera Farooq Runani, Ankita Ahmed Mir, Bilal Department of Computer Science and Engineering Shri Mata Vaishno Devi University J&K Katra India Department of Electrical Engineering Baba Ghulam Shah Badshah University J&K Rajouri India Faculty of Engineering University of Toyama Toyama City Toyama City Toyama Japan

Breast Malignancy (Tumor) presents a serious health risk as it is the primary cause of cancer-related death for women globally. To enhance the outcomes for patients and ensure optimal treatment, early testing and diagnosing are important. In recent years, deep learning (DL) techniques have shown great promise in aiding the detection and diagnosis of breast maliganacy. In this study, the Breast Echography scans were taken from Breast Ultrasound Scans (BUS) dataset and were resized into 224 × 224, and finally, data pre-processing steps were used. Next, the collective and iterative nature of Particle swarm optimization (PSO), where the swarm of particles collaboratively refined the solution space, to enables the dynamic convergence towards an optimal solution. Further, five transfer learning models were implemented for a 3-class Breast Malignancy classification task ResNet50, EfficientNetB4, MobileNetV2, and InceptionV3. We propose a hybrid model that integrates the three top-performing Transfer Learning (TL) architectures: EfficientNetB4 + MobileNetV2 + InceptionV3. These models’ outputs are combined into a single representation by flattening and concatenating them. To maximize the weight learning of the integrated models, deep dense layers (DDL) are then applied to this combined output. The proposed hybrid approach effectively categorizes medical images related to Breast Malignancyas either malignant, benign, or pertaining to healthy individuals without prior knowledge of the presence of a cancerous lesion. From the experimental evaluation, it is shown that the proposed novel stacking model gives the highest Accuracy of 95.92%, Precision of 95%, Recall of 93%, and F1 Score of 98.12%. © Bharati Vidyapeeth's Institute of computer Applications and Management 2024.

关键词： Breast echography Breast malignancy Breast ultrasound scans (BUS) EfficientNetB4 Feature space InceptionV3 MobileNetV2 Swarm-driven optimization

来源：评论

学校读者我要写书评

暂无评论

Secure and Efficient Deduplication for Multimedia Sharing in Privacy-Sensitive Cloud Ecosystems

Secure and Efficient Deduplication for Multimedia Sharing in...

引用

Intelligent Data Communication Technologies and Internet of Things (IDCIoT), International Conference on

作者： A. Suresh Erasappa Murali V. Anjali Sri E Sumana Gopaluni Venkata Sai Sathwik P. Devendra Reddy Department of Computer Science and Engineering Siddharth Institute of Engineering & Technology Puttur India

ISBN: (数字)9798331527549

ISBN: (纸本)9798331527556

Current cloud-based systems face the challenge of managing storage and eliminating redundant data associated with the exponentially growing multimedia content, where numerous semantically similar but nonidentical files are included. In this paper we introduce a novel secure deduplication framework for encrypted cloud storage. This proposed system employs similarity-preserving hashing, fuzzy PoW protocol, and attribute-based encryption to perform efficient deduplication without compromising security and providing ownership verification. Security and deduplication rate scalability are improved over brute-force and ownership-faking attacks by dynamically varying thresholds based on file characteristics. Experimental evaluation shows that a 48 % average deduplication rate and resistance to common attack vectors make it suitable for real-world cloud storage scenarios. The framework fills critical gaps in existing approaches and provides an innovative solution for dealing with similarity-aware data deduplication in privacy-sensitive environments.

关键词： Resistance Cloud computing Data privacy Protocols Scalability Semantics Throughput Solids Vectors Encryption

来源：评论

学校读者我要写书评

暂无评论

A Prototype for Secure and Efficient Real Estate Transactions Using IPFS and Block Chain Technology 4th

A Prototype for Secure and Efficient Real Estate Transaction...

引用

4th International Research Conference on Machine Learning, Deep Learning, and Blockchain, IRCICD 2023

作者： Peter, Aaron Rajeev, Ashwin Chooralil, Varghese S. Kumar, Aromal Anil Baiju, Ben Department of Computer Science and Engineering Rajagiri School of Engineering and Technology Kochi India

ISBN: (纸本)9783031882364

The global real estate market is a significant asset class, which was valued at over $6.27 billion dollars in 2020. It is anticipated to grow at a compound annual growth rate (CAGR) of 6.4% between 2021 and 2028. Due to its large size and relatively stable nature, real estate has become quite expensive, mainly owing to the involvement of numerous intermediaries. In this context, the utilization of innovative technologies such as Blockchain and IPFS has emerged as a potential solution to address these issues. Therefore, the objective of this paper is to investigate the applicability of these technologies in the real estate market and assess their potential to mitigate the existing challenges. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： engineering education

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：