检索结果-内蒙古大学图书馆

2024 Asia Pacific Conference on Innovation in technology, APCIT 2024

作者： Suryavanshi, Ankita Mehta, Shiva Gopal, Lisa Choudhary, Ankur Hariharan, Shanmugasundaram Chitkara University Institute of Engineering and Technology Chitkara University Punjab India Graphic Era Hill University Computer Science and Engineering Uttarakhand Dehradun India Graphic Era Deemed to Be University Computer Science and Engineering Uttarakhand Dehradun India Vardhaman College of Engineering Department of Computer Science and Engineering Hyderabad India

ISBN: (纸本)9798350361537

This research intends to apply Federated Learning (FL) with Convolutional Neural Networks (CNNs) to realize this goal, which is the accurate recognition and categorization of English handwritten scripts and to make a step in the direction of Handwritten English Script Recognition Systems (HESRS). The clients do this by horizontally storing local data, and then they group the data into five categories that reflect five levels of Complexity in English scripts. The investigation of real-world customer data provides encouraging findings: Client ms-1 obtained an efficacy of 80.91% precision and 80.98% recall, which yielded a finality of 80.92% and an accuracy of 92%. By the end of the whole process, ms clients ms-2 to ms-5 reached the best level very quickly, with ms-5 having the highest accuracy of 89.9% along with precision (89.96%), recall (89.99%) and precision (89.97%). The averages of the data sets that were obtained through the aggregation via average, the overall averages ranged from 0.7956 for mp-2 to 0.8997 for mp-5. Hence, the young averages had almost the same trend, and the micro averages were slightly higher. Therefore, such a rating implies a consensus in the personal judgment and performance rating. The sub-least micro average marks its extreme estimated 89.98% at mp-5, meaning superpower performance modelling with all sets of information. The individual macro averages I have achieved are between 80.94% and 89.97%. The group weighted averages stay at 81.04% to 89.97%, whereas those for micro averages from customer mp-1 to mp-5 place them at slightly higher values of 81% to 89.98%, respectively. © 2024 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Masked Siamese Prompt Tuning for Few-Shot Natural Language Understanding

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第2期5卷 624-633页

作者： Ni, Shiwen Kao, Hung-Yu Chinese Academy of Sciences Shenzhen Institute of Advanced Technology Shenzhen518000 China National Cheng Kung University Department of Computer Science and Information Engineering Tainan701401 Taiwan

Recently, prompt-based learning has shown excellent performance on few-shot scenarios. Using frozen language models to tune trainable continuous prompt embeddings has become a popular and powerful methodology. For few-shot natural language understanding, even if we freeze the parameters of the pretrained language model, the learned pseudo-prompts might still be overfitted. In this article, we propose a novel masked Siamese prompt tuning (MSP-tuning) to improve few-shot natural language understanding. Concretely, MSP-tuning masks randomly out part of the prompt tokens to get a pair of masked Siamese prompts for each sample. Each training sample is then fed to the model twice with the masked Siamese prompts. Finally, MSP-tuning minimizes the Jensen-Shannon-divergence (JSD) between the two output probability distributions of the pretrained language model to regularize the model further. Experiment results on the few-shot GLUE benchmark and SuperGLUE benchmark show that MSP-tuning outperforms previous approaches. Numerically, our MSP-tuning achieves an average improvement of 1.79% (BERT-base) and 1.39% (BERT-large) of the GLUE benchmark and 1.90% (RoBERTa-base) and 1.71% (RoBERTa-large) of the SuperGLUE benchmark compared to state-of-the-art method P-tuning. Our method facilitates applying large pretrained language models in natural language understanding. © 2023 IEEE.

关键词： Bit error rate

来源：评论

学校读者我要写书评

暂无评论

Ensemble Learning based Categorical Boosting for Customer Lifetime Value Prediction in Businesses 1

Ensemble Learning based Categorical Boosting for Customer Li...

引用

1st International Conference on Intelligent Systems and Computational Networks, ICISCN 2025

作者： Monica, K.M. Habelalmateen, Mohammed I. Shashidhara, K.S. Buckshumiyan, A. Muthukumaran, K. Vellore Institute of Technology School of Computer Science and Engineering Chennai India The Islamic University College of Technical Engineering Department of Computers Techniques Engineering Najaf Iraq Nitte Meenakshi Institute of Technology Department of Electronics and Communication Engineering Bengaluru India New Prince Shri Bhavani College of Engineering and Technology Department of Mechanical Engineering Chennai India Dhanalakshmi Srinivasan College of Engineering and Technology Department of Bio Medical Engineering Mamallapuram India

ISBN: (纸本)9798331529246

Nowadays, Customer Lifetime Value (CLV) prediction is becoming critical task for businesses to identify high-value customers and optimize the marketing strategies. However, the existing Deep Neural Networks (DNN) failed to predict CLV in business due to the dynamic and heterogeneous nature of customer behaviour. Hence, this research proposes an Ensemble Learning based Categorical Boosting (EL-CatBoost) to provide decision-making and strategic planning for businesses. Initially, the data is collected from online retail customer churn dataset and fed into preprocessing with K-Nearest Neighbor (KNN) imputation. The preprocessing technique effectively handles the missing values in the data by replacing the missing value with nearest similar one. Then, the Latent Semantic Analysis (LSA) is employed to extract semantic, sentiment, and contextual features from the preprocessed data. Finally, the proposed EL-CatBoost is employed to predict the CLV by capturing the complex relationships within the features. From the results, the proposed EL-CatBoost achieved better results in terms of terms of Rate of Investment (ROI) with 10-day prediction (ROI-10) of 0.1803 and ROI with 20-day prediction (ROI-20) of 0.3795 when compared with the existing Temporal Structural User Representation (TSUR) network. © 2025 IEEE.

关键词： Sales

来源：评论

学校读者我要写书评

暂无评论

The integration of IoT in learning environments: Assessing the impact of educational empowerment and addressing its challenges 4

The integration of IoT in learning environments: Assessing t...

引用

4th International Conference on Innovative Practices in technology and Management, ICIPTM 2024

作者： Rajora, Ricky Rajora, Aarju Sharma, Bhanu Aggarwal, Priyanshi Thapliyal, Siddhant Chitkara University Institute of Engineering and Technology Chitkara University Punjab India Chandigarh Engineering College Department of Computer Science & Engineering Mohali India Computer Science & Engineering Graphic Era Hill University Uttarakhand Dehradun India Computer Science & Engineering Graphic Era Deemed to Be University Uttarakhand Dehradun India

ISBN: (纸本)9798350307757

This article explores the diverse impact of the Internet of Things (IoT) on education. This research investigation examines the possible uses of the IoT in several educational domains, with a particular focus on the significant impact of IoT on student engagement, learning methodologies, and administrative efficacy. While explaining the considerable advantages, this discourse also acknowledges the significant obstacles pertaining to integration, security, and privacy. The study finishes by presenting a forward-looking model for smart campuses, providing a holistic viewpoint on the IoT impact on transforming education. The article examines the impact of the IoT on educational practices and emphasizes the significance of institutional adaptation to effectively incorporate this technological paradigm shift. The article presents an elaborate framework for the prospective development of intelligent campuses and higher education institutions. This article offers a comprehensive guide for effectively using the whole capabilities of the IoT within the realm of education. © 2024 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

PointBi-FPN: An Extention to Pointpillars for LiDAR 3D Object Detection in Autonomous Vehicles Using Bi-Directional Feature Pyramid Network

PointBi-FPN: An Extention to Pointpillars for LiDAR 3D Objec...

引用

2024 IEEE Region 10 Conference, TENCON 2024

作者： Soumya, A. Krishna Mohan, C. Cenkeramaddi, Linga Reddy Babu, Sobhan Indian Institute of Technology Department of Computer Science and Engineering Hyderabad India University of Agder Department of Information and Communication Technology Grimstad Norway

ISBN: (纸本)9798350350821

Autonomous vehicles increasingly rely on accurate three-dimensional (3D) object detection for safe navigation. While two-dimensional (2D) methods offer computational efficiency, the shift to 3D detection enhances precision in understanding environments. Point-based and voxel-based approaches are accurate but computationally intensive for onboard deployment. Pillar-based methods like PointPillars provide efficiency but may lack detection accuracy compared to voxel-based approaches. In this paper, we focus on enhancing the performance of PointPillars, a popular pillar-based detector, using LiDAR data. The proposed approach (PointBi-FPN) employs a bi-directional feature pyramid network (Bi-FPN) as a backbone, that aggregates multiscale features in the input data, providing a holistic view of the environment, which is crucial for detecting objects of varying sizes and distances accurately. Bi-FPN improves the model's understanding of complex scenes, making it more robust to occlusions. Through extensive experimentation on the KITTI dataset, the PointBi-FPN approach demonstrates better performance across all three detection benchmarks: BEV (Bird's Eye View), 3D (Three-Dimensional), and AOS (Average Orientation Similarity). Notably, the proposed approach exhibits significant improvements, particularly in accurately detecting objects labeled as 'hard' difficulty. © 2024 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

A KNN Algorithm Based Predictive Model for Heart Disease Progression

引用

Recent Advances in computer science and Communications 2023年第5期16卷 76-81页

作者： Jaffrin, Lijetha C. Srikanth, G. Umarani Department of Information Technology Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Chennai India Department of Computer Science and Engineering Bharath Institute of Higher Education and Research Chennai India

Background: Disease diagnosis is a useful phenomenon in healthcare. Machine learning classification methods would considerably improve the healthcare industry by providing a quick diagnosis of the disease. Thus, time could be saved for doctors. Nearly 17.9 million people expire due to heart disease every year. Objectives: World Health Organization (WHO) predicted that rate of death might increase by 24.5 million in 2030. Since heart illness was the major cause of death in comparison with other diseases today, it was the most challenging disease to diagnose. Methods: One of the reasons for death due to heart disease was due to the fact that risks were not identified in the earlier stage. Earlier diagnosis of disease was very much important. Machine Learning algorithms were used for predicting the prognosis of disease. Results: Here K-NN algorithm was used to predict the presence of heart disease in an individual. Thus, patients were classified as either positive or negative for heart disease and this model enhanced medical care and reduced the cost. This gave us significant knowledge that helps us to predict the patients with heart disease. Conclusion: The Python sci-kit library was used to implement this in Anaconda Navigator's Spyder Integrated Development Environment. Experiments revealed that technique worked well and was more accurate than before. © 2023 Bentham science Publishers.

关键词： Diseases

来源：评论

学校读者我要写书评

暂无评论

Challenges in Cyber Physical Social Systems and Internet of Things 5

Challenges in Cyber Physical Social Systems and Internet of ...

引用

5th International Conference on Smart Electronics and Communication, ICOSEC 2024

作者： Revathi, S. Raja, J. Mohanraj, M. Malathi, K. Mallala, Balasubbareddy Vidhya, R.G. B.S. Abdur Rahman Crescent Institute of Science and Technology Department of Computer Science Engineering Chennai India Vel Tech Rangarajan Dr. Sagunthala R&d Institute of Science and Technology School of Computing Department of Computer Science and Engineering Chennai India Sri Eshwar college of engineering Department of Computer Science and Business Systems Coimbatore India Saveetha Engineering College Department of Artificial Intelligence and Machine Learning Chennai India Chaitanya Bharathi Institute of Technology Department of Electrical and Electronics Engineering Telangana Hyderabad India Hkbk College of Engineering Department of Ece Bangalore India

ISBN: (纸本)9798331504403

This research analyses the complex dynamics of Cyber-Physical-Social Systems (CPSS), encompassing cyber-physical systems, cybersecurity, the Internet of Things (IoT), and social media. By exploring the interactions among these elements, this study analyses the vulnerabilities, risks, and potential benefits within the CPSS framework. This research study analyses the evolving landscape of cybersecurity by considering various aspects such as data integrity, privacy in the IoT era, and the influence of social media on user behaviour and influence dynamics. The study also introduces innovative strategies to enhance CPSS resilience, including methods to mitigate cyber threats and enhance overall system integrity. Additionally, we investigate how social media shapes CPSS behaviours, offering valuable insights into influence dynamics. This research advances the comprehensive understanding of CPSS to support future studies and advancements in the protection and improvement of interconnected systems. © 2024 IEEE.

关键词： Cyber-Physical Systems Cyber-Physical-Social Systems (CPSS) Cybersecurity Internet of Things (IoT) Social Media

来源：评论

学校读者我要写书评

暂无评论

Detection and Classification of Blood Cancer using an Auto Encoder and a Regularised Extreme Learning Machine 14

Detection and Classification of Blood Cancer using an Auto E...

引用

14th International Conference on Advances in Computing, Control, and Telecommunication Technologies, ACT 2023

作者： Sharath, M.N. Rajesh, T.M. Surabhi, K.R. Sanjay, M. Nikshepa, T. Sowmya, S.R. Department of Computer Science and Engineering Rajeev Institute of Technology Hassan India Department of Computer Science and Engineering Dayananda Sagar University Bengaluru India Department of Information Science and Engineering Dayananda Sagar Academy of Technology and Management Bengaluru India

Leukemia is by far the most common type of blood cancer in humans. The length of time it takes to recover from an illness is determines by how quickly it is recognized and treated. Leukemia develops when the bone marrow generates an abnormally large amount of white blood cells. Because images of human blood samples are studied under the microscope, hematologists use techniques such as microscopic color imaging, image segmentation, grouping, and classification to diagnose patients as quickly and precisely as possible. Microscopy can identify blood cancer in visible and immature white blood cells in a variety of methods. Early and correct diagnosis of leukemia is critical for clinicians to properly treat patients. Following the submission of the input image, Image processing in image. © Grenze Scientific Society, 2023.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Exploring The Synergy of Convolutional Neural Network For Enhancement Image Classification 2

Exploring The Synergy of Convolutional Neural Network For En...

引用

2nd IEEE International Conference on Trends in Quantum Computing and Emerging Business Technologies, TQCEBT 2024

作者： Suguna, R. Valantina, G. Mary Kanna, Rajesh Lakshmi, D. Muneeshwari, P. Vel Tech Rangarajan Dr. Sagunthala R&d Institute of Science and Technology Department of Computer Science and Engineering AvadiTamil Nadu Chennai India Saveetha University Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Computer Science and Engineering Tamil Nadu Chennai India Amet Deemed to Be University Department of Petroleum Engineering Chennai India Panimalar Engineering College Department of Computer Science and Engineering Tamil Nadu Chennai India Rajalakshmi Engineering College Department of Computer Science and Engineering Tamil Nadu Chennai India

ISBN: (纸本)9798350384277

This study explores the synergy of Convolutional Neural Networks (CNNs) for enhancing image classification through a comprehensive series of experiments. Four main aspects of our investigation include model architectures, the impact of loss functions, hyperparameter tuning, and data augmentation strategies. We conducted these experiments on a diverse and representative image dataset. In the first set of experiments, we evaluated various CNN architectures. The baseline CNN (CNN-1) achieved a training accuracy of 95% and validation accuracy of 91%, with a test accuracy of 90%. Introducing attention mechanisms (CNN-2) and residual blocks (CNN-3) yielded improved results, with test accuracies of 91% and 92%, respectively. The EfficientNet-B0 architecture outperformed others, achieving a test accuracy of 93%. In the second set of experiments, we examined the impact of different loss functions. While the standard Cross-Entropy loss achieved a test accuracy of 88%, the Focal Loss and Adversarial Loss improved accuracy to 90% and 89%, respectively. Our customized loss function further enhanced performance, achieving a test accuracy of 91%. In the third set of experiments, we conducted hyperparameter tuning. We tested various combinations, and the best-performing hyperparameter set included a learning rate of 0.0003, a batch size of 128, and a dropout rate of 0.25, resulting in a test accuracy of 93%. Finally, in the fourth set of experiments, we explored different data augmentation strategies. Rotation & Scaling and Translation & Cropping yielded the highest test accuracies of 92% and 93%, respectively, while Color Augmentation achieved 91%. The absence of data augmentation resulted in a test accuracy of 88%. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Framework for Logistics During Hajj Using Autonomous Mobile Robots 1

Framework for Logistics During Hajj Using Autonomous Mobile ...

引用

1st International Conference on Logistics, ICL 2024

作者： Nasti, Shoaib Mohd Chishti, Mohammad Ahsan Central University of Kashmir Department of Information Technology Ganderbal India National Institute of Technology Department of Computer Science and Engineering Srinagar India

ISBN: (纸本)9798331518844

Every year, Muslims worldwide travel to Saudi Arabia to Perform Hajj. Both government and private organizations provide various services and facilities to support these pilgrims and resolve any difficulties they may face. There is a scope for improvement in high-quality and smart logistical services for pilgrims. Smart logistic services have emerged as a practical solution to the growing complexity and volume of logistical operations. Artificial Intelligence, Internet of Things, and Robotics are all helping logistical operations to function more smoothly and efficiently. This paper briefly describes the background of the holy pilgrimage. Then, it contributes to the literature by introducing logistics services using autonomous mobile robots. Finally, the theoretical framework for the smart logistic system using an autonomous mobile robot is illustrated, which summarizes our survey on using the autonomous mobile robot for logistic services during the Hajj season. © 2024 IEEE.

关键词： Mobile robots

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：