检索结果-内蒙古大学图书馆

15th International Conference on Computing Communication and Networking Technologies, ICCCNT 2024

作者： Saxena, Aradhana Santhanavijayan, A. Department of Computer Science and Engineering National Institute of Technology Tamil Nadu Tiruchirappalli India

ISBN: (纸本)9798350370249

This study addresses the critical need for automated question difficulty classification in educational settings. By leveraging the DistilBERT model and Explainable AI (XAI) techniques, we aim to enhance the accuracy and interpretability of classifying questions into three difficulty levels: simple, average, and tough. We trained and evaluated the model using a dataset of 9041 questions, measuring performance with metrics including accuracy, precision, recall, F1 score, AUC, and cross-entropy loss. Our findings demonstrate significant improvements in classification accuracy with increasing data sizes, reaching an accuracy of 0.97 with 5000 samples. Additionally, we employed LIME to provide insights into the model’s decision-making process, enhancing the transparency and trustworthiness of the predictions. This research offers a novel approach to question difficulty classification, with potential applications in adaptive learning systems and automated assessment tools. ©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

Onboard Class Incremental Learning for Resource-Constrained scenarios using Genetic Algorithm and TinyML

Onboard Class Incremental Learning for Resource-Constrained ...

引用

2024 Genetic and Evolutionary Computation Conference Companion, GECCO 2024 Companion

作者： Pandey, Suraj Kumar Nair, Shivashankar B Department of Computer Science and Engineering Indian Institute of Technology Guwahati Assam Guwahati India

ISBN: (纸本)9798400704956

Deploying Machine Learning (ML) models in real-world settings over resource-constrained edge devices has always been a challenging task. While TinyML tackles this issue to an extent, by mostly using pre-trained Deep Learning (DL) models, the static nature of such models renders them ineffective for non-stationary data. A model having a low memory footprint that could allow onboard Class Incremental Learning (CIL) so as to accommodate data from new classes and also avoid catastrophic forgetting, is thus, the need of the day. The work described in this paper endeavours to meet this need by providing a method that utilises TinyML to accommodate a DL model onboard a resource-constrained device. To enable onboard CIL over the DL model, the method leverages Latent Replays (as exemplars) and a Genetic Algorithm (GA) to create a multi-fitness landscape that facilitates the inclusion of new class data, suppresses catastrophic forgetting and keeps a check on the quality of exemplars. Experiments and comparisons conducted for a gesture recognition task using time-series data with the proposed method deployed on a microcontroller, show the effectiveness of augmenting TinyML DL models with the GA for onboard CIL. © 2024 Copyright held by the owner/author(s).

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

A Deep Assessmen of Smart Logistics Kind of Parks with the Fog Based and Enabled Sensor Cloud Type Platforms 1

A Deep Assessmen of Smart Logistics Kind of Parks with the F...

引用

1st International Conference on Innovative Sustainable Technologies for Energy, Mechatronics and Smart Systems, ISTEMS 2024

作者： Yadav, Savita Gupta, Ashish Preethi, D. Noida Institute of Engineering and Technology Department of Internet of Things Uttar Pradesh Greater Noida India Tula's Institute Department of Computer Science and Engineering Dehradun India Department of Computer Science and Information Technology Bangalore India

ISBN: (纸本)9798350384246

The purpose of this research is to compare, in the environment of a clever logistics park, the effectiveness of fog-enabled sensor clouds (FSC) with traditional cloud computing with fog computing models. Expanding on our earlier work, we describe the physical instrument virtualization framework and scheme that are essential to the envisioned FSC. There is a careful construction of the network model and a mathematical explanation of the FSC parameters. We use a well-known logistics company in China is a case study to assess the efficacy of the suggested platform and clarify the network architecture in a real-world logistics situation. The experiment's results confirm that, with regard to bandwidth usage and service delay, the FSC for logistics parks is demonstrably superior to standard internet-based and fog computing models. © 2024 IEEE.

关键词： Fog

来源：评论

学校读者我要写书评

暂无评论

Beyond Information Exchange: An Approach to Deploy Network Properties for Information Diffusion 2nd

Beyond Information Exchange: An Approach to Deploy Network ...

引用

2nd International Conference on Big Data, Machine Learning, and Applications, BigDML 2021

作者： Das, Soumita Biswas, Anupam Kishore Devarapalli, Ravi Department of Computer Science and Engineering National Institute of Technology Assam Silchar788010 India

ISBN: (纸本)9789819934805

Information diffusion in Online Social Networks is a new and crucial problem in social network analysis field and requires significant research attention. Efficient diffusion of information is of critical importance in diverse situations such as pandemic prevention, advertising and marketing. Although several mathematical models have been developed till date, but previous works lacked systematic analysis and exploration of the influence of neighborhood for information diffusion. In this paper, we have proposed Common Neighborhood Strategy (CNS) algorithm for information diffusion that demonstrates the role of common neighborhood in information propagation throughout the network. The performance of CNS algorithm is evaluated on several real-world datasets in terms of diffusion speed and diffusion outspread and compared with several widely used information diffusion models. Empirical results show CNS enables better information diffusion both in terms of diffusion speed and diffusion outspread. © 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Marketing

来源：评论

学校读者我要写书评

暂无评论

A Key-Value pair Schema based Message Digest-5 Hash Algorithm for the Dynamic Data Masking 2

A Key-Value pair Schema based Message Digest-5 Hash Algorith...

引用

2nd IEEE International Conference on Data science and Network Security, ICDSNS 2024

作者： Gowda, Bharath Tumkur Shankare Raju, Channa Krishna Sri Siddhartha Institute of Technology Department of Computer Science and Engineering Maraluru Tumkur India

ISBN: (纸本)9798350373110

Dynamic data masking (DDM) is an efficient data security system for protecting personal, as well as other data from attackers or third parties. Reversible Data Hiding (RDH) is a scheme that hides secret data in a cover image through steganography images. However, when hiding a large amount of data in an embedding process, the data can be lost, leading to security problems. Hence, this research proposes the Key-Value pair schema-based Message Digest-5 (MD5) algorithm for dynamic data masking. The MD5 algorithm produces fixed size message digest of 128-bit and makes the process irreversible, thereby enhancing the security level over unauthorized users. Data masking is introduced on selected database areas to hide sensitive data in the group of query outcomes. The proposed Key-Value pair schema-based MD5 approach embeds one or multiple secret attributes about various hiding attributes in a similar relational database. The proposed method is validated using various performance metrics like Peak Signal-to-Noise Ratio (PSNR) of 8.383dB and Embedding Rate of 0.0065, as opposed to the existing methods, Quantum Particle Swarm Optimization (QPSO) and Multi-direction Pixel Swapping Approach (MPSA). © 2024 IEEE.

关键词： Particle swarm optimization (PSO)

来源：评论

学校读者我要写书评

暂无评论

Texture Recognition Using InceptionNeXt-based Texture Perception Network 15

Texture Recognition Using InceptionNeXt-based Texture Percep...

引用

15th International Conference on Computing Communication and Networking Technologies, ICCCNT 2024

作者： Sahoo, Tapan Kumar Mawari, Gaurav National Institute of Technology Warangal Department of Computer Science and Engineering Telangana Hanamkonda India

ISBN: (纸本)9798350370249

Recognizing textures remains challenging in images due to variations in fundamental elements (primates) and characteristics (attributes) caused by spatial context. Existing Convolutional Neural Network (CNN) approaches typically use detailed local descriptions with unordered aggregation for spatial layout inconsistency. However, these methods overlook the inherent structure relationships between primates and the semantic meanings conveyed by attributes, which are crucial for texture. This paper suggests a newfangled approach for texture cognition using Inception NeXt, a deep learning architecture known for its long-range modelling capabilities. Inception NeXt uses large-kernel convolutions to capture broader contextual information within textures. We address the computational efficiency boundaries associated with large kernels by incorporating an Inception-inspired decomposition strategy. This approach decomposes the large kernels into smaller, parallel branches, enhancing efficiency while maintaining performance. Our model extracts features by analyzing the relationships between these primates and attributes within the spatial context of the texture. The Inception NeXt architecture facilitates the capture of long-range dependencies between texture elements, leading to more robust feature representations. Experiments on two challenging texture databases demonstrate the effectiveness of our approach, achieving superior performance in terms of accuracy, robustness, and efficiency compared to traditional CNN-based methods. © 2024 IEEE.

关键词： Deep learning Large-kernel convolutions Robustness Spatial context Texture recognition

来源：评论

学校读者我要写书评

暂无评论

Automated Insect Detection and Classification Using Pelican Optimization Algorithm with Deep Learning on Internet of Enabled Agricultural Sector

引用

SN computer science 2024年第5期5卷 576页

作者： Assiri, Mohammed Elhameed, Elmouez Samir Abd Kumar, Arun Singla, Chinu Department of Computer Science College of Sciences and Humanities- Aflaj Prince Sattam bin Abdulaziz University Aflaj 16273 Saudi Arabia Department of Computer Science College of Post-Graduated Studies Sudan University of Science and Technology Khartoum Sudan Department of Electronics and Communication Engineering New Horizon College of engineering Bengaluru India Computer Engineering Thapar Institute of Engineering and Technology Punjab Patiala India

Recently, the combination of Deep Learning (DL) methods within the Internet of Things (IoTs) has developed in the agricultural field, especially in the domain of pest management. This study considers the implementation and development of an innovative method for Insect Detection and Classification using DL within the environment of the IoTs in agriculture. The developed system leverages advanced DL approaches for analyzing images captured by IoT-enabled devices, enabling real-time identification and categorization of insect pests. By continuously incorporating these technologies, this research goals to increase the efficiency and precision of pest monitoring, finally providing to sustainable agricultural technologies and increased crop yield. This study presents an Automated Insect Detection and Classification using Pelican Optimization Algorithm with Deep Learning (AIDC-POADL) technique on Internet of Enabled Agricultural Sector. The main aim of the AIDC-POADL system is to detect and categorize various kinds of insects exist in the agricultural field. In the primary stage, the AIDC-POADL technique involves DenseNet-121 model to learn complex features in the input images. Besides, the hyperparameter selection of the DenseNet-121 model developed by the POA. At last, multilayer perceptron (MLP) model can be applied to discriminate the insects into various classes. To validate the improved performance of the AIDC-POADL algorithm, a series of simulations were involved. The experimental outcomes stated that the AIDC-POADL technique offers enhanced recognition results over other approaches. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Agricultural region Deep learning Insect detection Internet of things Pelican optimization algorithm

来源：评论

学校读者我要写书评

暂无评论

Pothole Detection using Smartphone: A Driver Assistant

Pothole Detection using Smartphone: A Driver Assistant

引用

2023 IEEE International Conference on Modelling, Simulation and Intelligent Computing, MoSICom 2023

作者： Kumar, Deepak Kamalapuri, Amar Choudhury, Nehal Mukherjee, Shyamapada National Institute of Technology Silchar Department of Computer Science and Engineering Silchar India National Institute of Technology Rourkela Department of Computer Science and Engineering Rourkela India

ISBN: (纸本)9798350393415

India has an extensive road network that becomes degraded over time, causing potholes and other hazardous conditions for drivers. Potholes can cause physical injury and vehicle damage, so their detection is crucial. While some existing systems help drivers identify potholes, they lack precise localization and are location-dependent. In this paper, a cost-effective solution that utilizes mobile phone cameras and a deep neural network (Single Shot MultiBox Detector with MobileNet) for real-time pothole detection and localization is presented. The system alerts drivers about detected potholes instantly and achieved good mean Average Precision. Our solution aims to improve road safety and minimize vehicle damage. © 2023 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Video Frame Prediction Using Convolutional LSTM Encoder and Decoder 1

Video Frame Prediction Using Convolutional LSTM Encoder and ...

引用

1st International Conference on Data, Computation and Communication, ICDCC 2024

作者： Revathy, G. Jagadeesan, Pavithra Vajubunnisa Begum, R. Vijay Anand, P. Jasmin, H. Gayathri, S. Department of Computer Science and Engineering Chennai India Karpaga vinayaga college of engineering and technology Department of Computer Science and Engineering Chennai India Justice Basheer Ahmed Sayeed College for Women Affiliated by University of Madras Department of Electronics and Communication Science Teynampet Chennai India CMR Institute of Technology Department of Artificial Intelligence and Data Science Bengaluru India K. Ramakrishnan College of Technology Department of Computer Science and Engineering Tamilnadu Trichy India

ISBN: (纸本)9798331532956

With rapid growth of audiovisual content now spreading through social networking sites, chances of exposure towards violent and mature content by the youth are growing. This project addresses the risk of detection of such content using advanced video frame prediction techniques. We present here an encoder-decoder model of Convolutional LSTMs in the film for violent and obscene sequence detection. The spatiotemporal nature of video data encourages our algorithm to exploit accurate predictions of future frames from past content, which can be used in real time for dangerous material detection. The study relies on recent literature that highlights the psychological implications arising from the viewing of violent content by teenagers, thus making the need for efficient automated detection mechanisms ever more pertinent. Furthermore, we present a complete definition of 'violence' for the standardized detection mechanisms and to cross-compare various studies. Using Violent Scene Detection, our results indicate high improvements in the detection accuracy and therefore become effective in this context for parental control and content moderation. Such results help safeguard children in the increasingly digital environment: traditional as well as currently ever-increasing. © 2024 IEEE.

关键词： Video signal processing

来源：评论

学校读者我要写书评

暂无评论

Single Use Plastic Bottle Recognition and Classification Using Yolo V5 and V8 Architectures 4th

Single Use Plastic Bottle Recognition and Classification Usi...

引用

4th EAI International Conference on Cognitive Computing and Cyber Physical Systems, IC4S 2023

作者： Matta, Venkata Durgarao Mudunuri, K. A. Venkata Ramana Raju Sai Baba, B. Ch. S. N. L. S. Kiran, Kompella Bhargava Veenadhari, C. H. Lakshmi Prasanthi, B.V. Computer Science and Engineering Department Vishnu Institute of Technology Andhra Pradesh Bhimavaram India

ISBN: (纸本)9783031488900

Improper disposal of single use plastic bottles leads to many problems including danger to marine life and land pollution. Burning of plastic in turn releases dioxins and polychloride biphenyls. They are very harmful if inhaled and are threat to vegetation too. Manual sorting of plastic bottles and safe disposal is not an easy task. A lot of recycling initiatives use manual sorting for plastic recycling, which depends on plant staff visually identifying and selecting plastic bottles as they move along the conveyor belt. Automatic sorting of plastic bottles has advantage of non-intrusive sorting, speed, consistency, cost effectiveness in long run and even prevents health hazards to workers working in recycling environment. As a result, it is imperative to replace human sorting systems with intelligent automated systems. In this study, convolutional neural network architectures such as YOLOv5 and YOLOv8 were utilized to detect plastic bottles in images. Despite YOLOv8 having more parameters and requiring more computation time, it was found that YOLOv8 outperformed YOLOv5 in accurately identifying plastic bottles in the images. © 2024, ICST Institute for computer sciences, Social Informatics and Telecommunications engineering.

关键词： Cost effectiveness

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：