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： Kushwaha, Piyush Murmu, Mahendra Kumar National Institute of Technology Department of Computer Engineering Haryana Kurukshetra India

ISBN: (纸本)9798350370249

In today's digital age, effective surveillance and anomaly detection systems are crucial. YOLOv5, a cutting-edge real-time object detection model that stands out for its remarkable speed and accuracy, making it ideal for these applications. This paper outlines the essential libraries and dependencies for the project, highlighting the importance of setting up the correct environment. It details the process of importing datasets in the YOLO format and provides insights on effective dataset preparation and curation. The exploratory data analysis (EDA) section covers visualizing sample images with bounding boxes, analyzing class distribution, and assessing image size and resolution for YOLOv5 compatibility. The core discusses applying the YOLOv5 algorithm, including model loading, training, and fine-tuning, along with evaluating model performance. Finally, practical applications of the trained model, such as real-time and batch inference, are explored, demonstrating its utility in detecting suspicious activities for security and surveillance. This paper serves as a comprehensive resource for implementing YOLOv5, ensuring the successful deployment of a robust object detection system. © 2024 IEEE.

关键词： Anomaly Detection computer Vision Deep Learning Exploratory Data Analysis (EDA) Security Systems YOLOv5

来源：评论

学校读者我要写书评

暂无评论

Biomarkers to Predictive Modeling: Leveraging Feature Selection and Ensemble Learning for Ovarian Cancer Prognosis 2024

Biomarkers to Predictive Modeling: Leveraging Feature Select...

引用

16th International Conference on Contemporary Computing, IC3 2024

作者： Srivastava, Jaya Rathi, Megha Department of Computer Science & Engineering Jaypee Institute of Information Technology India

ISBN: (纸本)9798400709722

Treatment outcomes and patient survival rates are greatly improved by early identification of ovarian cancer. However, to increase diagnostic accuracy, effective predictive modeling is required due to the biomarkers' variability and complexity. This research proposes a complex methodology that combines feature selection and ensemble learning approaches to create a forecast model for the diagnosis of ovarian cancer. Recursive feature elimination (RFE), a thorough feature selection technique, was originally used to identify important biomarkers in order to lower dimensionality and improve model interpretability. We then used ensemble learning techniques, which fuse together several machine learning models, to boost the prognosis's stability and predictive performance. Using a dataset containing clinical and genetic data from ovarian cancer patients, ensemble techniques called Adaboost, Catboost, Random Forest, and Gradient boosting were put into practice and validated. Performance metrics like precision, accuracy, F1-score, recall and AUC-ROC curve were used to measure the techniques' efficacy. The findings show that our approach not only performs better than conventional single-predictor models, with results that are roughly 99% correct, but it also offers important insights into the biomarkers which are most symptomatic of the course of ovarian cancer. © 2024 Copyright held by the owner/author(s).

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Maximizing Profit of the Service Providers through Efficient Content Placement and Retrieval in an IoT-enabled Edge-Cloud Platform 18

Maximizing Profit of the Service Providers through Efficient...

引用

18th IEEE International Conference on Advanced Networks and Telecommunications Systems, ANTS 2024

作者： Gandhi, Preet Mahanta, Abhishriya Ghose, Manojit Sahu, Aryabartta Indian Institute of Technology Department of Computer Science and Engineering Guwahati India Indian Institute of Technology School of Computing and Electrical Engineering Mandi India Indian Institute of Information Technology Department of Computer Science and Engineering Guwahati India

ISBN: (纸本)9798350391725

In today's 5G era, the amount of data generated by the Internet of Things (IoT) devices is enormous. Data is processed and stored in the cloud under a traditional cloud computing architecture, and real-time processing and retrieval of these data is time-consuming. Therefore, an edge-cloud framework comes in handy where the content generated is placed on the edge nodes of the network, and it can reach the endusers in a time-bound manner. A majority of the previous research works on content placement under the edge-cloud architecture primarily aimed at improving the latency of the system. However, the approach proposed in this paper aims to maximize the profit of the service provider by effectively placing the content in a hierarchically organized edge-cloud network. We propose two strategies and compare them with random and greedy approaches for both synthetic and real data sets. The experimental results demonstrate the effectiveness of the proposed strategies, and the profit of the service providers increases by 38.81% (maximum) and 10.51% (on average) for the synthetic dataset and by 65.74% (maximum) and 22.83% (on average) for the real-world dataset when compared with the greedy approach. © 2024 IEEE.

关键词： Cloud platforms

来源：评论

学校读者我要写书评

暂无评论

Machine Learning Models for Heart Disease Prediction: Balancing Accuracy and Transparency 2

Machine Learning Models for Heart Disease Prediction: Balanc...

引用

2nd IEEE International Conference on IoT, Communication and Automation technology, ICICAT 2024

作者： Gupta, Sukriti Tripathi, A. Srivastava, Chaynika Buddha Institute of Technology Department of Computer Science & Engineering Gorakhpur India

ISBN: (纸本)9798350368109

Heart disease continues to be the world's leading cause of death, so precise and trustworthy predictive models are required to enable early diagnosis and individualized treatment. This paper addresses the escalating global burden of heart failure, marked by increasing morbidity, mortality, and strain on healthcare systems, stemming from left ventricular systolic and diastolic dysfunction. When compared to conventional methods, machine learning proves to be a more accurate tool for predicting heart diseases. Machine learning (ML) models have shown great promise in predicting heart disease, but their complexity often results in a trade-off between accuracy and interpretability. This study investigates various interpretable machine learning approaches, including decision trees, Mtree, Random Forest, K-means, to predict heart disease, while maintaining transparency. Utilizing a dataset comprising patient health records, including attributes like age, blood pressure, cholesterol levels, and electrocardiogram results, preprocess the data through normalization and feature selection to enhance model accuracy. Our study showcases the implementation of distinct algorithms, demonstrating their effectiveness in accurately detecting heart failure with an impressive accuracy range of approximately 89% to 92%. The study underscores the importance of balancing accuracy with transparency in developing ML models for healthcare applications, particularly in high-stakes scenarios like heart disease prediction. Keywords: Heart disease, Machine learning (ML), random forest, health records, decision trees, MTree. © 2024 IEEE.

关键词： Heart

来源：评论

学校读者我要写书评

暂无评论

Artificial Intelligence Applications in Accounting and Finance

Artificial Intelligence Applications in Accounting and Finan...

引用

2024 ASU International Conference in Emerging Technologies for Sustainability and Intelligent Systems, ICETSIS 2024

作者： Rahim, Rahila Chishti, Mohammad Ahsan National Institute of Technology Department of Computer Science Engineering Srinagar India

ISBN: (纸本)9798350372229

The world of Industry 4.0 embraces the transformation that calls for strategic investments, cooperative partnerships, and a proactive approach to overcome obstacles to successfully transition to smarter, more connected, and more efficient in- dustrial ecosystems. The accounting and finance industry has benefited greatly from process modernization, which has resulted in a significant decrease in the use of the traditional system through the integration of digital technologies. The use of various artificial intelligence technologies such as expert systems for audit and taxation, intelligent agents for customer service, Natural Language Processing (NLP), Robotic Process Automation (RPA), Blockchain, data analytics, and machine learning for decision- making has been proposed to automate, streamline, and improve the efficiency of various tasks and processes in accounting and finance operations. The study specifically investigates how the incorporation of AI technologies affects the reliability & timeliness of financial & accounting processes. The paper also presents some use cases and AI applications relevant to this field. © 2024 IEEE.

关键词： accounting AI finance ML NLP

来源：评论

学校读者我要写书评

暂无评论

Navigation Assistance for Visually Impaired People using Machine Learning 15

Navigation Assistance for Visually Impaired People using Mac...

引用

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

作者： Phatangare, Sheetal Rajeshirke, Siddhi Mule, Parth Pawar, Tanishka Morey, Sanskruti Vishwakarma Institute of Technology Department of Computer Engineering Pune411037 India

ISBN: (纸本)9798350370249

Navigational aids for visually impaired individuals have seen remarkable advancements owing to the integration of cutting-edge technologies. This paper introduces an innovative approach aimed at enhancing blind navigation systems by using computer vision, depth estimation, and auditory feedback. The proposed system leverages the robustness of the YOLOv3 object detection model for real-time localization and identification of surrounding objects. Concurrently, depth estimation is achieved through a U-Net architecture, providing essential spatial information vital for navigation assistance. YOLOv3 (You Only Look Once version 3) is a fast and accurate object detection algorithm that scans an image once to identify and locate objects by predicting bounding boxes and class probabilities for different regions. U-Net is a neural network for image segmentation, classifying each pixel in an image. Its U-shaped structure with encoder and decoder paths allows it to effectively capture and reconstruct image details, making it suitable for detailed analysis tasks. The integration of Raspberry Pi, Pi Cam, and audio output complements the system, delivering real-time auditory cues to users about detected objects and their estimated distances. The smooth integration of these technologies provides a thorough and pragmatic answer to help visually impaired individuals navigate their environment with increased safety and independence. © 2024 IEEE.

关键词： Depth perception

来源：评论

学校读者我要写书评

暂无评论

Person Re-Identification: A Retrospective Study, Recent Trends and Future Scope 1

Person Re-Identification: A Retrospective Study, Recent Tren...

引用

1st International Conference on Pioneering Developments in computer Science and Digital Technologies, IC2SDT 2024

作者： Soni, Shashank Kumar Kumar, Gautam National Institute of Technology Department of Computer Science and Engineering Delhi India

ISBN: (纸本)9798350365016

Person re-identification (Re-ID) searches several non-overlapping cameras for a person of interest. The development of deep neural networks combined with the increasing need for intelligent video surveillance has led to a huge surge in interest in the field of computer vision. This article highlights the state-of-the-art techniques used for person re-identification (reID), including the latest advancements and challenges. It outlines the different person re-identification datasets that are available in the literature, evaluates the effectiveness of the best techniques across a range of datasets, and clarifies the important restrictions on person re-identification, including the effects of changing environmental conditions and data limitations. Furthermore, the article points out the future research direction, focusing on creating more realistic datasets and designing robust algorithms that can effectively handle real-world conditions in order to drive the growth and development of re- ID technology. © 2024 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Audio Spoof Detection using Deep Residual Networks based Feature Extraction: Unveiling Synthetic, Replay and Mimicry Threats

Audio Spoof Detection using Deep Residual Networks based Fea...

引用

2024 International Conference on Recent Innovation in Smart and Sustainable technology, ICRISST 2024

作者： Chakravarty, Nidhi Dua, Mohit National Institute of Technology Kurukshetra Department of Computer Engineering Haryana India

ISBN: (纸本)9798350309140

With the increasing prevalence of artificial intelligence-driven audio systems, the vulnerability to adversarial attacks, specifically synthetic, replay and mimicry, has become a critical concern. This paper has proposed a deep Residual Network as a feature extractor to resolve this problem. This proposed approach is divided into two phases: Frontend and Backend. During the initial phase, the audio data underwent Mel spectrogram generation, and a modified ResNet50 architecture has been employed for feature extraction from the resulting Mel spectrograms. Subsequently, Linear Discriminant Analysis (LDA) has been applied as a dimensionality reduction technique to optimize the complexity of the extracted features. In the second phase, the features selected by LDA have been utilized to train machine learning (ML) algorithms, including Support Vector Machine (SVM), Random Forest (RF), K-Nearest Neighbour (KNN), and Naive Bayes (NB). The training process involved utilizing the ASVspoof 2019 Logical Access (LA) and Physical Access (PA) training partitions, while the proposed model's validation and evaluation have been conducted using the development and evaluation partitions of the same dataset. Also, the Voice Impersonation Corpus in Hindi Language (VIHL) dataset has been used to detect mimicry attacks. The proposed model has achieved an EER of 2.4%, 2.2%, and 4.1% for synthetic, replay and mimicry audio attack detection, respectively. © 2024 IEEE.

关键词： Dimensionality reduction

来源：评论

学校读者我要写书评

暂无评论

Integrity Evaluation Model for Object Oriented Software: A Developer Perspective 1

Integrity Evaluation Model for Object Oriented Software: A D...

引用

1st IEEE Karachi Section Humanitarian technology Conference, Khi-HTC 2024

作者： Gupta, Vivek Gupta, C.L.P. Bansal Institute of Engineering and Technology Department of Computer Science Lucknow India

ISBN: (纸本)9798350373042

Software integrity stands as a paramount element in software development. Evaluating integrity through design properties proves more fitting, and its validation underscores the genuine impact of structural and functional aspects in object-oriented design software. Integrity serves as a crucial quality indicator, and gauging it opens avenues for streamlining and enhancing the maintenance process. Software integrity offers valuable insights throughout the phases of design, coding, testing, and quality assurance. Both researchers and practitioners emphasize the desirability and significance of the integrity aspect in developing secure and robust software. Despite its vital role in the software development process, integrity is often inadequately managed. This paper underscores the necessity and significance of integrity during the design phase and endeavors to establish a substantial correlation between integrity and design properties. In light of this, a model is proposed for evaluating integrity in object-oriented design through the establishment of multiple linear regressions. Ultimately, the proposed model undergoes validation through experimental trials. © 2024 IEEE.

关键词： Multiple linear regression

来源：评论

学校读者我要写书评

暂无评论

An Optimal Weighted Ensemble of 3D CNNs for Early Diagnosis of Alzheimer’s Disease

引用

SN computer Science 2024年第2期5卷 252页

作者： Dharwada, Sriram Tembhurne, Jitendra Diwan, Tausif Department of Computer Science and Automation Indian Institute of Science Bangalore India Department of Computer Science and Engineering Indian Institute of Information Technology Maharashtra Nagpur 441108 India

Alzheimer disease (AD) is a chronic neurological disorder in which the loss of brain cells causes dementia. Early and accurate diagnosis of AD will lead to better treatment of the disease before irreversible brain damage has been occurred. This paper proposes the classification of Alzheimer's disease using 3D structural Magnetic Resonance Imaging (sMRI) images through 3D convolutional neural networks (CNNs). Most existing methods utilizing 3D subject-level CNNs for Alzheimer's disease classification design a single model which relies on a very large training dataset for improved generalization. Herein, we address this issue through 3D transfer learning which makes use of knowledge gained from a pre-trained task. We train 3D versions of five classical 2D image classification architectures—ResNet, ResNeXt, SE-ResNet, SE-ResNeXt, and SE-Net—by initializing each model with pre-trained weights from their 2D counterparts, and combine their predictions through a weighted average method. The weights assigned to each model of the ensemble are optimized to achieve a performance better than any single 3D CNN model. With a relatively smaller training dataset, the proposed model obtains 97.27%, 82.33%, 90.41%, 84.22%, 84.26%, and 77.1% accuracies for the Alzheimer’s disease (AD) versus cognitively normal (CN), early mild cognitive impairment (EMCI) versus CN, late mild cognitive impairment (LMCI) versus CN, EMCI versus AD, LMCI versus AD, and EMCI versus LMCI classification tasks, outperforming current state-of-the-art methods, and indicating the effectiveness of our proposed model. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd 2024.

关键词： Alzheimer’s disease Convolutional neural networks Ensemble learning Transfer 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：