检索结果-内蒙古大学图书馆

Optimal feature extraction and classification-oriented medical insurance prediction model: machine learning integrated with the internet of things

引用

International Journal of computers and Applications 2022年第3期44卷 278-290页

作者： Chowdhury, Subrata Mayilvahanan, P. Govindaraj, Ramya School of Computer Science Vels Institute of Science Technology & Advanced Studies TN Chennai India School of Information Technology and Engineering Vellore Institute of Technology TN Vellore India

This paper plans to develop an effective machine learning system integrated with the Internet of Things (IoT) to predict the health insurance amount. IoT in healthcare enables interoperability, machine-to-machine communication, information exchange, and data movement that make healthcare service delivery effective. The model includes three phases (a) Feature Extraction, and (b) Weighted Feature Extraction, and (c) Prediction. The feature extraction process computes two statistical measures: First Order Statistics like mean, median, standard deviation, the maximum value of entire data, and minimum value of entire data, and Second-Order Statistics like Kurtosis, skewness, correlation, and entropy. The prediction process deploys a renowned machine learning algorithm called Neural Network (NN). As the main contribution, the weighted feature vector is developed here, where the weight optimally tuned by Modified Whale Optimization Algorithm (WOA). Also, the contribution relies on NN, where the training algorithm replaced with the same modified WOA for weight update. The modified WOA developed here is termed as Fitness dependent Randomized Whale Optimization Algorithm (FR-WOA). At last, the valuable experimental analysis using three datasets confirms the efficient performance of the suggested model. © 2020 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Enhancing Forest Fire Detection: Utilizing VGG16 for High-Accuracy Image Classification and Machine Learning Integration 8

Enhancing Forest Fire Detection: Utilizing VGG16 for High-Ac...

引用

8th International Artificial Intelligence and Data Processing Symposium, IDAP 2024

作者： Tan, Muhammet Bakir, Halit Sivas University of Science and Technology Institute of Graduate Studies Sivas Turkey Sivas University of Science and Technology Department of Computer Engineering Sivas Turkey

ISBN: (纸本)9798331531492

Forest fires cause significant problems for nature, communities. They damage forests, animals' homes, and places where people live. The bad effects of forest fires go beyond just hurting the air, water, and animals. They also make climate change worse and mess up the balance of nature for a long time. In our efforts to address this significant issue, we utilize trained Convolutional Neural Networks (CNNs) algorithm VGG16 model to classify images containing fire and smoke versus non fire images. To enhance the training performance and the model's ability to generalize to unseen data, we curated a dataset comprising approximately 12,000 images from each category. We proposed using the features extracted the VGG16 model for training a simple neural network and multiple machine learning algorithms. Following the training process of the neural network model, our model achieved a training accuracy of 98.4231% and a validation accuracy of 99.0717%. Subsequently, to assess the robustness of our model, we subjected the output layer of our trained model to 13 different machine learning algorithms. Remarkably, all 6 models surpassed a performance threshold of 99% across all metrics. © 2024 IEEE.

关键词： Premixed flames

来源：评论

学校读者我要写书评

暂无评论

ParallelSFL: A Novel Split Federated Learning Framework Tackling Heterogeneity Issues 24

ParallelSFL: A Novel Split Federated Learning Framework Tack...

引用

30th International Conference on Mobile Computing and Networking, ACM MobiCom 2024

作者： Liao, Yunming Xu, Yang Xu, Hongli Yao, Zhiwei Huang, Liusheng Qiao, Chunming School of Computer Science and Technology University of Science and Technology of China China Suzhou Institute for Advanced Research University of Science and Technology of China China Department of Computer Science and Engineering University at Buffalo State University of New York United States

ISBN: (纸本)9798400704895

Mobile devices contribute more than half of the world's web traffic, providing massive and diverse data for powering various federated learning (FL) applications. In order to avoid the communication bottleneck on the parameter server (PS) and accelerate the training of large-scale models on resource-constraint workers in edge computing (EC) system, we propose a novel split federated learning (SFL) framework, termed ParallelSFL. Concretely, we split an entire model into a bottom submodel and a top submodel, and divide participating workers into multiple clusters, each of which collaboratively performs the SFL training procedure and exchanges entire models with the PS. However, considering the statistical and system heterogeneity in edge systems, it is challenging to arrange suitable workers to specific clusters for efficient model training. To address these challenges, we carefully develop an effective clustering strategy by optimizing a utility function related to training efficiency and model accuracy. Specifically, ParallelSFL partitions workers into different clusters under the heterogeneity restrictions, thereby promoting model accuracy as well as training efficiency. Meanwhile, ParallelSFL assigns diverse and appropriate local updating frequencies for each cluster to further address system heterogeneity. Extensive experiments are conducted on a physical platform with 80 NVIDIA Jetson devices, and the experimental results show that ParallelSFL can reduce the traffic consumption by at least 21%, speed up the model training by at least 1.36X, and improve model accuracy by at least 5% in heterogeneous scenarios, compared to the baselines. © 2024 Copyright is held by the owner/author(s). Publication rights licensed to ACM.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

'A Mobile Application for Women's Safety: Myguard' 5

'A Mobile Application for Women's Safety: Myguard'

引用

5th IEEE International Conference on Advances in Computing, Communication Control and Networking, ICAC3N 2023

作者： Mishra, Nitish Puri, Ananya Rishita Singh, Meeta Manav Rachna International Institute Of Research & Studies Department Of Computer Science Haryana Faridabad India

ISBN: (纸本)9798350330861

This paper addresses the issue of women's safety, which is a growing concern in many countries, particularly India. The police often face significant challenges in responding promptly to distress calls due to various constraints, including a lack of information regarding the location and existence of the crime. Additionally, women may find it difficult to discreetly and securely reach out to law enforcement. To address these problems, MyGuard, a revolutionary mobile app, has been developed to assist women in feeling more secure in their everyday lives. The app allows users to trigger a panic alert using a preset word with just a few taps. If the user fails to arrive at their destination on time, the app automatically notifies preselected emergency contacts and the closest police station with their geographic location. This paper provides an overview of the development and technical implementation of MyGuard, which is a game-changer for women's safety [16]. © 2023 IEEE.

关键词： Law enforcement

来源：评论

学校读者我要写书评

暂无评论

Enhancing Credit Card Fraud Detection With Sbpmc-Oa Based Feature Optimization

Enhancing Credit Card Fraud Detection With Sbpmc-Oa Based Fe...

引用

International Conference on Inventive Computation Technologies (ICICT)

作者： V. Suganthi J. Jebathangam Department of Computer Science Vels Institute of Science Technology & Advanced Studies(VISTAS) Chennai India

ISBN: (数字)9798331512248

ISBN: (纸本)9798331512255

Credit Card Fraud Detection (CCFD) protects cardholders from money loss during transactions. However, the prevailing works could not adapt to new attack techniques, leading to fraudulent transactions. Hence, the Bias Wrapper C1 Gated Recurrent Unit (BWC1-GRU) is proposed. Initially, to train the CCFD model, the credit card transaction details are collected and pre-processed. Next, the data augmentation is done using the Synthetic Heuristic Minority Over-Sampling Technique (S-HeMOTE) technique. After augmentation, the features are extracted and using the Secretary Bird Principle of Maximum Continuous Optimization Algorithm (SBPMC-OA), optimal features are selected. Then, the CCFD is done using the BWC1-GRU method. Thus, the proposed framework detected Credit Card Fraud (CCF) with an accuracy of 97.721 %, which outperformed the existing works.

关键词： Adaptation models Accuracy Feature extraction Credit cards Data augmentation Birds Data models Fraud Mobile applications Optimization

来源：评论

学校读者我要写书评

暂无评论

Facial Expression Based User Concentration Alert System

Facial Expression Based User Concentration Alert System

引用

2023 International Conference on Advances in Computation, Communication and Information Technology, ICAICCIT 2023

作者： Arora, Ginni Adlakha, Garima Joshi, Rashmi Batra, Mohit Sarkar, Tanisha Luthra, Meghna Manav Rachna International Institute of Research and Studies Department of Computer Science and Engineering Faridabad India

ISBN: (纸本)9798350344387

Nowadays, due to more social platform usage, the user concentration on major activities has been declined. It can be seen in the education sector as the students are unable to pay attention in online sessions. Therefore, there is a need of an automatic alert system to make a student engaged. This paper focus on implementing face detection, face recognition and face emotion recognition in the education sector. With the concepts of Machine Learning and Neural Networks such as ResNet, AlexNet, VGG-16 as pretrained network helps in recognizing the human face. The emotions of the recognized face are then analysed through the machine learning model i.e. face emotion detector which detects facial emotions like happy, sad, confused, focused, angry, etc. This facial expression system help to send alert to the parents and students for evaluation and for preventing them from getting distracted during sessions. © 2023 IEEE.

关键词： Deep Neural Network Emotion detection Face Expression Machine learning

来源：评论

学校读者我要写书评

暂无评论

MergeSFL: Split Federated Learning with Feature Merging and Batch Size Regulation 40

MergeSFL: Split Federated Learning with Feature Merging and ...

引用

40th IEEE International Conference on Data Engineering, ICDE 2024

作者： Liao, Yunming Xu, Yang Xu, Hongli Wang, Lun Yao, Zhiwei Qiao, Chunming University of Science and Technology of China School of Computer Science and Technology China Suzhou Institute for Advanced Research University of Science and Technology of China China University at Buffalo Department of Computer Science and Engineering The State University of New York United States

ISBN: (纸本)9798350317152

Recently, federated learning (FL) has emerged as a popular technique for edge AI to mine valuable knowledge in edge computing (EC) systems. To boost the performance of AI applications, large-scale models have received increasing attention due to their excellent generalized abilities. However, training and transmitting large-scale models will incur significant computing and communication burden on the resource-constrained workers, and the exchange of entire models may violate model privacy. To relax the burden of workers and protect model privacy, split federated learning (SFL) has been released by integrating both data and model parallelism. Despite resource limitations, SFL also faces two other critical challenges in EC systems, i.e., statistical heterogeneity and system heterogeneity. In order to address these challenges, we propose a novel SFL framework, termed MergeSFL, by incorporating feature merging and batch size regulation in SFL. Concretely, feature merging aims to merge the features from workers into a mixed feature sequence, which is approximately equivalent to the features derived from IID data and is employed to promote model accuracy. While batch size regulation aims to assign diverse and suitable batch sizes for heterogeneous workers to improve training efficiency. Moreover, MergeSFL explores to jointly optimize these two strategies upon their coupled relationship to better enhance the performance of SFL. Extensive experiments are conducted on a physical platform with 80 NVIDIA Jetson edge devices, and the experimental results show that MergeSFL can improve the final model accuracy by 5.82% to 26.22%, with a speedup by about 1.39× to 4.14×, compared to the baselines. © 2024 IEEE.

关键词： Merging

来源：评论

学校读者我要写书评

暂无评论

Natural Language Processing for Prediction of Disaster Tweets using Ensemble Learning

Natural Language Processing for Prediction of Disaster Tweet...

引用

2023 International Conference on Advances in Computation, Communication and Information Technology, ICAICCIT 2023

作者： Chowdhary, Mohit Nandal, Poonam Manav Rachna International Institute of Research and Studies Department of Computer Science and Engineering Faridabad India

ISBN: (纸本)9798350344387

People who have access to the internet now often utilize social media as a result of the growing digitization. The widespread usage of social media has considerably hastened the propagation of social events while also accelerating human connection. It seemed obvious that the broad and quick use of information would have advantages for social scientists and computer scientists. People's sentences on social media may be classified based on the emotional emotions they include, making it simple to utilize them to describe and forecast events. NLP (Natural Language Processing) approaches may carry out this estimate procedure. In this paper, we used the ensemble learning approach to evaluate tweets that included terms associated with catastrophes using various pre-processing techniques and TF-IDF (Term Frequency-Inverse Document Frequency) feature extraction method. We compared these classification algorithms (Gaussian Navies Bayes, Logistic Regression, Random Forest,) according to their f1 scores, accuracy, precision and Recall. © 2023 IEEE.

关键词： Ensemble Learning Information Retrieval Natural Language Processing. Prediction

来源：评论

学校读者我要写书评

暂无评论

Performance Analysis of Liver Disease Prediction Using Supervised Machine Learning Classification Models 1

Performance Analysis of Liver Disease Prediction Using Super...

引用

1st International Conference on advanced Computing and Emerging Technologies, ACET 2024

作者： Singh, Ramveer Sharma, Rahul Chaudhary, Shalbha Yadav, Deepika Jain, Jaishree Kumar, Pawan Raj Kumar Goel Institute of Technology Department of Computer Science & Engineering UP Ghaziabad India Raj Kumar Garg Institute of Technology Department of Computer Science & Engineering UP Ghaziabad India R.D. Engineering College Department of Management Studies UP Ghaziabad India Bharati Vidyapeeth's College of Engineering Department of Computer Science & Engineering New Delhi India Teerthanker Mahaveer University Department of Computer Science & Engineering UP Moradabad India Satyug Darshan College of Engineering Department of Computer Science Faridabad India

ISBN: (纸本)9798350367720

Liver is most fundamental part of human body. It is highly dependable to performing several body functions such as helping in food digestion, waste filtering, enzyme activation, detoxification, nutrition and glycogen storage, carbohydrate metabolism, vitamin and mineral storage etc. It has been found from various survey reports that across the world, 58 million people are suffering from liver disorders and nearly 01 million liver disorder patients are present in India. The growth rate of liver disorder patients is increasing rapidly worldwide due to a sharp rise in obesity, eating more junk foods, more alcohol consumption and unhealthy lifestyles. People who have liver disorders problem need medical treatment at the correct time and early disease prediction helps to reduce the number of deaths. The research focuses on designing an intelligent system that helps for prediction liver disorder by using various classification models of supervised machine learning and it also helps to prevent becoming India a global hub of liver diseases by 2030. The research emphasis on comparative performance analysis of different classification algorithms and ensemble methods. It also helps to select best accurate model for prediction of the liver diseases. © 2024 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Utilizing Gradient Boosting Models to Identify Risk Factors for Stroke 2

Utilizing Gradient Boosting Models to Identify Risk Factors ...

引用

2nd IEEE International Conference on Advances in Information Technology, ICAIT 2024

作者： Tamilselvi, P. Harini, C. Rao, Guduri Padma Narasimha Raju, P.V. Subbulakshmi, R. Channabasava, Ugranada Department of Computer Science Vels Institute of Science Technology and Advanced Studies Chennai India Maisammaguda Secunderabad India SRKR Engineering College Andhra Pradesh Bhimavaram India Sagi Rama Krishnam Raju Engineering College Bhimavaram India Department of Computer Science and Engineering Karpagam Institute of Technology Coimbatore India Department of Information Science and Engineering Don Bosco Institute of Technology VTU Karnataka Belagavi India

ISBN: (纸本)9798350383867

This research looks at the usage of the models of gradient boosting to find out the ones that are relevant in the stroke incidence. With the help of a large dataset ranging from demographics to clinical and imaging characteristics, the three most effective gradient boosting algorithms (XGBoost with LightGBM and CatBoost) were evaluated in terms of their ability for stroke risk prediction. Model performance metrics, including accuracy, area under the receiver operating characteristic curve (AUC-ROC), sensitivity, specificity, precision, and F1 score, were estimated through re-scheduling validation proses. Through feature importance analysis, I discovered which of predictors, such as age, hypertension, diabetes and others contribute to the stroke risk prediction and where age and hypertension occupy the highest positions. The result of external validation in separate cohorts was the confirmation of the predictive properties of the data assortment as well as the generalizability of the presented models in variety of populations. Also, the role of these models has been examined from the perspective of their clinical usefulness and ability to inform the implementation of targeted preventive interventions that will result in improved outcomes regarding heart health. We evaluated various lifestyle adjustments, medications and precision monitoring plans as being the most likely to have a positive effect on stroke rates and patient outcomes. Overall the present study proves the problem-solving capability of the gradient boosting model in a tailored risk assessment of stroke and stroke prevention, enabling us to implement the necessary interventions to minimize the impact of the diseases on the healthcare systems and individuals. © 2024 IEEE.

关键词： Risk assessment

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：