检索结果-内蒙古大学图书馆

12th international Advanced computing conference, IACC 2022

作者： Deo, Siddhi Mehta, Simran Jain, Digha Tiwari, Charu Thorat, Aniket Mahara, Sudhanshu Gonge, Sudhanshu Joshi, Rahul Gite, Shilpa Kotecha, Ketan Department of Computer Science and Engineering Symbiosis Institute of Technology Pune India

ISBN: (纸本)9783031356438

Digital recordings are now affordable, quick to make, and straightforward to upload on web platforms because of the widespread availability of video recording technology in smartphones and other digital devices. The truth of films published on social media cannot be trusted. With the latest tools used for video editing tools, movies may now be readily changed (faked) for political ends or unlawful profit. Over the years, a lot of effort has gone into creating novel methods for identifying various forms of video manipulation. The objectives of video tampering detection include identifying signs of change and assessing the validity and integrity of the video file. These methods are divided into two groups: active & passive (blind) techniques. Based on the forgeries that passive video tampering detection systems can see, they may be divided into three groups: region tampering detection, double or multiple compressed video detection, and video inter-frame forgery detection. The previous studies under consideration are provided in a condensed tabular format, together with the criteria they employed and any limitations they had. The research work concluded by outlining a number of problems that might be used to inspire new research topics for the identification of passive video manipulation. In this paper, few AI/ML algorithms were tested for video tampering detection to find out which one gives out the most accurate results. © 2023, Springer Nature Switzerland AG.

关键词： Digital devices

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Epidemic Alert System using machine learning for Dengue and Chikungunya 10

Comparative Analysis of Epidemic Alert System using Machine ...

引用

10th international conference on Cloud computing, Data Science and Engineering (Confluence)

作者： Dhaka, Aahhas Singh, Prabhishek Amity Univ Uttar Pradesh Amity Sch Engn & Technol Dept CSE Noida India

ISBN: (纸本)9781728127910

The Rapid spread of a disease is known as an epidemic. The catastrophe brought by an epidemic not only effects the people of an area, hut also brings about a lot of distress in every sector of social strata. An epidemic alerting system has a potential to carve the path how medical surveillance could become more efficient. The epidemic causing diseases arc usually vector borne. The diseases arc spread by pathogens present in these vectors. An epidemic alerting system could predict how the weather conditions and several other factors effect the growth and propagation of these vectors. The weather conditions could be predicted using the high-end instruments and satellites currently available. Using this prediction, we could forecast the next targets of the epidemic. To implement this epidemic alert system, four algorithms are used namely Random Forest Regression, Decision Tree Regression, Support Vector Regression and Multiple Linear Regression. For dengue, the state wise cases data of the year 2013 to 2017 has been used in the system while for chikungunya the data used is of the year 2013 to 2016. This dataset has been downloaded from a government website, i.e., https://*** the case of dengue, the model has been trained on the data of the year 2013 to 2016 and predictions of the year 2017 have been done. On the other hand, the model has been trained on the data of the year 2013 to 2015 and predictions for the year 2017 have been made regarding Chikungunya. At last, a contrastive analysis has been made on the four algorithms used for both the diseases.

关键词： epidemic alert system dengue chikungunya machine learning

来源：评论

学校读者我要写书评

暂无评论

machine learning-Based Profiling Attack Method in RSA Prime Multiplication

Machine Learning-Based Profiling Attack Method in RSA Prime ...

引用

2020 ACM international conference on Intelligent computing and its Emerging Applications, ICEA 2020

作者： Park, Han-Byeol Sim, Bo-Yeon Han, Dong-Guk Dept. of Financial Information Security Kookmin University Korea Republic of Dept. of Mathematics Kookmin University Korea Republic of Dept. of Information Security Cryptology and Mathematics Kookmin University Korea Republic of

ISBN: (纸本)9781450383042

In this paper, we propose a machine learning-based profiling attack on the prime multiplication operation of RSA's key generation algorithm. The proposed attack takes advantage of the fact that a prime word value, which is the data storage unit, is loaded in the process of the multiplication operation for generating a modulus. We selected a commonly used product-scanning method as a multiplication algorithm. Then we collected the power consumption traces and constructed a profile of the secret prime value based on machine learning. In addition, the success rate of the attack was measured within a single trace to perform a realistic attack during the key generation operation. The secret prime values were derived with a maximum success rate of 99.8% in a single trace. Based on this, this paper suggests that if the secret value is an operand of the multiplication operation, there may be vulnerability against side-channel attacks because of the characteristics of the multiplication algorithm.1 © 2020 ACM.

关键词： machine learning

来源：评论

学校读者我要写书评

暂无评论

ParSecureML: An Efficient Parallel Secure machine learning Framework on GPUs 20

ParSecureML: An Efficient Parallel Secure Machine Learning F...

引用

49th international conference on Parallel Processing (ICPP)

作者： Chen, Zheng Zhang, Feng Zhou, Amelie Chi Zhai, Jidong Zhang, Chenyang Du, Xiaoyong Renmin Univ China Key Lab Data Engn & Knowledge Engn MOE Beijing Peoples R China Renmin Univ China Sch Informat Beijing Peoples R China Shenzhen Univ Guangdong Prov Engn Ctr China made High Performan Shenzhen Peoples R China Tsinghua Univ Dept Comp Sci & Technol BNRist Beijing Peoples R China

ISBN: (纸本)9781450388160

machine learning has been widely used in our daily lives. Large amounts of data have been continuously produced and transmitted to the cloud for model training and data processing, which raises a problem: how to preserve the security of the data. Recently, a secure machine learning system named SecureML has been proposed to solve this issue using two-party computation. However, due to the excessive computation expenses of two-party computation, the secure machine learning is about 2x slower than the original machine learning methods. Previous work on secure machine learning mostly focused on novel protocols or improving accuracy, while the performance metric has been ignored. In this paper, we propose a GPU-based framework ParSecureML to improve the performance of secure machine learning algorithms based on two-party computation. The main challenges of developing ParSecureML lie in the complex computation patterns, frequent intra-node data transmission between CPU and GPU, and complicated inter-node data dependence. To handle these challenges, we propose a series of novel solutions, including profiling-guided adaptive GPU utilization, fine-grained double pipeline for intra-node CPU-GPU cooperation, and compressed transmission for inter-node communication. As far as we know, this is the first GPU-based secure machine learning framework. Compared to the state-of-the-art framework, ParSecureML achieves an average of 32.2x speedup. ParSecureML can be downloaded from https://***/ZhengChenCS/ParSecureML.

关键词： Security GPU machine learning Two-Party Computation

来源：评论

学校读者我要写书评

暂无评论

LSTM-based Network Attack Detection: Performance Comparison by Hyper-parameter Values Tuning 7

LSTM-based Network Attack Detection: Performance Comparison ...

引用

7th IEEE international conference on Cyber Security and Cloud computing (CSCloud) / 6th IEEE international conference on Edge computing and Scalable Cloud (EdgeCom)

作者： Hossain, Md Delwar Ochiai, Hideya Fall, Doudou Kadobayashi, Youki Nara Inst Sci & Technol Div Informat Sci Ikoma Nara Japan Univ Tokyo Grad Sch Info Sci & Tech Tokyo Japan

ISBN: (纸本)9781728165509

Network attacks have been around since the beginning of the Internet and they are still relevant due to the numerous attempts of independent hackers, cybercrime organizations and state-sponsored hacking squads to intrude into others computer networks. Similarly, the industry and academia have been providing constantly evolving solutions to try to thwart the ever-sophisticated network attack techniques. machine learning (ML) and deep learning have shown promising results for detecting and preventing network attacks. In this paper, we investigate a Long Short-Term Memory (LSTM)-based network attack detection. Our goal is to find the optimal values for network attack detection by fine-tuning different LSTM hyper-parameters: optimizers, loss functions, learning rates and activation functions, and by comparing their performance by using the CICIDS2017 labeled dataset. Our results show that LSTM can effectively detect the network attacks with high accuracy and reasonable detection rates. With the optimized hyper-parameter values, the proposed LSTM model can detect DoS attacks with an effective detection accuracy of 99.08% and a detection rate of 0.93. We achieved BoT, DDoS, and port scan attacks detection accuracy of 99.54% and detection rate of 0.84.

关键词： Network Attack Detection Cybersecurity Deep learning Long Short-Term Memory Intrusion Detection System

来源：评论

学校读者我要写书评

暂无评论

Application of Quantum Generative Adversarial learning in Quantum Image Processing 2

Application of Quantum Generative Adversarial Learning in Qu...

引用

2nd international conference on Information Technology and Computer Application, ITCA 2020

作者： Ren, Wanghao Li, Zhiming Li, Hailing Li, Yang Zhang, Chunwei Fu, Xiaoqian University of Jinan School of Information Science and Engineering Jinan250022 China Shandong College of Electronic Technology Jinan250022 China

ISBN: (纸本)9780738111414

Quantum machine learning is a hot topic in the quantum community recently. The combination of quantum machine learning and image processing is also one of the researchers' concerns. With the promotion of deep learning to process images, the field of image processing has also shown amazing potential. The combination of quantum computing and artificial intelligence can not only exert the computing power of quantum computing but also find more applications in the field of image processing. This article combines quantum generative adversarial learning with the field of image processing. A quantum generative adversarial network is designed to load and learn classical image data. Numerical simulations intuitively show that quantum algorithms can also effectively process images. In our scheme, using N qubits can load 2N classical bits of image data. © 2020 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

A Deep learning Framework for Stock Prediction Using LSTM 1

引用

international conference on Decision Economics, DECON 2019

作者： Lin, Yaohu Liu, Shancun Yang, Haijun Wu, Harris School of Economics and Management Beihang University Beijing China Key Laboratory of Complex System Analysis Management and Decision Beihang University Ministry of Education Beijing China Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University Beijing China Strome College of Business Old Dominion University NorfolkVA United States

ISBN: (数字)9783030382278

ISBN: (纸本)9783030382261

In order to test the predictive power of the deep learning model, several machine learning methods were introduced for comparison. Empirical case results for the period of 2000 to 2017 show the forecasting power of deep learning technology. With a series of linear regression indicator measurement, we find LSTM networks outperform traditional machine learning methods, i.e., Linear Regression, Auto ARIMA, KNN. © Springer Nature Switzerland AG 2020.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

A User-Centered Active learning Approach for Appliance Recognition 6

A User-Centered Active Learning Approach for Appliance Recog...

引用

6th IEEE international conference on Smart computing (SMARTCOMP)

作者： Shin, Eura Khamesi, Atieh R. Bahr, Zachary Silvestri, Simone Baker, D. A. Univ Kentucky Dept Comp Sci Lexington KY 40506 USA Missouri Univ Sci & Technol Dept Psychol Sci Rolla MO 65409 USA

ISBN: (纸本)9781728169972

Smart homes offer new possibilities for energy management. One key enabler of these systems is the ability to monitor energy consumption at the appliance level. Existing approaches rely mainly on data from aggregated smart meter readings, but lack sufficient accuracy to recognize several appliances. Conversely, smart outlets are a suitable alternative since they can provide accurate electrical readings on individual appliances. Previous approaches for appliance recognition based on smart outlets use passive machine learning, which are deficient in the flexibility and scalability to work with highly heterogeneous appliances in smart homes. In this paper, we propose a stream-based active learning approach, called K-Active-Neighbors (KAN), to address the problem of appliance recognition in smart homes. KAN is an interactive framework in which the user is asked to label signatures of recently used appliances. Differently from previous work, we consider the realistic case in which the user is not always available to participate in the labeling process. Therefore, the system simultaneously learns the signatures and also the user willingness to interact with the system, in order to optimize the learning process. We develop an Arduino-based smart outlet to test our approach. Results show that, compared to previous solutions, KAN achieves higher accuracy in up to 41% less time.

关键词： Appliance Recognition Stream-based Active learning User-Centered machine learning Labeler Abstention

来源：评论

学校读者我要写书评

暂无评论

Deep and shallow feature fusion and recognition of recording devices based on attention mechanism 12th

Deep and shallow feature fusion and recognition of recording...

引用

12th international conference on Intelligent Networking and Collaborative Systems, INCoS 2020

作者： Zeng, Chunyan Zhu, Dongliang Wang, Zhifeng Yang, Yao Hubei Key Laboratory for High-Efficiency Utilization of Solar Energy and Operation Control of Energy Storage System Hubei University of Technology WuhanHubei430068 China Department of Digital Media Technology Central China Normal University WuhanHubei430079 China

ISBN: (纸本)9783030577957

In the device source identification task, there is a lot of redundant information in the traditional MFCC, GSV and i-vector features, which affects the accuracy of device source identification. In this paper, in order to extract the key information from the device source, we propose a multi-feature fusion attention mechanism network to achieve improved accuracy. First, the multi-feature fusion attention mechanism network we proposed directly compresses the MFCC, GSV and i-vector features that characterize the source of the device using convolutional neural networks, reducing the dimension and computation of the feature vector and improving the feature vector Representativeness Second, we use a feature attention mechanism to adaptively rescale feature characteristics by considering the interdependence between features. Finally, we also use a combination of deep learning and machine learning. The back-end uses SVM models for classification. Extensive experiments show that our proposed method achieves better accuracy in device source recognition tasks. © Springer Nature Switzerland AG 2021.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Improving deep learning performance with missing values via deletion and compensation

引用

NEURAL computing & APPLICATIONS 2020年第17期32卷 13233-13244页

作者： Sanchez-Morales, Adrian Sancho-Gomez, Jose-Luis Martinez-Garcia, Juan-Antonio Figueiras-Vidal, Anibal R. Univ Politecn Cartagena Dept Tecnol Informac & Comunicac Plaza Hosp1 Edificio Cuartel Antigones Murcia 30202 Spain Univ Carlos III Madrid Dept Teoria Senal & Comunicac Avda Univ 30 Madrid 28911 Spain

Missing values in a dataset is one of the most common difficulties in real applications. Many different techniques based on machine learning have been proposed in the literature to face this problem. In this work, the great representation capability of the stacked denoising auto-encoders is used to obtain a new method of imputating missing values based on two ideas: deletion and compensation. This method improves imputation performance by artificially deleting values in the input features and using them as targets in the training process. Nevertheless, although the deletion of samples is demonstrated to be really efficient, it may cause an imbalance between the distributions of the training and the test sets. In order to solve this issue, a compensation mechanism is proposed based on a slight modification of the error function to be optimized. Experiments over several datasets show that the deletion and compensation not only involve improvements in imputation but also in classification in comparison with other classical techniques.

关键词： Missing values Imputation Classification Deep 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：