检索结果-内蒙古大学图书馆

15th International Conference on Advances in Computing, Control, and Telecommunication Technologies, ACT 2024

作者： Wankhede, Disha Athanikar, Abhishek Borude, Yash Chougule, Raviraj Diware, Om Department of Computer Engineering Vishwakarma Institute of Information Technology India

ISBN: (纸本)9798331300579

Software programmers may get gradually precise at expecting consequences without being clearly coded using machine learning techniques. Machine learning is based on the idea that models and algorithms may collect input data then utilize statistical investigation to determine an output, while updating results as information become obtainable, as underlying principle. For example, models may be adapted and trained to meet management expectations so that correct measures are followed to reach a certain goal. Wall Mart, a one-stop shopping mall, has been used in this system to estimate sales of various products and to study impact that various variables have on sales of products. Predictive models may be built using different features of a Wall Mart dataset and methods used to construct them, and these findings can be used to make better business choices. This Predictive Algorithm is used in many fields over the market world for their profitability. Recently, Lot of companies are going to used this Machine Learning for better AI and new technologies in market. This report Includes combination of machine learning algorithm and Features engineering to predict Walmart sales. The research topic of this paper primarily solves the problems of Walmart sales forecasting that is provided by the Kaggle competition. © Grenze Scientific Society, 2024.

关键词： Input output programs

来源：评论

学校读者我要写书评

暂无评论

Enhanced Non-Defended Website Fingerprinting Attack Model Using Deep Learning

Enhanced Non-Defended Website Fingerprinting Attack Model Us...

引用

2024 International Conference on Intelligent Systems for Cybersecurity, ISCS 2024

作者： Sinha, Minal Dave, Mayank National Institute of Technology Department of Computer Engineering Kurukshetra India

ISBN: (纸本)9798350375237

Onion routing (Tor) is a routing protocol that uses multiple-layer encryption to protect privacy and security for authentic users during web browsing. The website fingerprinting (WF) attack, which may be used to identify which websites a user is browsing are one of the vulnerabilities present in the Tor browser, despite its security. There are many modern and advanced attack models to build a WF attack model but in this work, Convolutional Neural Network (CNN) is presented considering two different scenarios Close World (CW) and Open World (OW). In CW scenario, the proposed attack model attains 98.20% accuracy on traffic traces without utilizing any defense model and when the same attack model is used in the OW scenario, it attains 97.10% accuracy. In OW scenario, the precision comes out to be 0.99 and the recall is 0.90, even in the presence of imbalanced data without any defenses. Overall, in this work, the proposed non-defended attack model can identify monitored websites with a precision of 0.99, even in the real OW scenario. © 2024 IEEE.

关键词： Websites

来源：评论

学校读者我要写书评

暂无评论

Hybrid Steganographic and Encryption Approach using Color Analysis for Secure Image Communication 2

Hybrid Steganographic and Encryption Approach using Color An...

引用

2nd International Conference on Advances in Computation, Communication and Information technology, ICAICCIT 2024

作者： Jadhav, Swati Darekar, Smeet Deshmukh, Anurag Deore, Bhavesh Datir, Dev Department of Computer Engineering Vishwakarma Institute of Technology Pune India

ISBN: (纸本)9798331541217

Digital privacy and secure communication are under increasing threats, making strong methods of protection of sensitive data ever more important. Steganography is the art of concealing information within seemingly innocuous media, but the traditional techniques usually cannot resist advanced steganalysis methods. One such hybrid steganographic method has been proposed, embedding encryption onto color analysis for serving the dual purpose of ensuring more security and making the data hidden less perceivable. This technique tries to enhance both imperceptibility and resistance towards detection by embedding encrypted messages into strategically selected parts of the image where modifying is less likely to be detected by both the naked eye observer and the automated detection mechanisms. This is achieved by carrying out a color analysis that selects the optimal embedding regions with regard to texturing complexity and luminance contrast. In this work, a proposal for secure transmission is also developed using steganalysis keys that allow only the authorized parties to decrypt and reach the hidden information. Consequently, the strong point of this approach considers this to be a highly secure and imperceptible means for covert data communication, suitable for applications in the military, government, and digital watermarking. © 2024 IEEE.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

PerceptiGuard: Unveiling DeepFakes Through Enhanced Media Perception

PerceptiGuard: Unveiling DeepFakes Through Enhanced Media Pe...

引用

2024 IEEE Silchar Subsection Annual Conference, SILCON 2024

作者： Phatangare, Sheetal Tekade, Om Thakare, Prathamesh Wankhade, Abhishek Wankhade, Purva Vishwakarma Institute of Technology Department of Computer Engineering Pune India

ISBN: (纸本)9798331540821

This research presents a comprehensive approach to developing a multi-modal deepfake detection system leveraging advanced deep learning architectures. The primary objectives include the creation of a robust detection model, evaluation across various manipulation techniques, and integration into practical applications. Utilizing state-of-the-art neural networks, the system aims to enhance detection accuracy and offer a user-friendly interface for real-world deployment. The proposed architecture incorporates preprocessing steps such as video splitting, detection, and face cropping, followed by feature extraction and video classification using a combination of ResNet and LSTM networks. The model's performance is assessed through confusion matrix evaluation, ensuring reliability across diverse deepfake scenarios. © 2024 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Audio Classification Using Ensemble Model

Audio Classification Using Ensemble Model

引用

2024 IEEE International Conference on Intelligent Signal Processing and Effective Communication Technologies, INSPECT 2024

作者： Gupta, Neelabh Ahlawat, Priyanka National Institute of Technology Department of Computer Engineering Kurukshetra India

ISBN: (纸本)9798350379525

Audio classification is essential for numerous ap-plications, including environmental sound monitoring, speech recognition systems and music genre classification. The ability to accurately classify audio data improves the functionality of these systems, leading to improved user experiences and more effective automation. Accurate audio classification can improve emergency response systems by detecting distress sounds, en-hance smart home devices with better sound recognition and advance healthcare through precise classification of respiratory sounds. Machine learning has transformed audio classification by enabling systems to automatically learn patterns in audio data, enhancing applications like speech recognition and environmental sound monitoring. Convolutional Neural Networks(CNNs), in particular, have proven highly effective for audio classification by extracting hierarchical features from audio spectrograms. This paper proposes an efficient ensemble model involving VGG16, ResNet50 and DenseNet121. The audio data are first converted to spectrograms, which are then used as inputs for the CNN models. The individual pre-trained models are modified to adapt their classifiers to the specific audio classification task, with the number of output classes determined by the dataset. The ensemble model combines the outputs from the three networks, incorporating batch normalization and dropout to improve generalization. By leveraging the complementary strengths of VGG16, ResNet50 and DenseNet121, the ensemble model shows an impressive accuracy of 97.4% on the UrbanSound8K dataset. © 2024 IEEE.

关键词： Speech enhancement

来源：评论

学校读者我要写书评

暂无评论

Using Unsupervised Learning to detect Fraud in Cryptocurrency 3

Using Unsupervised Learning to detect Fraud in Cryptocurrenc...

引用

3rd IEEE International Conference on Blockchain and Distributed Systems Security, ICBDS 2024

作者： Chordia, Sankalp Nilesh Shinde, Sandeep Vishwakarma Institute of Technology Department of Computer Engineering Pune India

ISBN: (纸本)9798350354348

Bitcoin, the most popular form of cryptocurrency, operates independently of central banks or governmental entities. The emphasis on anonymity complicates the task of tracing cryptocurrency payments linked to illicit activities or criminal transactions. Consequently, ransomware developers exploit this environment to extort ransom by encrypting valuable user information. Numerous research endeavors have explored the application of machine learning techniques to detect fraudulent activities within cryptocurrency transactions. However, there is a notable absence of studies showcasing the practical implementation of such methods using real-time blockchain data. The present study is structured into two main sections. Firstly, we delineate the development of various features utilized for training purposes. Subsequently, our approach will be validated using real-time Bitcoin data. The other section highlights the use of open source investigation tool (OSINT) to carry forward the cryptocurrency investigation by taking the input from previous findings. OSINT involves the retrieval of information from a range of publicly accessible data sources, while offering a viable approach to address the difficulties associated with extracting and aggregating intelligence from diverse publicly available information channels and social platforms. Thus narrowing down the search investigation by exploiting the pseudonymity of Bitcoin and tracing back the wallet and IP addresses' digital and physical identity. © 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

Forecasting Renewable Energy Generation: A Predictive Modelling Approach 2

Forecasting Renewable Energy Generation: A Predictive Modell...

引用

2nd International Conference on Self Sustainable Artificial Intelligence Systems, ICSSAS 2024

作者： Mondal, Priotosh Bhatia, Aditi Panjwani, Roshini Dokare, Indu Ves Institute of Technology Department of Computer Engineering Mumbai India

ISBN: (纸本)9798350368413

India's rapid urbanization demands innovative solutions to address energy consumption patterns while reducing reliance on fossil fuels. This research paper explores the application of predictive modelling techniques like Multilayer Perceptron and Polynomial Regression to forecast solar energy generation and Long Short-Term Memory (LSTM) models to forecast wind energy generation, thereby facilitating efficient energy planning in major Indian cities. The proposed system aims to create a web-based platform that integrates these predictive models to display real-time temperature conditions and the corresponding amount of energy that solar and wind sources can provide in specific locations, thus promoting smart cities and smart homes. By utilizing the website, energy planners will be able to compare the generated energy from wind and solar sources, enabling informed decisions on which resource best meets the energy requirements of housing settlements. © 2024 IEEE.

关键词： Wind forecasting

来源：评论

学校读者我要写书评

暂无评论

Sign Language Translation Across Multiple Languages 4

Sign Language Translation Across Multiple Languages

引用

4th International Conference on Emerging Systems and Intelligent Computing, ESIC 2024

作者： Antad, Sonali M. Chakrabarty, Siddhartha Bhat, Sneha Bisen, Somrath Jain, Sneha Vishwakarma Institute of Technology Department of Computer Engineering Pune India

ISBN: (纸本)9798350349856

Sign Language serves as the only mode of communication for people who are speech and hearing disabled. Despite this, they face many difficulties in communication due to different sign languages across the world using different signs and symbols. This project aims to bridge this gap and make communication easy and accessible for all. Through the utilization of Convolutional Neural Networks (CNN) and other deep learning algorithms, the project provides a platform to people to choose between Indian Sign Language and American Sign Language and further provides translation to multiple Indian regional languages to improve communication between people. This serves a two-way purpose by enabling people who are trained in different sign languages to communicate efficiently as well as providing written text in multiple languages for people who cannot interpret sign language and are proficient only in regional languages. This comprehensive strategy strives to eliminate communication barriers, contributing to a more equitable society. This innovative project has the potential to significantly enhance communication inclusivity and accessibility for the global deaf and hard of hearing community. © 2024 IEEE.

关键词： Audition

来源：评论

学校读者我要写书评

暂无评论

Stock Indexes Community Identification Using BAT-Modified Optimization Algorithm

引用

SN computer Science 2024年第8期5卷 1-14页

作者： Aggarwal, Kirti Arora, Anuja Department of Computer Science and Engineering and Information Technology Jaypee Institute of Information Technology Noida India

Stock price accelerates interest and preference of the young generation to explore the stock market with elicit interest. An autopilot system is needed where users choose beneficial stocks of their choice without paying attention to the manual inspection or implication. This study aims to form highly correlated stock communities based on fluctuations in stock price which will be helpful in the prediction and decision-making of users about their expenditure in the stock market. For the enormous volume of stock price time series data, the opted data representation method is the network which is generated with the help of a data statistical correlation test. Another research challenge was to detect the community in an optimized manner due to complex stock networks. For the same, BAT optimization algorithm and a proposed BAT-modified (BAT-M) algorithm are applied for complex stock network community formation. The better community formation results showcase that BAT-M gives better performance in comparison to the standard BAT algorithm on various performance evaluation factors. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： BAT Community detection Correlation Modularity Nature-inspired Optimization Stock indexes Stock network

来源：评论

学校读者我要写书评

暂无评论

Satellite Image to Map Translation using GANs 15

Satellite Image to Map Translation using GANs

引用

15th International Conference on Advances in Computing, Control, and Telecommunication Technologies, ACT 2024

作者： Phatangare, Sheetal Khalifa, Mohammed Mansoor Kharche, Sejal Khatib, Amaan Kshirsagar, Atharva Vishwakarma Institute of Technology Computer Engineering Department Pune India

ISBN: (纸本)9798331300579

Generative adversarial networks (GANs) are a promising method for learning deep representations without spending a lot of time on data for annotated training. They accomplish this by obtaining signals that are back propagated through a competitive mechanism in-transforming two networks. The possible representations of a variety of applications may employ the knowledge gathered by GANs, image synthesis, semantic image editing, fashion categorization, transmission, and image super-resolution. The goal of this research article is to give an overview of GANs. the signal processing community, utilizing well-known whenever feasible, use concepts and analogies. Additionally, to recognize various training techniques and building GANs, we also highlight lingering issues with their theory and utilization. © Grenze Scientific Society, 2024.

关键词： Gallium nitride

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：