检索结果-内蒙古大学图书馆

IEEE International Conference on Big Data

作者： Xinyang Wu Andrew Horner Dept. Computer Science and Engineering Hong Kong University of Science and Technology Hong Kong

ISBN: (数字)9798350362480

ISBN: (纸本)9798350362497

Music mashups integrate elements from different songs to create surprising and engaging listening experiences. Typically, a mashup combines the vocal track of a base song with the instrumental tracks of complementary songs. Automating the production of mashups has been an area of research for decades. Traditional approaches utilize rule-based methods, such as matching tempo and harmonic similarity, to select optimal segments for mashup generation. More recent techniques leverage neural networks to classify segment compatibility. However, both approaches primarily focus on layering segments that are generally compatible, without ensuring their detailed integration and alignment with the vocal track. In contrast, we introduce a novel approach using Graph Neural Networks (GNNs) that learns to rearrange instrumental segments to better align with the base vocal track, resulting in more surprising and accurate music blends for mashup generation. Additionally, we conducted subjective listening tests to evaluate our generated mashups against a baseline model using the same song pairs and the original base songs, assessing our model’s performance. Generated mashups used for evaluation can be found in https://***/***/.

关键词： Accuracy Mashups Instruments Music Production Big Data Harmonic analysis Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

AdR-Gaussian: Accelerating Gaussian Splatting with Adaptive Radius

arXiv

引用

arXiv 2024年

作者： Wang, Xinzhe Yi, Ran Ma, Lizhuang Dept. of Computer Science and Engineering Shanghai Jiao Tong University China

3D Gaussian Splatting (3DGS) is a recent explicit 3D representation that has achieved high-quality reconstruction and real-time rendering of complex scenes. However, the rasterization pipeline still suffers from unnecessary overhead resulting from avoidable serial Gaussian culling, and uneven load due to the distinct number of Gaussian to be rendered across pixels, which hinders wider promotion and application of 3DGS. In order to accelerate Gaussian splatting, we propose AdR-Gaussian, which moves part of serial culling in Render stage into the earlier Preprocess stage to enable parallel culling, employing adaptive radius to narrow the rendering pixel range for each Gaussian, and introduces a load balancing method to minimize thread waiting time during the pixel-parallel rendering. Our contributions are threefold, achieving a rendering speed of 310% while maintaining equivalent or even better quality than the state-of-the-art. Firstly, we propose to early cull Gaussian-Tile pairs of low splatting opacity based on an adaptive radius in the Gaussian-parallel Preprocess stage, which reduces the number of affected tile through the Gaussian bounding circle, thus reducing unnecessary overhead and achieving faster rendering speed. Secondly, we further propose early culling based on axis-aligned bounding box for Gaussian splatting, which achieves a more significant reduction in ineffective expenses by accurately calculating the Gaussian size in the 2D directions. Thirdly, we propose a balancing algorithm for pixel thread load, which compresses the information of heavy-load pixels to reduce thread waiting time, and enhance information of light-load pixels to hedge against rendering quality loss. Experiments on three datasets demonstrate that our algorithm can significantly improve the Gaussian Splatting rendering speed. Copyright © 2024, The Authors. All rights reserved.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

Efficient Host Threat Detection Using The Procsage Method

Efficient Host Threat Detection Using The Procsage Method

引用

Emerging Smart Computing and Informatics (ESCI), Conference on

作者： Shashikumar D R Madhura N S Lekha Reddy Navya M Purav S Shreyas H R Dept. Computer Science & Engineering Sai Vidya Institute of Technology Bangalore India

ISBN: (数字)9798331515683

ISBN: (纸本)9798331515690

Advanced Persistent Threats (APTs) pose a significant challenge to modern cybersecurity due to their stealth and complexity. This paper introduces a scalable host threat detection framework that utilizes provenance graphs and graph representation learning to overcome limitations of traditional defense mechanisms. By optimizing process- and thread-centric analyses within provenance graphs, the system minimizes computational overhead while maintaining robust behavioral insights. Leveraging semantic and topological feature extraction with GraphSAGE-based learning, the framework achieves precise anomaly detection via clustering techniques. Validation on benchmark datasets demonstrates substantial improvements in accuracy, efficiency, and scalability, showcasing the framework's potential for real-world applications.

关键词： Representation learning Computational modeling Scalability Instruction sets Semantics Benchmark testing Feature extraction Threat assessment computer security Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

A Comparative Analysis of Machine Learning Prediction Algorithms for Detecting IoT Botnet Activities

A Comparative Analysis of Machine Learning Prediction Algori...

引用

Intelligent Systems for Cybersecurity (ISCS), International Conference on

作者： Happy Rita Chhikara Neeti Kashyap Dept. of Computer Science & Engineering The NorthCap University Gurugram Haryana India

ISBN: (数字)9798350375237

ISBN: (纸本)9798350375244

As the prevalence of Internet of Things (IoT) products keeps increase, the need for robust security measures becomes increasingly vital. Our paper addresses this concern by conducting a detailed comparative analysis of various machine learning methods, assessing their effectiveness in detecting and predicting malicious activities associated with IoT botnets. This paper meticulously examines initial identification methods for IoT botnet operations using advanced machine learning (ML) prediction techniques. To achieve this, we utilize the CICIoT2023 Dataset, a real-world IoT dataset obtained from networks that captures diverse device interactions and communication patterns. This dataset acts as the framework for constructing and evaluating numerous machine learning techniques, including support vector machines (SVM), k-nearest neighbours (k-NN), Naive Bayes, random forest (RF), logistic regression (LR), and decision trees (DT) approaches. Performance metrics such as accuracy, precision, F1-score, ROC curve, confusion matrix, and recall are employed to gain insights into the algorithms’ capabilities in botnet detection. Furthermore, this paper delves into an examination of the trade-offs between computational complexity and detection accuracy. This analysis aids in selecting the most suitable ML techniques tailored to specific IoT security scenarios. This comparative analysis lays the groundwork for the advancement of IoT botnet discovery strategies, providing essential insights to researchers, practitioners, and industry experts working to strengthen IoT ecosystems against growing cyber threats. We anticipate that our findings will spark more conversations and developments in the sector, promoting the establishment of more robust and adaptable security measures across the IoT landscape.

关键词： Support vector machines Radio frequency Machine learning algorithms Accuracy Botnet Oral communication Prediction algorithms

来源：评论

学校读者我要写书评

暂无评论

An Advanced computer Vision Model for Aiding Automobile in Traffic Sign Classification

An Advanced Computer Vision Model for Aiding Automobile in T...

引用

Computing and Data science (ICCDS), International Conference on

作者： Veera Satya Pavan Vignesh Kukkadapu Sai Prajith Goddati Parthav Gokul Cherukuri Adarsh Goluguri NV Rajareddy Dept. of Computer Science & Engineering GITAM Deemed to be University Visakhapatnam India

ISBN: (数字)9798350365337

ISBN: (纸本)9798350365344

Accurate traffic sign recognition is crucial for autonomous vehicles. This paper proposes a novel deep learning approach for sign classification that addresses limitations in training data. Our method employs a three-step process: strategic data augmentation with rotation limitations for realistic variations, feature extraction using parallel atrous convolution layers with varying dilation rates to capture multi-scale information, and robust feature map generation through output concatenation. This approach achieves a remarkable accuracy of 94.65%, surpassing the performance of established pre-trained models like VGG-16, ResNet-50, and AlexNet. This research contributes a significant advancement in real-world traffic sign classification in autonomous vehicles.

关键词： Training Solid modeling Accuracy Navigation Computational modeling Roads Transportation

来源：评论

学校读者我要写书评

暂无评论

A Likelihood Ratio Test of Genetic Relationship among Languages

arXiv

引用

arXiv 2024年

作者： Akavarapu, V.S.D.S. Mahesh Bhattacharya, Arnab Dept. of Computer Science and Engineering Indian Institute of Technology Kanpur India

Lexical resemblances among a group of languages indicate that the languages could be genetically related, i.e., they could have descended from a common ancestral language. However, such resemblances can arise by chance and, hence, need not always imply an underlying genetic relationship. Many tests of significance based on permutation of wordlists and word similarity measures appeared in the past to determine the statistical significance of such relationships. We demonstrate that although existing tests may work well for bilateral comparisons, i.e., on pairs of languages, they are either infeasible by design or are prone to yield false positives when applied to groups of languages or language families. To this end, inspired by molecular phylogenetics, we propose a likelihood ratio test to determine if given languages are related based on the proportion of invariant character sites in the aligned wordlists applied during tree inference. Further, we evaluate some language families and show that the proposed test solves the problem of false positives. Finally, we demonstrate that the test supports the existence of macro language families such as Nostratic and Macro-Mayan. © 2024, CC BY.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Metaheuristic Techniques Optimised LSTM Network for Improved Weather Prediction

Metaheuristic Techniques Optimised LSTM Network for Improved...

引用

Interdisciplinary Approaches in Technology and Management for Social Innovation (IATMSI), IEEE Conference on

作者： Shweta Mittal Om Prakash Sangwan Dept. of Computer Science and Engineering Guru Jambheshwar University of Science and Technology Hisar India

Tremendous amount of meteorological data is being generated on a daily basis from a number of sources such as weather stations, balloons, satellites, sensors etc. Timely weather prediction helps people plan everyday life events. Long Short Term Memory (LSTM) networks perform extremely well for capturing dependencies in time series datasets. Number of neurons in hidden layer highly impacts the performance of the LSTM network. The hit and Trial method for selecting the neurons consumes a lot of time and resources and might not lead to a global optimum solution. In this research work, two techniques, i.e. Genetic Algorithm optimized LSTM (LSTM_GA) and Artificial Bee Colony optimized LSTM (LSTM_ABC) have been proposed and implemented to automate the selection of the hidden layer’s neurons for improved weather prediction. Proposed techniques have been implemented using DeepLearning4j library making the techniques scalable on clusters of machines. For evaluating the performance of proposed techniques, 15 years Brazil’s weather dataset has been used. The experimental results proved that the proposed LSTM_GA and LSTM_ABC techniques have the reduced MAE value of 0.0074 and 0.0075 respectively while the LSTM network without any optimization has a relatively higher MAE value of 0.0086.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Deepfakes - Generating Synthetic Images, and Detecting Artificially Generated Fake Visuals Using Deep Learning

Deepfakes - Generating Synthetic Images, and Detecting Artif...

引用

International Conference on Confluence The Next Generation Information Technology Summit (Confluence)

作者： Ayushi Mishra Aadi Bharwaj Aditya Kumar Yadav Khushi Batra Nidhi Mishra Dept. of Computer Science and Engineering Amity University Noida Noida India

The abstract delves into the groundbreaking realm of DeepFakes, a revolutionary synthesis of deep learning and synthetic media. Central to DeepFake generation is the utilization of Generative Adversarial Networks (GANs), particularly the innovative mechanisms of face reenactment involving DCGANs (Deep Convolutional GANs) and Autoencoders. These technologies empower a generator to transform random noise into hyper-realistic visuals, capturing intricate details such as facial expressions and lighting conditions through latent space interpolation. The adversarial interplay between generator and discriminator continuously refines the authenticity of the generated content. While DeepFakes unlocks creative possibilities for artists and filmmakers, the abstract underscores the ethical concerns surrounding misinformation, privacy breaches, and trust erosion in digital media. The discourse navigates the delicate balance between creative freedom and responsible use, highlighting how DeepFakes, with their roots in advanced deep learning techniques, redefine our perception of synthetic media, challenging notions of reality in our increasingly digital world.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Leveraging Artificial Intelligence for Enhanced Data Generation in Addressing Imbalance in Binary Classification System 2

Leveraging Artificial Intelligence for Enhanced Data Generat...

引用

2nd IEEE International Conference on Interdisciplinary Approaches in Technology and Management for Social Innovation, IATMSI 2024

作者： Bandopadhyay, Srijita Dutta, Srimonti Haider, Imran Anuraag, Bhavaraju Zhu, Jerry Bazaz, Saad Ahmed University of Engg. & Management Dept. of Electronics and Communication Engineering Kolkata India Indian Institute of Technology Guwahati School of Agro and Rural Technology Guwahati India Indian Institute of Technology Kharagpur Dept. of Industrial and Systems Engineering Kharagpur India Vellore Institute of Technology Dept. of Electronics and Communication Engineering Vellore India University of Waterloo Faculty of Mathematics ON Canada National University of Computer & Emerging Sciences Dept. of Computer Science Islamabad Pakistan

ISBN: (纸本)9798350360523

This paper delves into the challenges of binary classification using imbalanced datasets, particularly when instances of interest are infrequent. It explores a comprehensive approach that integrates Synthetic Minority Over-sampling Technique (SMOTE), Generative Adversarial Networks (GANs), and Variational Autoencoders (VAEs) to enhance classification outcomes. Traditional classification models tend to favor the majority class, while the impact of imbalanced misclassification costs is often overlooked. The integration of SMOTE, GANs, and VAEs in binary classification, or SMOTE-GAN-VAE, addresses these challenges by generating synthetic instances, refining data representations, and capturing latent features. To evaluate the effectiveness of various data generation methods, a credit card fraud dataset is used. The performance metrics considered include F0.5-score, F1-score, and F2-score, which account for both precision and recall. The results indicate that SMOTE-GAN-VAE outperforms individual methods, such as SMOTE, GANs, and VAEs, demonstrating its potential to enhance data representation and classification accuracy, and outperformed the ß- VAE filtered approach employed in previous literature. © 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

Deciphering Deception: Unmasking Fake News in Multilingual Contexts

Deciphering Deception: Unmasking Fake News in Multilingual C...

引用

Computing, Power and Communication Technologies (IC2PCT), IEEE International Conference on

作者： Ashutosh Agarwal Yash Pratap Singh Vipin Rai Dept. of Computer Science & Engineering Galgotias University Greater Noida India

A major global worry in the ever-expanding digital ecosystem is the spread of incorrect information. The ability to recognize false news in non-English languages is still lacking, despite great advancements in the identification of false news in English. This research looks at the unique opportunities and problems that these languages bring. It offers a thorough assessment and comparative analysis of existing approaches, highlighting their advantages and disadvantages. By doing comparative study of various differ techniques we have found that Multilingual-Fake Model currently works the best. The results of the study highlight how critical it is to use inclusive and linguistically varied tactics in order to successfully counter disinformation. It emphasizes how crucial it is to create trustworthy tools and processes that can adjust to the subtleties of many languages and their socio-cultural settings. It highlights how crucial it is to comprehend these components completely in order to create detection methods that are more potent. The study admits that the dissemination of incorrect information is an issue that is deeply rooted in the social environment of linguistic groups and transcends language borders. A discussion of possible future study areas closes the report. In a multilingual culture, this study marks a substantial breakthrough toward the development of more thorough and efficient techniques for spotting bogus news.

关键词：

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：