检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Sarmin, Fatima Jahan Sarkar, Atiquer Rahman Wang, Yang Mohammed, Noman Dept. of Computer Science University of Manitoba Canada Dept. of Computer Science and Software Engineering Concordia University Canada

Synthetic data has been considered a better privacy-preserving alternative to traditionally sanitized data across various applications. However, a recent article challenges this notion, stating that synthetic data does not provide a better trade-off between privacy and utility than traditional anonymization techniques, and that it leads to unpredictable utility loss and highly unpredictable privacy gain. The article also claims to have identified a breach in the differential privacy guarantees provided by PATE-GAN and PrivBayes. When a study claims to refute or invalidate prior findings, it is crucial to verify and validate the study. In our work, we analyzed the implementation of the privacy game described in the article and found that it operated in a highly specialized and constrained environment, which limits the applicability of its findings to general cases. Our exploration also revealed that the game did not satisfy a crucial precondition concerning data distributions, which contributed to the perceived violation of the differential privacy guarantees offered by PATE-GAN and PrivBayes. We also conducted a privacy-utility tradeoff analysis in a more general and unconstrained environment. Our experimentation demonstrated that synthetic data indeed achieves a more favorable privacy-utility trade-off compared to the provided implementation of k-anonymization, thereby reaffirming earlier conclusions. © 2024, CC BY.

关键词： Anonymity

来源：评论

学校读者我要写书评

暂无评论

Project Elements: A computational entity-component-system in a scene-graph pythonic framework, for a neural, geometric computer graphics curriculum

Project Elements: A computational entity-component-system in...

引用

44th Annual Conference on European Association for computer Graphics, EUROGRAPHICS 2023

作者： Papagiannakis, G. Kamarianakis, M. Protopsaltis, A. Angelis, D. Zikas, P. Foundation of Technology Hellas - ICS Heraklion Greece University of Crete Dept. of Computer Science Heraklion Greece University of Western Macedonia Dept. of Electrical and Computer Engineering Kozani Greece University of Geneva Dept. of Computer Science Geneva Switzerland ORamaVR SA Geneva Switzerland ORamaVR SA Heraklion Greece

ISBN: (纸本)9783038682103

We present the Elements project, a lightweight, open-source, computational science and computer graphics (CG) framework, tailored for educational needs, that offers, for the first time, the advantages of an Entity-Component-System (ECS) along with the rapid prototyping convenience of a Scenegraph-based pythonic framework. This novelty allows advances in the teaching of CG: from heterogeneous directed acyclic graphs and dept.-first traversals, to animation, skinning, geometric algebra and shader-based components rendered via unique systems all the way to their representation as graph neural networks for 3D scientific visualization. Taking advantage of the unique ECS in a a Scenegraph underlying system, this project aims to bridge CG curricula and modern game engines (MGEs), that are based on the same approach but often present these notions in a black-box approach. It is designed to actively utilize software design patterns, under an extensible open-source approach. Although Elements provides a modern (i.e., shader-based as opposed to fixed-function OpenGL), simple to program approach with Jupyter notebooks and unit-tests, its CG pipeline is not black-box, exposing for teaching for the first time unique challenging scientific, visual and neural computing concepts. © 2023 The Authors.

关键词： Open source software

来源：评论

学校读者我要写书评

暂无评论

Children and Information Access: Fostering a Sense of Belonging

Children and Information Access: Fostering a Sense of Belong...

引用

Joint Workshops of IUI 2023: HAI-GEN - Human-AI Co-Creation with Generative Models, ITAH - Workshop on Interactive Technologies for AI in Healthcare, MILC - Workshop on Intelligent Music Interfaces for Listening and Creation, SHAI - Workshop on Designing for Safety in Human-AI Interactions, SketchRec - Workshop on Sketch Recognition and SOCIALIZE - Social and Cultural Integration with Personalized Interfaces

作者： Pera, Maria Soledad Wright, Katherine Landau Kennington, Casey Fails, Jerry Alan Web Information Systems TU Delft Netherlands Dept. of Literacy Language and Culture Boise State University United States Dept. of Computer Science Boise State University United States

In this vision paper, we spotlight children as often underserved users in the digital ecosystem. With online search as a use case, we discuss the need for a multi-perspective approach to designing interactive interfaces and technologies that can enable information access systems to better respond to children's requirements while respecting the cultural and social norms impacting their upbringing. © 2023 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). CEUR Workshop Proceedings (***)

关键词： Information retrieval

来源：评论

学校读者我要写书评

暂无评论

Detecting and Mitigating Adversarial Perturbations to Improve E-Commerce Security 11

Detecting and Mitigating Adversarial Perturbations to Improv...

引用

11th IEEE/ACM International Conference on Big Data Computing, Applications and Technologies, BDCAT 2024

作者： Tariq, Usman Jaafar, Fehmi Malik, Yasir Prince Sattam Bin Abdulaziz University College of Business Administration Dept. of Management Information Systems Saudi Arabia Université du Québec à Chicoutimi Dept. d'Informatique et de mathématique Canada Bishop's University Dept. of Computer Science Canada

ISBN: (纸本)9798350367300

E-commerce platforms face the critical challenge of adversary events, including fraudulent transactions and fake reviews, which can lead to significant financial and reputational damage. Addressing this, our research introduces a hybrid Deep Learning model, tailored for the detection of such adversarial activities. This innovative approach leverages spatial and sequential data processing capabilities, enhancing the identification of subtle adversarial manipulations across diverse e-commerce contexts. Our findings indicate a high detection rate of 93 percent for adversarial attacks, with precision, recall, and Matthews Correlation Coefficient metrics underscoring the model's efficacy. This work significantly contributes to e-commerce security by advancing the robustness of detection systems against a spectrum of adversarial threats, including account takeovers and deceptive reviews. While demonstrating a notable improvement over existing methods, our research also acknowledges the potential for evasion by sophisticated attacks, highlighting areas for future work in enhancing model resilience. This balance of innovation and critical insight provides a solid foundation for further advancements in the field of e-commerce security © 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

Heuristic approaches for hiding sensitive frequent itemsets in uncertain databases 5

Heuristic approaches for hiding sensitive frequent itemsets ...

引用

5th International Conference on Pattern Analysis and Intelligent systems, PAIS 2023

作者： Lefkir, Mira Nouioua, Farid Fournier-Viger, Phillipe University Mohamed El Bachir El Ibrahimi of Bordj Bou Arreridj BBA Dept. of Computer Science Algeria Shenzhen University College of Computer Science and Software Engineering Shenzhen China

ISBN: (纸本)9798350381450

Pattern mining is a core objective within data mining, involving the detection of frequent itemsets (collections of values) within databases. This process serves to extract valuable insights from the data, facilitating informed decision-making. In recent times, a range of algorithms has emerged to adapt this task to the realm of uncertain *** has a wide range of applications, such as analyzing medical data to identify correlated symptoms based on their existential probabilities. However, pattern mining algorithms also pose the risk of revealing sensitive information. As far as we know, there is no existing solution that addresses the inadvertent disclosure of sensitive itemsets when pattern mining algorithms are employed on uncertain databases. This paper addresses this problem by proposing three heuristic approaches, which ensure that sensitive expected frequent itemsets are not discovered while minimizing undesirable side effects for other itemsets. The first heuristic approach (called aggregate) deletes some transactions from an uncertain database. The second approach (named disaggregate) removes some selected items from each transaction. And the third approach (called the hybrid approach) combines the two previous approaches, and deletes some selected items from some selected transactions. An experimental evaluation is presented to assess and compare the effectiveness of the three approaches. © 2023 IEEE.

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

Recognition of Data Breach Method from Story Using Advanced Deep Learning 3

Recognition of Data Breach Method from Story Using Advanced ...

引用

3rd International Conference for Innovation in Technology, INOCON 2024

作者： Jeba, Samiha Maisha Aurpa, Tanjim Taharat Alom, Md Samiul Joy, Md. Wadud Mahmud Ahamed, Asm Shakil Dash, Satya Ranjan Ahmed, Md Shoaib International University of Business Agriculture and Technology Dept. of Computer Science and Engineering Bangladesh Bangabandhu Sheikh Mujibur Rahman Digital University Department of Data Science Bangladesh Kiit University School of Computer Applications Odisha India Boise State University Department of Computer Science BoiseID United States

ISBN: (纸本)9798350381931

We often open our eyes in the morning with the appalling news of data breaches of different popular companies. This is a significant threat not only to giant companies but also to the general people's privacy. Various methods are maneuvered for data breaches, and identifying these methods in a short can help understand the loss and take proper steps against these incidents. Researchers nowadays focus on the data breaches issue to reduce its threats to the tech industry. In this research, we have used advanced deep learning methods(e.g., BERT, XLNet, and Electra) to predict data breach methods using the data breach story. The proposed classifier has provided an accuracy of 92.86%. © 2024 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Demonstration of Effectiveness of Nativeness in Stepwise Learning by Performing Sentiment Analysis

Demonstration of Effectiveness of Nativeness in Stepwise Lea...

引用

Next-Generation Computing, IoT and Machine Learning (NCIM), International Conference on

作者： Al-Mahmud Kazutaka Shimada Dept. of Computer Science and Systems Engineering Kyushu Institute of Technology Japan

Nowadays, more people and entities express opinions on online platforms. With this increasing trend of online opinion mass information, it is impossible to analyze them manually. For this reason, an automatic approach to opinion and sentiment analysis is essential and useful in the real world. Several approaches based on machine learning techniques in sentiment analysis have been proposed. However, machine learning essentially needs massive data to generate a robust model. On the other hand, collecting the training data is costly, especially in low-resource language. To solve this problem, we introduce stepwise learning. Stepwise learning is a technique that improves the model capacity for a target task (i.e., main task) by exploiting auxiliary information from source tasks. It is effective when the data in the target tasks are inadequate. The target task of this paper is sentiment analysis for reviews in the Bangla language. Our proposed method consists of two stages: source and target tasks. For the source task, we train a transformers-based model, such as BanglaBERT, by a large dataset translated into the Bangla language; the original dataset is written in English. For the target task, we retrain and evaluate the model with a relatively small target dataset. We further investigate the effect of opinion expression style of writers between source and target data in stepwise learning. We call it “nativeness.” The experimental result shows that nativeness has an important role in improving performance.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Hybrid FA for High Performance Arithmetic Application

A Hybrid FA for High Performance Arithmetic Application

引用

2023 IEEE International Conference on Device Intelligence, Computing and Communication Technologies, DICCT 2023

作者： Kandpal, Jyoti Gowari, R. Dubey, V.P. Hussain, Md. Shahbaz Joshi, Jyoti Chamoli, Vivek Graphic Era Hill University Dept. of Electronics and Communication Engineering Dehradun India Aligarh Muslim University Z.H College of Engineering and Technology Dept. of Electronics Engineering India Graphic Era Hill University Dept. of Computer Science and Engineering Dehradun India

ISBN: (纸本)9781665474924

This research introduces a novel hybrid CMOS design model for a 1-bit full adder that addresses the requirements for improved noise-robustness, enhanced drivability, and reduced power consumption in deep submicron technology. The adoption of a hybrid style based on CMOS was explored to fulfil the desired performance of implementing new, fully-featured adders. More freedom to concentrate on different applications allows designers to drastically cut design efforts. There exist various unique designs for the full adder (FA), but this research proposes a new FA design implemented using CMOS technology with 90nm. The design employs an XOR-XNOR circuit that produces exclusive OR/exclusive NOR full-swing outputs and is utilized to construct FA is presented first. The proposed work offers improvement in latency and power delay products by 6.27 to 38.24 and 13.27 percent, respectively, over the prior design. © 2023 IEEE.

关键词： Adders

来源：评论

学校读者我要写书评

暂无评论

Using Artificial Neural Networks to Study Dependencies in Real-Time Audio and Video Conferencing

Using Artificial Neural Networks to Study Dependencies in Re...

引用

Communications, Information, Electronic and Energy systems (CIEES), International Conference on

作者： Vladislav Hinkov Georgi Krastev Dept. of Computer Systems and Technologies University of Ruse Ruse Bulgaria

ISBN: (数字)9798350352863

ISBN: (纸本)9798350352870

Real-time research of modern audio and video conferencing in operational conditions is important for revealing bottlenecks in their functioning. On this basis, recommendations can be made to improve performance indicators, such as productivity, reliability, mode of service, etc. In the study, the architecture of the artificial neural network (ANN) of the type with forward feed and backpropagation of errors was chosen. These networks are made up of several layers of neurons: an input layer, several hidden layers, and an output layer. The study used the Levenberg-Marquardt learning algorithm, which combines gradient descent methods and the Gaussian and Newtonian algorithm. The relationship between "active audio streams - active users" with two neural networks was investigated: The first number of neurons in the hidden layer is 5; The second number of neurons in the hidden layer is 6. The activating function that is used for the neurons of the hidden layer is the tangent hyperbolic, and for the output neuron – the linear function. The neural network with six neurons in the hidden layer in the scope of the training sample, the prediction of ANN is very close to the cloud of points of the original data. It can also be seen that in proximity (about 10% of the training sample range) ANN also has some extrapolation ability, reproducing the behavior of the original data.

关键词： Training Productivity Neurons Artificial neural networks Real-time systems Reliability Biological neural networks Streams Videoconferences Convergence

来源：评论

学校读者我要写书评

暂无评论

Diagnosis of Parkinson’s Diseases with MDVP Features Using Machine Learning Approaches

Diagnosis of Parkinson’s Diseases with MDVP Features Using ...

引用

2024 IEEE International Conference on Biomedical engineering, computer and Information Technology for Health, BECITHCON 2024

作者： Sikder, Iftikhar Ahammad Uddin, Khandaker Mohammad Mohi Nur-A-Alam, Md. Based, Md Abdul Dept. of Computer Science and Engineering Dhaka International University Dhaka Bangladesh Dept. of Computer Science and Engineering Southeast University Dhaka Bangladesh Dept. of Computer Science and Engineering Sunamgonj Science and Technology University Dhaka Bangladesh

ISBN: (纸本)9798331534356

Parkinson's disease is a neurodegenerative disorder that poses a significant global health challenge. Its prevalence has prompted the urgent need for fast and immediate diagnosis to enable timely intervention and treatment planning. Parkinson's disease causes both motor and non-motor symptoms necessitating a comprehensive approach to prediction. This research explores the intricacies of Parkinson's disease prediction, highlighting the importance of multidimensional data analysis. Acknowledging the evolving landscape of medical research, the study advocates for a comprehensive approach to enhance the early detection of Parkinson's disease, provides significant insights on the possibilities of machine learning techniques for improved patient outcomes, particularly focusing on the distinctive features offered by MDVP (Multidimensional Voice Program). Including MDVP features enhances the predictive capabilities of the models, offering a novel perspective for accurate early detection and applying machine learning models like AdaBoost, Multilayer Perceptron, Decision Tree, Random Forest, Logistic Regression, K-nearest Neighbors, Support Vector Machine (SVM), Gradient Boosting, & Naive Bayes. This research explores the intricacies of Parkinson's disease prediction. Results highlight the remarkable performance of the Support Vector Machine, achieving an accuracy of 98.98% & Gradient Boosting, 97.28% accuracy with the feature selection technique XGBoost and SelectKBest underscoring the significance of MDVP features in refining the predictive accuracy. This study leads to continuous efforts for the improvement of diagnostic accuracy and prognosis for Parkinson's disease, emphasizing the importance of integrating MDVP features within the machine learning framework. © 2024 IEEE.

关键词： Support vector machines

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：