检索结果-内蒙古大学图书馆

An Extensive Overview of anamoly Detection Analysis in FL-Based IDS

Procedia computer science 2025年 259卷 326-335页

作者： Himani Tyagi Abhilasha Singh Research Scholar Department of Computer Science and Engineering SRM Institute of Science and Technology Delhi- NCR Campus Ghaziabad Associate Professor Department of Computer Science and Engineering SRM Institute of Science and Technology Delhi- NCR Campus Ghaziabad

In this paper, we present a Federated Learning-based Intrusion Detection System (FL-IDS) aimed at enhancing security in Connected Autonomous Vehicles (CAVs) within IoT environments. The novelty lies in the dual utilization of Support Vector Machine (SVM) and Convolutional Neural Networks (CNN) for anomaly detection, leveraging the Flower framework for privacy-preserving, decentralized model training. Unlike traditional centralized approaches, FL-IDS ensures data confidentiality by sharing only model updates. Using the NSL-KDD dataset, our evaluation highlights superior accuracy and loss metrics, validating the efficiency of both machine learning and deep learning methods for detecting DDoS, Probe, and U2R attacks. This work contributes to the advancement of secure and efficient vehicular communication systems, providing scalable solutions for real-world IoT deployments.

关键词： Federated Learning Deep Learning Machine learning CAV IDS Edge Computing

来源：评论

学校读者我要写书评

暂无评论

T5LSTM-RNN based Text Summarization Model for Behavioral Biology Literature

引用

Procedia computer science 2023年 218卷 585-593页

作者： Shivangi Chaurasia Debalay Dasgupta Rajeshkhannan Regunathan School of Computer Science and Engineering Vellore Institute of Technology Vellore Professor School of Computer Science and Engineering Vellore Institute of Technology Vellore

Behavioral biology is one of the crucial and trending topics these days which needs proper attention by scholars for rapid development in this field. Therefore, the purpose of this paper is to ease the process of collecting all kinds of relevant and vital data available on the internet regarding this topic in different forms of media, such as news articles, research papers, and YouTube lecture videos, all into one place into a single document in a proper summarized form. For proper training of the LSTM model, the lengthy video and journal datasets are pre-processed using the T5 transformer model to generate a uniform training dataset. So, in this work, a comprehensive approach is proposed based on an abstractive form of text summarization using the seq2seq encoder-decoder model combined with a stacked LSTM layer with an attention mechanism and a T5 transformer model pre-processor. Therefore, a proper hybrid model, T5LSTM-RNN, is implemented to generate the summarized data.

关键词： Seq2Seq Text Summarization Abstractive Summary T5 Transformer Behavior Biology LSTM RNN

来源：评论

学校读者我要写书评

暂无评论

Filtering Decision Rules Driven by Sequential Forward and Backward Selection of Attributes: An Illustrative Example in Stylometric Domain 18

Filtering Decision Rules Driven by Sequential Forward and Ba...

引用

18th Conference on computer science and Intelligence Systems, FedCSIS 2023

作者： Zielosko, Beata Stanczyk, Urszula Jablonski, Kamil Institute of Computer Science University of Silesia in Katowice Bedzínska 39 Sosnowiec41-200 Poland Department of Graphics Silesian University of Technology Computer Vision and Digital Systems Akademicka 2A Gliwice44-100 Poland

ISBN: (纸本)9788396744784

The paper presents investigations concerning the decision rule filtering process controlled by the estimated relevance of available attributes. In the conducted study, two search directions were used, sequential forward selection and sequential backward elimination. The steps of sequential search were governed by three rankings obtained for variables, all related to characteristics of data and rules that can be induced, as follows, (i) a ranking based on the weighting factor referring to the occurrence of attributes in generated decision reducts, (ii) the OneR ranking exploiting short rule properties, and (iii) the proposed ranking defined through the operation of greedy algorithm for rule induction. The three rankings were confronted and compared from the perspective of their usefulness for the selection of rules performed in the two directions and with two strategies for rule selection. The resulting sets of rules were analysed with respect to the properties of the constituent decision rules and from the point of performance for all constructed rule-based classifiers. Substantial experiments were carried out in the stylometric domain, treating the task of authorship attribution as classification. The results obtained indicate that for all three rankings and search paths it was possible to obtain a noticeable reduction of attributes while at least maintaining the power of inducers, at the same time improving characteristics of rule sets. © 2023 Polish Information Processing Society.

关键词： Information use

来源：评论

学校读者我要写书评

暂无评论

Secured Digital and QR Code Payments: A Comparative Analysis of Payment Models

Secured Digital and QR Code Payments: A Comparative Analysis...

引用

Sustainable Expert Systems (ICSES), International Conference on

作者： Bhavna Galhotra Shalini Bhaskar Bajaj Aman Jatain Vivek Jaglan Computer science dept Amity University JIMS Computer science dept Amity University Department of Computer Science & Technology Manav Rachna University Research Graphics Era University Dehradun

ISBN: (数字)9798331540364

ISBN: (纸本)9798331540371

The internet's explosive expansion has paved the way for a wide range of smartphone applications that use QR codes and digital wallets to facilitate online payments. Security is a major concern for web-based applications since they are being delivered over the Internet at a rapid pace. Due to unlawful access to many accounts, which causes problems with end-to-end security, the security of these applications has grown to be a top concern for both customers and business owners. This work has addressed the several models that are accessible for security applications, also provided a comparative analysis of all the models, taking into account their security privacy, adaptability, accuracy reliability, vulnerabilities and suitability for digital wallets and online payment systems through QR codes.

关键词： Adaptation models Analytical models Privacy Biological system modeling QR codes Logic gates Tokenization Mobile applications Reliability Standards

来源：评论

学校读者我要写书评

暂无评论

PFMRC: A Unified MRC Framework for Chinese Named Entity Recognition via Semantic Feature Fusion 4

PFMRC: A Unified MRC Framework for Chinese Named Entity Reco...

引用

4th International Conference on Artificial Intelligence, Robotics, and Communication, ICAIRC 2024

作者： Deng, Zhenrong Zhang, Jinglin Xiao, Yannan Wei, Shiwei Guilin University of Electronic Technology Guangxi Key Laboratory of Images and Graphics Intelligent Processing Guilin China Nanning Research Institute Guilin University of Electronic Technology Guilin China Guangxi Key Laboratory of Images and Graphics Intelligent Processing Guilin University of Electronic Technology Guilin China School of Computer Science and Engineering Guilin University of Aerospace Technology Guilin China

ISBN: (纸本)9798331531225

The landscape of natural language processing (NLP) has been revolutionized by advancements that enable the upstream training of knowledge-rich models for downstream tasks, particularly named entity recognition (NER). Traditional NER approaches, treating it as a sequence labeling problem, often fail to deeply understand the whole text and to capture syntactic nuances of text. To address these issues, we introduce PFMRC, a novel machine reading comprehension (MRC) framework that leverages the strengths of both contextual semantics and syntactic information for enhanced NER performance. PFMRC employs a dual-pronged feature extraction strategy, utilizing RoBERTa and BiGRU to encode text features for contextual semantic information and representing them as node features of the graph structure, and a weighted embedding representation to translate syntactic dependencies into graph-structured edge features. Our model further incorporates Graph Attention Networks (GAT) to harmonize node and edge features, facilitating a nuanced understanding of text. The integration of prior knowledge and the modeling of intricate syntactic relationships through Graph Attention Networks (GAT) not only improves entity recognition accuracy but also enriches the feature representation, leading to more robust NER outcomes. Extensive experiments on three Chinese datasets demonstrate PFMRC's superior performance over state-of-the-art models. © 2024 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Research on the behaviour of Bluetooth Low Energy protocol in the heart rate monitoring application

引用

Procedia computer science 2023年 225卷 63-69页

作者： Zuzanna Balas Krzysztof Tokarz Bartłomiej Zieliński Tomasz Guźniczak Department of Graphics Computer Vision and Digital Systems Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Gliwice Poland

In the paper, the authors check the behaviour of Bluetooth Low Energy protocol in a popular smart wristband and a microcontroller in the heart rate monitoring application. The measurements were collected using a development board with the ESP32 System on Chip. The authors tested measurement period stability and measurement reliability in various conditions.

关键词： Smart wristband Internet of Things Health monitoring Bluetooth Low Energy Wireless data transmission

来源：评论

学校读者我要写书评

暂无评论

Domain-specific data characteristics: A study on meaning of stylometric sub-concepts and in-class imbalance

引用

Procedia computer science 2024年 246卷 4421-4430页

作者： Urszula Stańczyk Department of Computer Graphics Vision and Digital Systems Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Akademicka 2A 44-100 Gliwice Poland

In the context of data imbalance probably the most investigated problem is imbalance of classes, as learning from the data with this characteristic makes detection of existing patterns for all classes more difficult. However, other problems related to imbalance also exists and the paper addresses such cases where classes are balanced, but there is in-class imbalance. Such imbalance can be caused by uneven representation of sub-concepts. When there is a noticeable difference between the numbers of samples belonging to sub-concepts, this can turn the under-represented sub-concepts into disjuncts. Data irregularities of this type can hinder recognition, therefore actions are typically taken to restore balance. In the investigations described, the issue was studied in the stylometric domain and various classifiers were applied to the data that was balanced, then imbalanced, and finally with restored balance. The experiments show that the specifics of the domain of application can put its own mark on the data which is difficult to overcome by standard processing such as under- or oversampling. Observed dependence on a learner and dataset makes the issue even more complex and layered, and shows the need for deeper studies.

关键词： Attribute domain Data irregularities Sub-concept Disjunct Pattern recognition Stylometry Authorship attribution

来源：评论

学校读者我要写书评

暂无评论

Real-Time Sign Language Recognition System for Physically Challenged Community Using Deep Learning Technique 8th

Real-Time Sign Language Recognition System for Physically Ch...

引用

8th International Conference on Microelectronics Electromagnetics and Telecommunications, ICMEET 2023

作者： Devi, V. Anjana Bhuvaneswari, E. Tummala, Rama Krishna Professor Department of Computer Science and Engineering Rajalakshmi Institute of Technology Tamil Nadu Chennai India Associate Professor Department of Artificial Intelligence and Data Science Panimalar Engineering College Tamil Nadu Chennai India Accenture Solutions Pvt. Ltd Chennai India

ISBN: (纸本)9789819784219

Language disparities have always existed in human civilization. In their community, deaf individuals use sign languages to communicate with one another, but it can be difficult to do so when interacting with the wider population. The project comprises a real-time sign language recognition system that enables hearing-impaired individuals and persons who do not know sign language to effortlessly communicate. The most used sign language in South Asian nations is Indian sign language. Convolution neural networks, specifically Tensor Flow object detection, are utilized to assess and train the model on the images that are used. The transfer training approach is used to train the deep learning model with various hand movements. Loss is anticipated to be decreased. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Observations of data characteristics and irregularities through domain-oriented transformations of attributes

引用

Procedia computer science 2024年 246卷 2694-2703页

作者： Urszula Stańczyk Grzegorz Baron Department of Computer Graphics Vision and Digital Systems Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Akademicka 2A 44-100 Gliwice Poland

The paper presents research dedicated to observations of relations between attribute properties and discretisation. In the investigations described, the gradually increasing sets of features were discretised by selected approaches, and several variants of data were constructed. The continuous, partially discrete, and completely translated datasets were explored by the chosen classifiers and their performance studied in the context of a number of discretised attributes, discretisation procedures, and the way of processing of features and datasets. The stylometric problem of authorship attribution was the machine learning task under study. The experimental results enable to observe closer the specificity of style-markers employed as characteristic features, and indicate conditions for efficient recognition of authorship. They can be extended to other application domains with similar characteristics.

关键词： Attribute domain Discretisation Data representation Data irregularities Pattern recognition Stylometry Authorship attribution

来源：评论

学校读者我要写书评

暂无评论

Attribute Relevance and Discretisation in Knowledge Discovery: A Study in Stylometric Domain 23rd

Attribute Relevance and Discretisation in Knowledge Discov...

引用

23rd International Conference on Computational science, ICCS 2023

作者： Stańczyk, Urszula Zielosko, Beata Baron, Grzegorz Department of Graphics Computer Vision and Digital Systems Silesian University of Technology Akademicka 2A Gliwice44-100 Poland Institute of Computer Science University of Silesia in Katowice Bȩdzińska 39 Sosnowiec41-200 Poland

ISBN: (纸本)9783031360206

The paper demonstrates the research methodology focused on observations of relations between attribute relevance, displayed by rankings, and discretisation. Instead of transforming all continuous attributes before data exploration, the variables were gradually processed, and the impact of such a change on the performance of a classifier was studied. Considerable experiments carried out on stylometric data illustrate that selective discretisation could be more advantageous to predictive accuracy than some uniform transformation of all features. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Metadata

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：