检索结果-内蒙古大学图书馆

Combining Ensemble Learning based Hybrid Classification Algorithm with Effective Feature Selection to identify Electricity Fraud in Advanced Metering Infrastructure of Smart Grid

Combining Ensemble Learning based Hybrid Classification Algo...

引用

2024 IEEE International Conference on Computing, Applications and Systems, COMPAS 2024

作者： Sikder, Md. Masum Ahmed, Boshir Rajshahi University of Engineering and Technology Department of Computer Science and Engineering Rajshahi Bangladesh

ISBN: (纸本)9798331529765

Consuming electricity through Advanced Metering Infrastructure (AMI) has become an irrefutable part of our daily micro-moments. In AMI, electricity fraud is regarded as one of the most significant Nontechnical losses (NTLs) for power companies which requires substantial investigation to effectively detect theft instances with minimal false positives. In this paper, we investigated how the meticulous selection of top-ranked attributes enhances the accuracy of fraud identification. We proposed a modified-ReliefF feature selection algorithm to find the rank of attributes of our developed dataset and observed significant improvement in electricity fraud identification. Moreover, we designed a hybrid classifier called ElectroGuard SVM Bagging, which combines SVM with Bootstrap Aggregation. This approach, paired with our modified-ReliefF feature selection, outperformed three standard supervised classifiers. Finally, using 10-fold cross-validation, we compared our hybrid model's performance against the best result obtaining conventional classifier, i.e. SVM and achieved 5.94% (before applying feature selection) and 6.19% (after applying modified-ReliefF feature selection) accuracy improvement for ElectroGuard SVM Bagging. © 2024 IEEE.

关键词： Advanced metering infrastructures

来源：评论

学校读者我要写书评

暂无评论

Unleashing the power of pseudo-code for binary code similarity analysis

引用

Cybersecurity 2023年第2期6卷 44-61页

作者： Weiwei Zhang Zhengzi Xu Yang Xiao Yinxing Xue School of Computer Science and Engineering University of Science and Technology of ChinaHefeiChina School of Computer Science and Engineering Nanyang Technological UniversitySingaporeSingapore Institute of Information Engineering Chinese Academy of SciencesBeijingChina School of Cyber Security University of Chinese Academy of SciencesBeijingChina

Code similarity analysis has become more popular due to its significant applicantions,including vulnerability detection,malware detection,and patch *** the source code of the software is difficult to obtain under most circumstances,binary-level code similarity analysis(BCSA)has been paid much attention *** recent years,many BCSA studies incorporating Al techniques focus on deriving semantic information from binary functions with code representations such as assembly code,intermediate representations,and control flow graphs to measure the ***,due to the impacts of different compilers,architectures,and obfuscations,binaries compiled from the same source code may vary considerably,which becomes the major obstacle for these works to obtain robust *** this paper,we propose a solution,named UPPC(Unleashing the Power of Pseudo-code),which leverages the pseudo-code of binary function as input,to address the binary code similarity analysis challenge,since pseudocode has higher abstraction and is platform-independent compared to binary *** selectively inlines the functions to capture the full function semantics across different compiler optimization levels and uses a deep pyramidal convolutional neural network to obtain the semantic embedding of the *** evaluated UPPC on a data set containing vulnerabilities and a data set including different architectures(X86,ARM),different optimization options(O0-O3),different compilers(GCC,Clang),and four obfuscation *** experimental results show that the accuracy of UPPC in function search is 33.2%higher than that of existing methods.

关键词： Binary code similarity Machine learning Software security Pseudo-code

来源：评论

学校读者我要写书评

暂无评论

Marine Life Analysis Based on Ocean Water Level Rise and Climate Change Using Underwater Imaging Techniques

引用

Remote Sensing in Earth Systems sciences 2024年第4期7卷 657-669页

作者： Mate, Anuja Dinesh Arulananth, T.S. Sathiya, T. Jyothi, B. Swarna Raja, A. Wasim Yalawar, Mrutyunjaya S. Computer Technology Sou.Venutai Chavan Polytechnic Pune India Department of Electronics and Communication Engineering MLR Institute of Technology Telangana Hyderabad 500043 India Department of Computer Science and Engineering Sona College of Technology Tamil Nadu Salem India Computer Science & amp Engineering Madanapalle Institute of Technology & amp Science Andhra Pradesh Madanapalle India Artificial Intelligence and Data Science Sri Krishna College of Engineering and Technology Coimbatore India Department of Computer Science and Engineering CMR Engineering College Hyderabad India

Forecasting sea levels is crucial for harbour operations and coastal structure design. The oceans make up two-thirds of Earth’s surface;therefore, historically, the marine economy has been extremely diversified as well as abundant. In order to explore as well as harness this difficult environment, prominent businesses and ecological groups are developing and deploying novel gadgets backed by state-of-the-art marine mechatronics solutions. This research proposes a novel method of marine life monitoring with ocean water level rise analysis based on climate changes by underwater imaging using machine learning techniques. Here, the input has been collected as underwater images as well as processed for noise removal and normalisation. Then, these images were segmented for monitoring marine life using U-Net graph cut fuzzy clustering with a spatial neural network. Then this segmented image has been classified using multilayer region-based Q-belief vector neural networks for the analysis of sea level rise analysis based on climate change training. Experimental analysis has been carried out for various marine life monitoring datasets and climate change datasets in terms of training accuracy, specificity, recall, AUC, and Kappa-coefficient. Proposed techniques achieved specificity 93%, training accuracy 91%, AUC 95%, recall 96%, and Kappa-coefficient 90%. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2024.

关键词： Climate changes Machine learning techniques Marine life monitoring Q-belief vector Water level rise analysis

来源：评论

学校读者我要写书评

暂无评论

Influence of Access Points’ Height and High Signal Relation in WLAN Fingerprinting-Based Indoor Positioning Systems’ Accuracy 1

引用

26th International Conference on Distributed computer and Communication Networks: Control, Computation, Communications, DCCN 2023

作者： Nicholaus, Mrindoko R. Ruambo, Francis A. Masanga, Elijah E. Muthanna, Mohammed Saleh Ali Lashchev, Andrei Computer Science and Engineering Department College of Information and Communication Technology Mbeya University of Science and Technology 131 Mbeya Tanzania United Republic of Information Systems and Technology Department College of Information and Communication Technology Mbeya University of Science and Technology 131 Mbeya Tanzania United Republic of School of Cyber Science and Engineering Huazhong University of Science and Technology Wuhan430074 China Research and Development Department Aifrruis Laboratories Mbeya Tanzania United Republic of Institute of Computer Technologies and Information Security Southern Federal University Taganrog347922 Russia

ISBN: (数字)9783031504822

ISBN: (纸本)9783031504815

Wireless Local Area Network (WLAN) Fingerprinting-based Indoor Positioning Systems (IPS) offer several key advantages over other indoor positioning technologies, including cost-effectiveness with high accuracy and passive tracking of entities’ location without Global Positioning System (GPS) dependence. However, achieving a satisfactory accuracy performance of the WLAN fingerprinting-based IPS is still challenging, as it is affected by several factors, such as obstacles (e.g., walls and furniture) that can block or weaken wireless signals. Additionally, the IPS’s accuracy can be affected by environmental changes, such as the height, addition, or removal of a WLAN Access Point (AP) or changes in radio frequency interference. This paper presents a novel algorithm incorporating a probabilistic analytical model employing only high signal relations that mitigate the effect of low signal relation on the WLAN fingerprinting-based IPS, thus improving the accuracy performance. Furthermore, it analyses the impact of AP heights on the accuracy performance of the WLAN IPS. Upon extensive experiments, the proposed algorithm improves the accuracy of the IPS by an average of 32.74% in terms of Root Mean Square Error (RMSE). These results imply that incorporating APs’ height and high signal relations can significantly improve the accuracy performance of the WLAN fingerprinting-based IPS. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Cost effectiveness

来源：评论

学校读者我要写书评

暂无评论

Flexible and Effective Cellular Traffic Data Synthesis with Large Language Model

Flexible and Effective Cellular Traffic Data Synthesis with ...

引用

2024 IEEE Global Communications Conference, GLOBECOM 2024

作者： Duan, Sijing Lyu, Feng Cen, Jinfeng Ren, Ju Yang, Peng Zhang, Yaoxue Tsinghua University Department of Computer Science and Technology Beijing China Central South University School of Computer Science and Engineering Changsha China Huazhong University of Science and Technology School of Electronic Information and Communications Wuhan China

ISBN: (纸本)9798350351255

Cellular traffic data hold significant potential for applications such as network planning, traffic prediction, mobility modeling, and personalized recommendations. However, limited data accessibility hinders more open data-driven research. Previous studies have explored data synthesis, while exhibiting flexible limitations in supporting conditional traffic synthesis, and are vulnerable to multidimensional data modeling. In this paper, we present LLMCell, a flexible and effective framework that leverages the arbitrary conditioning and contextual understanding capabilities of the large language model (LLM) to generate high-quality synthetic cellular traffic data. The LLMCell comprises three key components: i) a textual encoder for converting raw cellular traffic data into textual representations, ii) a generative model learner to fine-tune pre-trained LLM based on encoded textual representation for cellular traffic generation, and iii) a synthetic data sampling module for final synthetic data sampling and textual-to-data transformation. Experiments conducted on a large-scale dataset demonstrate the superior fidelity and utility of LLMCell over state-of-the-art baselines, and the synthetic data can effectively preserve user privacy. We release our synthetic dataset to the public to benefit future research in the wireless network community1 © 2024 IEEE.

关键词： Cellular Traffic Flexible Data Synthesis Large Language Model

来源：评论

学校读者我要写书评

暂无评论

Economical Calculation Basis of Synthetic Gas Production using Alternate Renewable Resources in Combining of Hydrogen and Carbon Dioxide 65

Economical Calculation Basis of Synthetic Gas Production usi...

引用

65th IEEE Annual International Scientific Conference on Power and Electrical engineering of Riga Technical University, RTUCON 2024

作者： Borisovs, Aleksandrs Dolgicers, Aleksandrs Riga Technical University Faculty of Computer Science Information Technology and Power Engineering Riga Latvia

ISBN: (纸本)9798350365771

The main task of this work is to establish the consumption of energy from renewable sources such as solar and wind energy for the production of such a combustible substance as hydrogen, and further optimal calculation of the thermodynamic process of heating carbon dioxide and hydrogen to produce economic synthetic gas. © 2024 IEEE.

关键词： energy gas hydrogen renewable source synthetic

来源：评论

学校读者我要写书评

暂无评论

Intelligent Intrusion Detection Framework for Anomaly-Based CAN Bus Network Using Bidirectional Long Short-Term Memory

引用

Journal of The Institution of Engineers (India): Series B 2024年第3期105卷 541-564页

作者： Kishore, Ch. Ravi Rao, D. Chandrasekhar Nayak, Janmenjoy Behera, H.S. Department of Information Technology Veer Surendra Sai University of Technology Burla768018 India Department of Computer Science Maharaja Sriram Chandra Bhanja Deo University Odisha Baripada757003 India

The modern vehicle depends on controller area network (CAN) bus technology to facilitate communication among electronic control units. Although the CAN bus technology is significant, it lacks the typical authentication and authorization protocols, allowing interactions susceptible to unauthorized access. Attackers can easily target the CAN bus network because of this vulnerability. Attackers have several options when it comes to infiltrating a CAN bus system, and several of them are focused on either denial of service, fuzzing, or spoofing attacks. To ensure current autos are secure, it is critical to developing ways to guard against these threats. In order to identify and eliminate vulnerabilities affecting CAN bus networks, prior research has developed several works, including intrusion detection systems (IDSs) based on (ML) and deep learning (DL) approaches such as the long short-term memory (LSTM)-based IDS. Many previous studies employed ML or DL methodologies to identify and mitigate CAN bus network assaults. Technological advancements increase attack variety. An IDS has been designed for the CAN bus that is more precise and accurate. CAN bus network assaults are being detected and mitigated in this study using a Bidirectional_LSTM-based IDS. Using the suggested framework, non-attack and attack records can be discriminated and precisely characterized. With improved accuracy and an intelligent IDS creating process, the suggested model outperforms other standard ML approaches. The proposed method outperforms other competitive methods with an accuracy of 99.09% and signifies the efficacy of the developed method. © The Institution of Engineers (India) 2024.

关键词： Authentication

来源：评论

学校读者我要写书评

暂无评论

SoulMate: Short-Text Author Linking Through Multi-Aspect Temporal-Textual Embedding (Extended Abstract) 40

SoulMate: Short-Text Author Linking Through Multi-Aspect Tem...

引用

40th IEEE International Conference on Data engineering, ICDE 2024

作者： Najafipour, Saeed Hosseini, Saeid Hua, Wen Kangavari, Mohammad Reza Zhou, Xiaofang School of Information Technology and Electrical Engineering Iran University of Science and Technology Tehran Iran Sohar University Faculty of Computing and Information Technology Sohar Oman Hong Kong Polytechnic University Department of Computing Hong Kong Hong Kong University of Science and Technology Department of Computer Science and Engineering Hong Kong

ISBN: (纸本)9798350317152

We devise a neural network-based temporal-textual framework that generates subgraphs with highly correlated authors from short-text contents. Our approach computes the relevance score (edge weight) between authors by considering a portmanteau of contents and concepts. It then employs a stack-wise graph-cutting algorithm to extract communities of related authors. Experimental results show that our multi-aspect vector space model can gain higher performance than other knowledge-centered competitors in linking short-text authors. © 2024 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Cluster-preserving sampling algorithm for large-scale graphs

引用

science China(information sciences) 2023年第1期66卷 60-76页

作者： Jianpeng ZHANG Hongchang CHEN Dingjiu YU Yulong PEI Yingjun DENG National Digital Switching System E&T Research Center Information Engineering University Network Systems Department of the Strategic Support Force School of Computer Science and Technology Eindhoven University of Technology Center for Applied Mathematics Tianjin University

Graph sampling is a very effective method to deal with scalability issues when analyzing largescale graphs. Lots of sampling algorithms have been proposed, and sampling qualities have been quantified using explicit properties(e.g., degree distribution) of the sample. However, the existing sampling techniques are inadequate for the current sampling task: sampling the clustering structure, which is a crucial property of the current networks. In this paper, using different expansion strategies, two novel top-leader sampling methods(i.e., TLS-e and TLS-i) are proposed to obtain representative samples, and they are capable of effectively preserving the clustering structure. The rationale behind them is to select top-leader nodes of most clusters into the sample and then heuristically incorporate peripheral nodes into the sample using specific expansion strategies. Extensive experiments are conducted to investigate how well sampling techniques preserve the clustering structure of graphs. Our empirical results show that the proposed sampling algorithms can preserve the population's clustering structure well and provide feasible solutions to sample the clustering structure from large-scale graphs.

关键词： graph sampling clustering structure top-leader nodes expansion strategies large-scale graphs

来源：评论

学校读者我要写书评

暂无评论

Detection of Compromised Accounts using Machine Learning Based Boosting Algorithms- AdaBoost, XGBoost, and CatBoost 14

Detection of Compromised Accounts using Machine Learning Bas...

引用

14th International Conference on Computing Communication and Networking Technologies, ICCCNT 2023

作者： Pandey, Arti Jain, Arti Jaypee Institute of Information Technology Computer Science & Engineering and Information Technology Uttar Pradesh Noida India

ISBN: (纸本)9798350335095

In today's digital era, Online Social Media (OSM) like Twitter, Facebook, and Instagram are used to share information. Everyone is trusted on these social network platforms for exchanging information such as personal, professional, financial, etc. Due to this implicit faith in these social platforms, users broadcast their personal information on them. By taking advantage of the same, unauthorized users use the account of legitimate users for performing malicious activities. They gain valuable information and legitimate accounts are trapped in malicious hands. These types of accounts are known as Compromised Accounts. In order to create a safe and secure social media environment, it is, therefore, essential to identify these compromised accounts. In this paper, for the detection of Compromised Accounts, we have applied machine learning-based three boosting algorithms- AdaBoost, XGBoost, and CatBoost. These algorithms are compared with the state-of-art Compromised Account Detection using the Authorship Verification (CAD-AV) approach over the real-world Twitter dataset. The results have shown that all three algorithms- AdaBoost, XGBoost, and CatBoost have performed better in comparison to CAD-AV. Among them, CatBoost has performed the best (Accuracy: 94%) in the detection of compromised accounts. © 2023 IEEE.

关键词： Adaptive boosting

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：