检索结果-内蒙古大学图书馆

A Detection Framework Against CPMA Attack Based on Trust Evaluation and Machine Learning in IoT Network

IEEE INTERNET OF THINGS JOURNAL 2021年第20期8卷 15249-15258页

作者： Liu, Liang Xu, Xiangyu Liu, Yulei Ma, Zuchao Peng, Jianfei Nanjing Univ Aeronaut & Astronaut Coll Comp Sci & Technol Nanjing 210016 Peoples R China

Internet of Things (IoT) network is vulnerable to various cyberattacks, especially insider attacks. Most existing studies mainly detect nontargeted insider attackers, who manipulate all packets forwarded by them with a probability. Compared with nontargeted attackers, targeted attackers only manipulate specific packets, which makes them more efficient and covert. In this article, we propose a targeted insider attack model called conditional packets manipulation attack (CPMA), in which attackers maliciously manipulate the packets whose attribute values meet specific conditions with a probability. When resisting the CPMA attack, most existing detection algorithms are inefficient to find such malicious behavior. Also, they detect malicious nodes by collecting and analyzing the overall behavior of nodes, which are not appropriate for energy-constrained nodes in the IoT network. To solve these problems, we present CPMAED, a malicious nodes detection framework against CPMA attack. CPMAED maintains some partial trust metrics for each relay node, which indicate the probability of launch attacks when forwarding the packets with different attribute values. Also, our scheme leverages regression and clustering algorithms to evaluate the trust values of nodes and classify them into benign or malicious. In order to obtain higher detection accuracy, we optimize the routing of transmitted packets and inject the packets to collect more information about nodes to enhance detection. The experimental results show that our proposed scheme utilizing support vector machine and K-means can achieve good detection performance and identify malicious nodes' attack modes with high accuracy.

关键词： Internet of Things Routing Relays Machine learning Sensors Measurement clustering algorithms Insider attacks Internet of Things (IoT) network machine learning trust evaluation

来源：评论

学校读者我要写书评

暂无评论

Algorithmic performance-data classification of condensing heat exchangers

引用

NUMERICAL HEAT TRANSFER PART A-APPLICATIONS 2021年第1期79卷 1-20页

作者： Pacheco-Vega, Arturo Avila, Gabriela Calif State Univ Los Angeles Dept Mech Engn Los Angeles CA 90032 USA Univ Autonoma San Luis Potosi CIEP Fac Ciencias Quim San Luis Potosi San Luis Potosi Mexico

We describe a methodology able to extract the regimes of operation from condensing heat exchanger data. The methodology applies a Gaussian mixtures clustering algorithm to determine the number of groups directly from the data, and a maximum likelihood decision rule to classify such data into these clusters. Published measurements visually classified as dry-surface, dropwise condensation, and film condensation, are used to illustrate the applicability of the classification technique. Since some results from the algorithmic and visual classifications differ, the second part presents an independent evaluation of the allocation process by a procedure based on artificial neural networks (ANNs) and a variant of cross-validation. Results from the ANN approach to the same data show remarkable agreement with the clustering technique confirming that an algorithmic classification of complex physical phenomena is not only possible but also accurate, and represents an excellent alternative to typical visual-based procedures for the analysis of thermal and other systems.

关键词： clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

Protein Complexes Detection Based on Semi-Supervised Network Embedding Model

引用

IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS 2021年第2期18卷 797-803页

作者： Zhu, Jia Zheng, Zetao Yang, Min Fung, Gabriel Pui Cheong Huang, Changqin South China Normal Univ Sch Comp Sci Guangzhou 510631 Peoples R China Chinese Acad Sci Shenzhen Inst Adv Technol Beijing 100864 Peoples R China Chinese Univ Hong Kong Dept SEEM Hong Kong Peoples R China Zhejiang Normal Univ Dept Educ Technol Jinhua 321004 Zhejiang Peoples R China

A protein complex is a group of associated polypeptide chains which plays essential roles in the biological process. Given a graph representing protein-protein interactions (PPI) network, it is critical but non-trivial to detect protein complexes, the subsets of proteins that are tightly coupled, from it. Network embedding is a technique to learn low-dimensional representations of vertices in networks. It has been proved quite useful for community detection in social networks in recent years. However, unlike social networks, PPI network does not contain rich metadata, so that existing network embedding methods cannot fully capture the network structure of PPI to improve the effect of protein complexes detection significantly. We propose a semi-supervised network embedding model by adopting graph convolutional networks to detect densely connected subgraphs effectively. We compare the performance of our model with state-of-the-art approaches on three popular PPI networks with various data sizes and densities. The experimental results show that our approach significantly outperforms other approaches on all three PPI networks.

关键词： Proteins Protein engineering clustering algorithms Biological system modeling Computational modeling Social networking (online) Data models Protein complexes detection network embedding deep learning

来源：评论

学校读者我要写书评

暂无评论

BS-SC: An Unsupervised Approach for Detecting Shilling Profiles in Collaborative Recommender Systems

引用

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2021年第4期33卷 1375-1388页

作者： Cai, Hongyun Zhang, Fuzhi Yanshan Univ Sch Informat Sci & Engn Qinhuangdao 066000 Hebei Peoples R China Hebei Univ Sch Cyber Secur & Comp Baoding 071000 Peoples R China Key Lab Comp Virtual Technol & Syst Integrat Hebe Qinhuangdao 066000 Hebei Peoples R China

Collaborative recommender systems are vulnerable to shilling attacks. To address this issue, many methods including supervised and unsupervised have been proposed. However, supervised detection methods require training classifiers and they only apply to detect known types of attacks. The existing unsupervised detection methods need to know the prior knowledge of attacks, otherwise they suffer from low detection precision. In this paper, we present BS-SC, an unsupervised approach for detecting shilling profiles, which does not need to know the attack size or to label the candidate spammers. BS-SC starts from an in-depth analysis of user behaviors and uses two key mechanisms (i.e., behavior features extraction and behavior similarity matrix clustering) to distinguish shilling profiles from genuine ones. The behavior features reflect the behavior difference between genuine and shilling profiles, and the behavior similarity matrix clustering is to cluster shilling profiles based on their highly similar behaviors. Experimental results on the MovieLens and the sampled Amazon review datasets indicate that BS-SC outperforms the baseline unsupervised approaches, even when the prior knowledge is given for them.

关键词： Feature extraction Recommender systems Collaboration Training Correlation clustering algorithms Detection algorithms Collaborative recommender systems shilling attacks detection behavior analysis spectral clustering

来源：评论

学校读者我要写书评

暂无评论

A Cluster-Principal-Component-Analysis-Based Indoor Positioning Algorithm

引用

IEEE INTERNET OF THINGS JOURNAL 2021年第1期8卷 187-196页

作者： Li, Ang Fu, Jingqi Shen, Huaming Sun, Sizhou Shanghai Univ Dept Mechatron Engn & Automat Shanghai 200444 Peoples R China

Indoor location-based service is emerging as the crucial application of the Internet of Things, which promotes the advance of relevant technology in the indoor scenario. Several positioning algorithms are proposed for different indoor configurations in recent years. The fingerprint-based indoor positioning algorithm has drawn much attention because of the good positioning performance without additional hardware. However, the false fingerprint matching frequently incurs due to the complexity of the indoor positioning environment and affects the positioning accuracy. In this article, a principal component analysis (PCA)-assisted indoor positioning algorithm based on the adaptive hierarchical clustering algorithm (PAHC) is proposed, which can improve the positioning accuracy through aggregating the reference points (RPs) and conducting the cluster-based PCA (C-PCA) features extraction. More specifically, a clustering termination method is proposed to obtain reasonable RPs clusters adaptively according to the preset RPs. A two-stage fingerprint matching algorithm is proposed based on the C-PCA to further increase the difference between similar RPs and thus improving the positioning accuracy. To verify the proposed algorithm, an indoor wireless system is established in the practical indoor scenario. The experimental results indicate that the proposed PAHC algorithm can increase the positioning accuracy by 9.3% compared with the conventional fingerprint algorithm.

关键词： clustering algorithms Principal component analysis Interference Estimation Internet of Things Wireless sensor networks Feature extraction Fingerprint matching algorithm hierarchical clustering indoor positioning algorithm principal component analysis (PCA) sums-of-squared deviations

来源：评论

学校读者我要写书评

暂无评论

Research on Medical Question Answering System Based on Knowledge Graph

引用

IEEE ACCESS 2021年 9卷 21094-21101页

作者： Jiang, Zhixue Chi, Chengying Zhan, Yunyun Univ Sci & Technol LiaoNing Sch Comp & Software Anshan 114031 Peoples R China Technol Univ Dublin Coll Sci & Hlth Dublin D08 X622 Ireland

To meet the high-efficiency question answering needs of existing patients and doctors, this system integrates medical professional knowledge, knowledge graphs, and question answering systems that conduct man-machine dialogue through natural language. This system locates the medical field, uses crawler technology to use vertical medical websites as data sources, and uses diseases as the core entity to construct a knowledge graph containing 44,000 knowledge entities of 7 types and 300,000 entities of 11 kinds. It is stored in the Neo4j graph database, using rule-based matching methods and string-matching algorithms to construct a domain lexicon to classify and query questions. This system has specific practical value in the medical field knowledge graph and question answering system.

关键词： Medical diagnostic imaging Diseases Knowledge discovery Knowledge based systems Data mining clustering algorithms Semantics Natural language processing knowledge graph question and answer system medical knowledge

来源：评论

学校读者我要写书评

暂无评论

Blockchain-Based Data Dissemination Scheme for 5G-Enabled Softwarized UAV Networks

IEEE TRANSACTIONS ON GREEN COMMUNICATIONS AND NETWORKING

引用

IEEE TRANSACTIONS ON GREEN COMMUNICATIONS AND NETWORKING 2021年第4期5卷 1712-1721页

作者： Gupta, Rajesh Patel, Mohil Maheshkumar Tanwar, Sudeep Kumar, Neeraj Zeadally, Sherali Nirma Univ Dept Comp Sci & Engn Inst Technol Ahmadabad 382481 Gujarat India Thapar Inst Engn & Technol Dept Comp Sci & Engn Patiala 147004 Punjab India Univ Petr & Energy Studies Sch Comp Sci Dehra Dun 248007 Uttarakhand India Univ Kentucky Coll Commun & Informat Lexington KY 40506 USA

Unmanned aerial vehicles (UAVs) are widely used in applications such as surveillance, healthcare, rescue, and crowdsensing. However, the network management of these applications is quite challenging, given the mobility of UAVs. In this scenario, the network softwarization, which decouples the hardware from the network control functions, becomes essential. Moreover, the fifth-generation (5G) communication network is used to disseminate the data from controller-to-UAV and UAV-to-Internet of Things (IoT) enabled devices. However, handling the security and privacy of data against the eavesdropper is a challenging aspect that needs to be resolved efficiently. Most of the solutions presented in the literature are based on a centralized architecture having a single point of failure and are also vulnerable to various security attacks such as controller hijacking and man-in-the-middle attacks. Motivated by these facts, we present a blockchain-based secure data dissemination scheme for softwarized UAV networks. The proposed scheme can detect an eavesdropper, analyze malicious data, and mitigate various security attacks on the SDN controller. The performance analysis shows that the proposed scheme achieves superior results in detecting the eavesdropper and anomalous data before adding them into the blockchain compared to other state-of-the-art schemes.

关键词： Blockchains Unmanned aerial vehicles Security 5G mobile communication Codes clustering algorithms Reliability Blockchain smart contract software-defined network softwarization unmanned aerial vehicles 5G

来源：评论

学校读者我要写书评

暂无评论

A Pixel Cluster CNN and Spectral-Spatial Fusion Algorithm for Hyperspectral Image Classification With Small-Size Training Samples

引用

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING 2021年 14卷 4101-4114页

作者： Dong, Shuxian Quan, Yinghui Feng, Wei Dauphin, Gabriel Gao, Lianru Xing, Mengdao Xidian Univ Sch Elect Engn Dept Remote Sensing Sci & Technol Xian 710071 Peoples R China Univ Paris XIII Inst Galilee Lab Informat Proc & Transmiss F-93430 Villetaneuse France Chinese Acad Sci Aerosp Informat Res Inst Key Lab Digital Earth Sci Beijing 100094 Peoples R China Xidian Univ Acad Adv Interdisciplinary Res Xian 710071 Peoples R China

Convolutional neural networks (CNNs) can automatically learn features from the hyperspectral image (HSI) data, avoiding the difficulty of manually extracting features. However, the number of training samples for the classification of HSIs is always limited, making it difficult for CNN to obtain effective features and resulting in low classification accuracy. To solve this problem, a pixel cluster CNN and spectral-spatial fusion (SSF) algorithm for hyperspectral image classification with small-size training samples is proposed in this article. First, spatial information is extracted by the gray level co-occurrence matrix. Then, spatial information and spectral information are fused by means of bands superposition, forming spectral-spatial features. To expand the number of training samples, the pixels after SSF are combined into pixel clusters according to a certain rule. Finally, a CNN framework is utilized to extract effective features from the pixel clusters. Experiments based on three standard HSIs demonstrate that the proposed algorithm can get better performance than the conventional CNN and also outperforms other studied algorithms in the case of small training set.

关键词： Feature extraction Training clustering algorithms Data mining Classification algorithms Hyperspectral imaging Support vector machines Convolutional neural network (CNN) hyperspectral image classification (HSIC) pixel cluster small training set spectral-spatial fusion (SSF)

来源：评论

学校读者我要写书评

暂无评论

Design of Fuzzy Ensemble Architecture Realized With the Aid of FCM-Based Fuzzy Partition and NN With Weighted LSE Estimation

引用

IEEE TRANSACTIONS ON FUZZY SYSTEMS 2021年第3期29卷 569-583页

作者： Roh, Seok-Beom Oh, Sung-Kwun Pedrycz, Witold Fu, Zunwei Univ Suwon Sch Elect & Elect Engn Hwaseong Si 18323 South Korea Linyi Univ Res Ctr Big Data & Artificial Intelligence Linyi 276005 Shandong Peoples R China Univ Alberta Dept Elect & Comp Engn Edmonton AB T6G 2R3 Canada Polish Acad Sci Syst Res Inst PL-01447 Warsaw Poland King Abdulaziz Univ Fac Engn Dept Elect & Comp Engn Jeddah 21589 Saudi Arabia

Neural networks (NNs) with least square error (LSE) estimation form a certain type of single hidden layer feed-forward NNs. In this class of networks, the input connections (weights) and the biases of hidden neurons are generated randomly and fixed after being generated. The output connections are estimated by the LSE method rather than the back-propagation method. The random generation of the input connection weights and the hidden biases results in the larger number of hidden neurons to assure the quality of classification performance. To reduce the number of neurons in the hidden layer while maintaining the classification performance, we apply a "divide and conquer" strategy in this article. In other words, we divide an overall input space into several subspaces by using information granulation technique (Fuzzy C-Means clustering algorithm) and determine the local decision boundaries among related subspaces. A decision boundary defined in the input space can be considered as being composed of several decision boundaries defined in subspaces that form the entire input space. For the decision boundaries defined in the subspaces, their nonlinearity becomes lower in comparison with the one being encountered when considering the entire input space. Through the weighted LSE estimation instead of using the LSE estimation method, the connections of several NNs can be estimated without interfering with each other. After estimating the weights, the decision boundaries defined in the related subspaces are merged to a single decision boundary by using fuzzy ensemble technique. Several machine learning datasets and one real world application dataset are used to evaluate and validate the proposed fuzzy ensemble classifier. Based on the experimental results, the proposed classifier shows better classification performance when compared with the performance of some selected classifiers.

关键词： Estimation Artificial neural networks Neurons clustering algorithms Fuzzy logic Error analysis Fuzzy clustering ensemble classifier neural networks (NNs) with least square error (LSE) estimation regularization factor weighted least square error estimation (WLSE)

来源：评论

学校读者我要写书评

暂无评论

Probing the Relation Between Community Evolution in Dynamic Residue Interaction Networks and Xylanase Thermostability

引用

IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS 2021年第2期18卷 686-696页

作者： Zhang, Chen Ding, Yanrui Jiangnan Univ Lab Media Design & Software Technol Wuxi 214122 Jiangsu Peoples R China Jiangnan Univ Lab Media Design & Software Technol Key Lab Ind Biotechnol Wuxi 214122 Jiangsu Peoples R China

Residue-residue interactions are the basis of protein thermostability. The molecular conformations of Streptomyces lividans xylanase (xyna_strli) and Thermoascus aurantiacus xylanase (xyna_theau) at 300K, 325K and 350K were obtained by Molecular Dynamics (MD) simulations. Dynamic weighted residue interaction networks were constructed and the rigid-communities were detected using the ESPRA algorithm and the Evolving Graph+Fast-Newman algorithm. The residues in the rigid-communities are primarily located in loop2, short helixes alpha 2',alpha 3',alpha 4' and helixes alpha 3 and alpha 4. Thus, the rigid-community is close to the N-terminus of xylanase, which is usually stabilized to increase thermostability using site-directed mutagenesis. The evolution of the rigid-community with increasing temperature shows a stable synergistic interaction between loop2, alpha 2',alpha 3' and alpha 4' in xyna_theau. In particular, the short helixes alpha 2' and alpha 3' form a "thermo helix" to promote thermostability. In addition, tight global interactions between loop2, alpha 2',alpha 3',alpha 3,alpha 4' and alpha 4 of xyna_theau are identified, consisting mainly of hydrogen bonds, van der Waals forces and pi-pi stacking. These residue interactions are more resistant to high temperatures than those in xyna_strli. Robust residue interactions within these secondary structures are key factors influencing xyna_strli and xyna_theau thermostability. Analyzing the rigid-community can elucidate the cooperation of secondary structures, which cannot be discovered from sequence and 3D structure alone.

关键词： Proteins Heuristic algorithms clustering algorithms Immune system Resistance heating Bridges Evolution (biology) Dynamic residue interaction networks rigid-community thermostability xylanase

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：