检索结果-内蒙古大学图书馆

Personalized Book Intelligent Recommendation System Design for University Libraries Based on IBCF Algorithm

IEEE ACCESS 2024年 12卷 82015-82032页

作者： Lin, Na Jilin Agr Univ Changchun 130118 Peoples R China

With the digital transformation and improvement of university library information technology, readers' demands for library services are increasingly diversified and personalized. They are no longer satisfied with the traditional borrowing services, but hope that the library can provide more accurate and personalized recommendation services. To solve these problems, this study first proposes an improved item-based collaborative filtering recommendation algorithm based on the mean model representation. Then, combining this algorithm with user-based collaborative filtering recommendation algorithm, an improved item-based collaborative filtering algorithm is designed. The results showed that the CPU usage of the whole system was not high during the operation of the improved item-based collaborative filtering recommendation algorithm, with an average usage rate of about 9.8%. The minimum root mean square error of the algorithm was 0.013 and the runtime was 12000 mu s. Compared with existing the similar systems, when the number of users exceeded 200, the response speed was significantly reduced by more than 50%, and the coverage rate reached more than 90%. In summary, the personalized intelligent book recommendation system for university library proposed in the study has the advantages of high coverage, low resource consumption, high accuracy and so on, which can provide readers with more accurate and personalized recommendation services.

关键词： Recommender systems clustering algorithms Collaborative filtering Prediction algorithms Filling Vectors SIBCF algorithm UBCF collaborative filtering book recommendations mean model vector representation

来源：评论

学校读者我要写书评

暂无评论

Analysis of Network Intrusion Detection Based on Semi-Supervised and SS-DGM

引用

IEEE ACCESS 2024年 12卷 170148-170160页

作者： Yu, Xiao Liu, Chang Wang, Jie Liu, Chang Tian, Li Zhou, Liang State Grid Hubei Elect Power Res Inst Wuhan 430074 Peoples R China

The rapid advancement of technology has made network security a hot topic of concern for researchers worldwide. Therefore, to improve the accuracy and real-time response capability of network intrusion detection systems, and to effectively detect and analyze network attack forms in complex network environments, starting from the drawbacks of existing network intrusion detection methods and the vulnerable attack modes of networks, this study first introduces variational auto-encoder to improve the semi-supervised intrusion detection algorithm. Subsequently, the labeled dataset is expanded using K-means clustering algorithm and collaborative training algorithm. Finally, a novel network intrusion detection model based on multi-level semi-supervised intrusion detection algorithm is proposed. The experimental outcomes denoted that the model achieved the highest classification accuracy of 93.08%, 92.12%, 91.22%, and 96.38% for four common types of network attacks: denial of service attacks, unauthorized access from remote machines, unauthorized access attacks by ordinary users to local superuser privileges, and cross-site scripting attacks, respectively. The results of the network attack confusion matrix indicated that the proposed model could detect and classify all types of network attacks, with excellent detection applicability and classification efficiency, and its scores were all above 60 points. In addition, the running time of the proposed model was only 25.64 seconds. From this, the proposed method effectively improves the dynamic adaptability and accuracy of network intrusion detection, providing an efficient and accurate solution for network security defense.

关键词： Accuracy Data models Deep learning Network security Feature extraction Decoding Classification algorithms Anomaly detection clustering algorithms Training Semisupervised learning Intrusion detection Encoding Semi-supervised SS-DGM variational auto-encoder clustering intrusion detection

来源：评论

学校读者我要写书评

暂无评论

Optimally Ordered Orthogonal Neighbor Joining Trees for Hierarchical Cluster Analysis

引用

IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS 2024年第8期30卷 5034-5046页

作者： Ge, Tong Luo, Xu Wang, Yunhai Sedlmair, Michael Cheng, Zhanglin Zhao, Ying Liu, Xin Deussen, Oliver Chen, Baoquan Shandong Univ Dept Comp Sci Jinan 250100 Shandong Peoples R China Univ Stuttgart D-70174 Stuttgart Germany SIAT Shenzhen VisuCA Key Lab Shenzhen 518055 Guangdong Peoples R China Cent South Univ Changsha 410017 Hunan Peoples R China Beijing Genom Inst BGI Shenzhen Shenzhen 518083 Guangdong Peoples R China Univ Konstanz D-78464 Constance Germany Peking Univ Beijing 100871 Peoples R China

We propose to use optimally ordered orthogonal neighbor-joining (O-3 NJ) trees as a new way to visually explore cluster structures and outliers in multi-dimensional data. Neighbor-joining (NJ) trees are widely used in biology, and their visual representation is similar to that of dendrograms. The core difference to dendrograms, however, is that NJ trees correctly encode distances between data points, resulting in trees with varying edge lengths. We optimize NJ trees for their use in visual analysis in two ways. First, we propose to use a novel leaf sorting algorithm that helps users to better interpret adjacencies and proximities within such a tree. Second, we provide a new method to visually distill the cluster tree from an ordered NJ tree. Numerical evaluation and three case studies illustrate the benefits of this approach for exploring multi-dimensional data in areas such as biology or image analysis.

关键词： Couplings clustering algorithms Layout Visualization Biology Germanium Sorting Neighbor joining leaf ordering orthogonal layout

来源：评论

学校读者我要写书评

暂无评论

A Novel clustering Strategy-Based Sink Path Optimization for Wireless Sensor Network

引用

IEEE SENSORS JOURNAL 2022年第20期22卷 20042-20052页

作者： Xie, Meng Pi, Dechang Dai, Chenglong Xu, Yue Li, Bentian Nanjing Univ Aeronaut & Astronaut Coll Comp Sci & Technol Nanjing 211106 Jiangsu Peoples R China Jiangnan Univ Coll Artificial Intelligence & Comp Sci Wuxi 610101 Jiangsu Peoples R China

To optimize the path of mobile sinks in the wireless sensor network (WSN) and use of 5G signals to transmit information, a new clustering strategy is proposed, which not only balances the number of cluster nodes but also shortens the path of mobile sinks. By moving rendezvous points (RPs) to change the cluster position, we first propose a metaheuristics clustering algorithm named the dynamic clustering-based rectangular evolutionary algorithm (DCREA). The algorithm adjusts the clustering structure of the network by limiting the cluster movement range;then, it is optimized in combination with the greedy algorithm. Extensive experimental results show that this method can significantly shorten the path of the mobile sink, increase the efficiency of the mobile sink, and improve the quality of signal transmission while maintaining the balance of the cluster structure. The experimental results show that the DCREA is the most effective and indicate the effectiveness of the algorithm by comparing it with related recent algorithms.

关键词： Sensors clustering algorithms Wireless sensor networks Periodic structures Metaheuristics Heuristic algorithms Energy consumption clustering metaheuristics mobile sink path optimization wireless sensor network (WSN)

来源：评论

学校读者我要写书评

暂无评论

Non-Euclidean Grid-Partitioning to Mitigate Cascading Risk in Multi-Infeed HVDC System

引用

IEEE TRANSACTIONS ON POWER SYSTEMS 2024年第5期39卷 6426-6440页

作者： Wang, Xiaohui Song, Kaige Hao, Quanrui Gao, Feng Shandong Univ Sch Elect Engn Key Lab Power Syst Intelligent Dispatch & Control Minist Educ Jinan 250061 Peoples R China Shandong Univ Sch Control Sci & Engn Jinan 250061 Peoples R China

The presence of multiple high voltage direct current (HVDC) systems in close proximity creates voltage-related cascading risks that are not adequately addressed by conventional grid-partitioning. This paper proposes an improved partitioning scheme to mitigate these new risks in addition to conventional objective of preventing parallel power flow from transferring adversely. Unlike classical partitioning approach, which relies solely on either optimization or clustering, our proposed bi-level architecture includes an additional HVDC clustering before optimization. However, this paper innovatively reveals that the distribution of correlation data to be used in clustering is non-Euclidean due to unusual equivalent reactance different from the normal operating condition, resulting from HVDC station's reactive power control. This non-Euclidean distribution makes heuristic clustering algorithms infeasible. To address this issue, an alternative solution is proposed to embed the correlation data into a dimension-reduced eigenspace spanned by selected eigenvectors, allowing clustering to be performed. The optimization implementing other objectives inherits the results of HVDC clustering as constraint, and the graphic betweenness weighted by power flows is presented to promote efficiency. Our proposed scheme is validated using cases studied in modified IEEE 118-bus benchmark system and practical regional grid, demonstrating its effectiveness in mitigating cascading risks in multi-infeed HVDC systems.

关键词： HVDC transmission clustering algorithms Impedance Couplings Correlation Optimization Partitioning algorithms Cascading failure clustering electromagnetic loop grid partitioning HVDC non-Euclidean parallel power flow reactive power control spectral embedding triangle inequality

来源：评论

学校读者我要写书评

暂无评论

Data-Driven Identification of Industrial Clusters: A Patent Analysis Approach

引用

IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT 2024年 71卷 15422-15437页

作者： Lin, Wenguang Wang, Ting Chen, Zhizhen Xiao, Renbin Xiamen Univ Technol Sch Mech & Automot Engn Xiamen 361024 Peoples R China Univ Greenwich Sch Business London SE10 9LS England Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Wuhan 430074 Peoples R China

Accurate identification of industrial clusters (IIC) serves as a reference for regional economic policymaking and enterprise development decision-making. Although data-driven methods have been extensively used in previous studies to support objective and effective work, both the data sources and research algorithms have significant shortcomings for IIC. To address these challenges, this article proposes a novel research framework that integrates patent mining and machine learning. Patents, with their quantifiable knowledge attributes and accessibility from public databases, are particularly suited for macrolevel analysis of innovation activities, providing robust support for identifying and analyzing clusters on a national scale, especially knowledge-intensive ones. This article introduces an improved density-based parameter adaptive algorithm designed to effectively carry out IIC based on the geographical location of patent applicants. Based on spatial cluster types defined by Markusen (1996), target clusters are classified using patent analysis. Four quantitative indexes-scale, output, efficiency, and quantity-are proposed to evaluate clusters based on their spatial structure and industrial organization. The practical application is demonstrated through a case study of China's flexible electronics industry. In addition, the Silhouette Coefficient index is employed to compare the effectiveness of the proposed algorithm against other methods. This article advances the theory of IIC, and provides foundation for scholars, calling for empirical research on industrial clusters from the perspective of individual enterprises. It also provides practical guidance for enterprises and policymakers on the application of IIC.

关键词： Patents clustering algorithms Technological innovation Industries Engineering management Economics Production Adaptation models Heuristic algorithms Accuracy Flexible electronics industry identification industrial clusters patent analysis spatial distribution

来源：评论

学校读者我要写书评

暂无评论

Understanding Pooling in Graph Neural Networks

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2024年第2期35卷 2708-2718页

作者： Grattarola, Daniele Zambon, Daniele Bianchi, Filippo Maria Alippi, Cesare Univ Svizzera italiana USI Fac Informat CH-6904 Lugano Switzerland UiT Arctic Univ Norway Dept Math & Stat N-9019 Tromso Norway Norwegian Res Ctr NORCE N-5838 Bergen Norway Politecn Milan Dept Elect Informat & Bioengn I-20133 Milan Italy

Many recent works in the field of graph machine learning have introduced pooling operators to reduce the size of graphs. In this article, we present an operational framework to unify this vast and diverse literature by describing pooling operators as the combination of three functions: selection, reduction, and connection (SRC). We then introduce a taxonomy of pooling operators, based on some of their key characteristics and implementation differences under the SRC framework. Finally, we propose three criteria to evaluate the performance of pooling operators and use them to investigate the behavior of different operators on a variety of tasks.

关键词： Task analysis Taxonomy Aggregates Point cloud compression Laplace equations Convolution clustering algorithms Dimensionality reduction graph neural networks (GNNs)

来源：评论

学校读者我要写书评

暂无评论

Leveraging Drone-Assisted Surveillance for Effective Forest Conservation: A Case Study in Australias Daintree Rainforest

引用

IEEE INTERNET OF THINGS JOURNAL 2024年第19期11卷 31167-31179页

作者： Melhim, Loai Kayed B. Jemmali, Mahdi Boulila, Wadii Alazab, Mamoun Rani, Shalli Campbell, Hamish Amdouni, Hajer Univ Hafr Al Batin Coll Appl Med Sci Dept Hlth Informat Management & Technol Hafar al Batin 39524 Saudi Arabia Univ Sharjah Coll Comp & Informat Comp Sci Dept Sharjah U Arab Emirates Prince Sultan Univ Robot & Internet Things Lab Riyadh 11586 Saudi Arabia Univ Manouba Natl Sch Comp Sci RIADI Lab Manouba 2010 Tunisia Charles Darwin Univ IT & Environm Casuarina NT 2904 Australia Chitkara Univ Inst Engn & Technol Patiala 147001 Punjab India Charles Darwin Univ Fac Sci & Technol Environm Sci Casuarina NT 0810 Australia Univ Tunis Higher Sch Econ & Commercial Sci Tunis 1089 Tunisia

Nowadays, there is global consensus on the threats to forests and their crucial role in mitigating global warming and its impact on Earth's biodiversity. Both private and public entities, alongside governments, have engaged the most advanced technologies to safeguard and monitor forests against encroachment. This article examines the application of various drone technologies in the surveillance of forest areas. The system described herein employs drones to continuously survey forests, recording any changes, particularly in instances of encroachment or fire. The data captured are transmitted to a control unit for subsequent analysis. To circumvent the risk of task failure due to technical challenges, monitoring tasks within a predefined flight duration are allocated to the available drones. Given the critical nature of timing in the success of these tasks, this study addresses the forest monitoring challenge by seeking to minimize the maximum time required to complete all monitoring tasks. This challenge was addressed through the development of a suite of enhanced algorithms aimed at optimizing task efficiency. The primary goal of the proposed methodology is to afford the monitoring system additional time, thereby enabling the handling of an increased volume of tasks and providing support to firefighting teams in responding to forest fires. The system's adaptability to new, unforeseen forest fire scenarios through the generation of novel solutions is also discussed. Extensive testing involving 1350 different scenarios has demonstrated the effectiveness of the proposed algorithms in reducing the maximum time needed for the completion of surveillance tasks by drones. The most effective algorithm was the two-group clustering algorithm (TGC), which achieved a success rate of 97.2%, with an average gap of less than 0.001 and an average computation time of 0.016 s. Furthermore, the application of this methodology to a case study of the Daintree Rainforest in Australia show

关键词： Forestry Drones clustering algorithms Surveillance Internet of Things Global warming Climate change Fire safety Fires Environmental monitoring Data processing decision making drone fire fighting information systems monitoring system

来源：评论

学校读者我要写书评

暂无评论

PICTURE-A Framework to Assess the Degree of Approximation of Summarized Time Series

引用

IEEE ACCESS 2024年 12卷 178717-178733页

作者： Bianchini, Devis Garda, Massimiliano Univ Brescia Dept Informat Engn I-25123 Brescia Italy

The analysis of time series data, which represents dynamic phenomena through sequences of observations, is greatly influenced by Big Data. Both the sheer volume and the advanced capabilities of Big Data significantly impact on how these analyses are conducted, enabling more comprehensive and detailed insights. Recent studies have promoted the use of data summarization techniques, for instance through incremental clustering, to address the challenges of Big Data volume. These techniques quickly capture data evolution, thereby helping domain experts make informed and proactive decisions by leveraging a concise representation of time series. However, although incremental clustering efficiently reduces data volume and retains key statistical information, it is important to evaluate the accuracy of the summarized version compared to the original time series data. This assessment is critical when the summarized data is used as the basis for complex analytical pipelines, such as those for pattern recognition and anomaly detection. Moved by these premises and starting from an empirical experience on the definition of a metric to assess the adherence of summarised time series to the original data stream, in this paper: (i) we propose a variant of a renowned quality metric for incremental clustering based on an abstract model of clustering data structures, to assess the extent to which the time series summary accurately captures the dynamics of the original data;(ii) we present PICTURE (Python-based Incremental clustering for Time series Representation and Evaluation) a framework featuring four widely used incremental clustering algorithms from the literature, equipped with modules for execution, representation, and evaluation of clustering results applied to time series according to the abstract model;(iii) we conduct an extensive qualitative and quantitative analysis of incremental clustering results on a synthetic and two real-world datasets using the PICTURE framework, to

关键词： Time series analysis Measurement clustering algorithms Telemetry European Space Agency Time measurement Data models Big Data Accuracy Accelerometers Time series data summarisation incremental clustering data stream analysis python framework

来源：评论

学校读者我要写书评

暂无评论

Fast Approximated Multiple Kernel K-Means

引用

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2024年第11期36卷 6171-6180页

作者： Wang, Jun Tang, Chang Zheng, Xiao Liu, Xinwang Zhang, Wei Zhu, En Zhu, Xinzhong Natl Univ Def Technol Sch Comp Changsha 410073 Peoples R China China Univ Geosci Sch Comp Sci Wuhan 430074 Peoples R China Xidian Univ State Key Lab Integrated Serv Networks Xian 710071 Peoples R China Qilu Univ Technol Shandong Comp Sci Ctr Natl Supercomp Ctr Jinan Shandong Prov Key Lab Comp NetworksShandong Acad Jinan 250000 Peoples R China Zhejiang Normal Univ Sch Comp Sci & Technol Sch Artificial Intelligence Jinhua 321004 Peoples R China

Multiple Kernel clustering (MKC) has emerged as a prominent research domain in recent decades due to its capacity to exploit diverse information from multiple views by learning an optimal kernel. Despite the successes achieved by various MKC methods, a significant challenge lies in the computational complexity associated with generating a consensus partition from the optimal kernel matrix, typically of size n x n , where n represents the number of samples. This computational bottleneck restricts the practical applicability of these methods when confronted with large-scale datasets. Furthermore, certain existing MKC algorithms derive the consensus partition matrix by fusing all base partitions. However, this fusion process may inadvertently overlook critical information embedded in individual base kernels, potentially leading to inferior clustering performance. In light of these challenges, we introduce an innovative and efficient multiple kernel k -means approach, denoted as FAMKKM. Notably, FAMKKM incorporates two approximated partition matrices instead of the original individual partition matric for each base kernel. This strategic substitution significantly reduces computational complexity. Additionally, FAMKKM leverages the original kernel information to guide the fusion of all base partitions, thereby enhancing the quality of the resulting consensus partition matrix. Finally, we substantiate the efficacy and efficiency of the proposed FAMKKM through extensive experiments conducted on six benchmark datasets. Our results demonstrate its superiority over state-of-the-art methods.

关键词： Kernel clustering algorithms Partitioning algorithms Computational complexity Optimization Linear programming Matrix decomposition Multi-view clustering partition learning multiple kernel k -means data fusion

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：