检索结果-内蒙古大学图书馆

2nd International Conference on Disruptive Technologies, ICDT 2024

作者： Kaushik, Abha Attri, Shree Harsh Jha, Ravi Shankar Computer Science and Engineering Gl Bajaj Institute of Technology and Management Greater Noida India Computer Science and Engineering Sharda University Greater Noida India School of Computing Dit University Dehradun India

ISBN: (纸本)9798350371055

Text summarization is a significant task in natural language processing (NLP) since it seeks to reduce massive amount of textual information into concise and logical summaries. The rapid expansion of digital content has raised the need for efficient text summarizing strategies. An in-depth evaluation of current trends and future directions in text summarization is provided during the investigations of summarizing techniques. This Survey begins by outlining established methodologies, such as extractive and abstractive methods, and noting their current challenges. The emergence of deep learning models in text summarization such as transformer-based models and encoderdecoder architectures will be discussed. These models have shown inspiring results in abstractive summarization and semantic meaning capture. Additionally, Investigation process carries current advances in text summarization, addressing the issues associated with summarizing information from numerous sources and languages. Along with advancement in text summarization, intriguing future avenue in text summarizing research, such as personalized summarization, domain-specific summarization, and the incorporation of external knowledge sources has also been identified. Consequences of these directions for real-world ap- plications and user preferences are analyzed. Objective is to emphasize the merits and demerits of existing approaches, identify significant difficulties, and propose potential future research directions. Finally, the goal of this review is to inspire and lead the future development of more robust and effective text summarizing algorithms. © 2024 IEEE.

关键词： deep learning models Natural language processing Text Summarization transformer-based models

来源：评论

学校读者我要写书评

暂无评论

Optimizing high-utility item mining using hybrid dolphin echolocation and Boolean grey wolf optimization

引用

Journal of Ambient Intelligence and Humanized computing 2023年第3期14卷 2327-2339页

作者： Pazhaniraja, N. Sountharrajan, S. Suganya, E. Karthiga, M. School of Computing and Science VIT Bhopal University Madhya Pradesh Bhopal India Department of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Chennai India Department of Information Technology Sri Sivasubramaniya Nadar College of Engineering Tamil Nadu Chennai India Bannari Amman Institute of Technology Tamilnadu Sathyamangalam India

High-utility item-set mining has become a very crucial topic of research in recent times. It is used to find profitable products utilizing both profit factors and the quantity as opposed to the association rule mining or the frequent item-set mining. There are many algorithms that are proposed for mining, such high-utility item-sets, and almost all algorithms will need to grip an exponential search space for learning the high-utility item-sets. To handle the exponential search space, metaheuristic algorithms are used to find optimal solutions. Here, an optimization method depending on the grey wolf optimization and dolphin echolocation-optimization has been proposed for finding high-utility item-sets. The grey wolf optimization algorithm can be an interesting one owing to the strategy of group foraging. All conventional grey wolf optimization will deal with the problems of continuous optimization. In dolphin echolocation-optimization, as soon as the dolphin manages to reach the food, the search is reduced, and there is an increasing click made that directs food to the site where food is. The procedure also feigns dolphin echolocation by means of preventive exploration, and this is relational its target distance. In connection to the problems in optimization, there can be a particular features-set, which include progress of various hunting preys that sing echo sounding among dolphins, those having problems in answering. A hybrid dolphin echolocation-optimization (hybrid dolphin echolocation-binary grey wolf optimization) is proposed with a novel mechanism for finding global solutions. In this work, hybrid dolphin echolocation-binary grey wolf optimization algorithm is proposed. The dolphin echolocation-optimization had taken as it can be a promising solution and the Boolean format of grey wolf optimization is used. It works as a global search solution to distinguish the high-utility item-sets. The results prove that the dolphin echolocation-binary grey wolf optimizat

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

Multi-model ensemble framework for analysis of psychopathic traits in heinous crime convicts

引用

Journal of Computational Social science 2025年第3期8卷 1-28页

作者： Singh, Aman Mohapatra, Subrajeet Department of CSE School of Computing MIT-ADT University Pune 412001 India Department of Computer Science and Engineering Birla Institute of Technology Mesra Jharkhand Ranchi 835215 India

The correlation between psychiatric disorder and criminality has been the subject of intense debate and scrutiny in recent years, in light of multiple violent incidents in India and other nations. To determine the severity of psychopathic traits or tendencies in heinous crime convicts, a revised method is proposed, and its results are correlated with those of the PCL-R administered by an experienced psychologist. With a focus on multidimensional behavioral and personality characteristics, these schemes have evaluated the degree of psychopathy in violent offenders. Utilizing a fuzzy mutual information-based feature selection method, it is established that 62 features are statistically significant out of 68. Accordingly, a set of five base classifiers are used to construct the first layer of the stacking ensemble, and a single meta-learner is used to develop the second layer of the proposed stacked ensemble model. The average accuracy for the proposed stacked ensemble model with SVM-NL as meta-learner 87.74% is the highest among all the configured stacked models. However, precision and f1-score for the proposed model are 86% and 85% respectively. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Machine learning Psychopathy risk markers Quantitative risk assessment Stacked ensemble Violent recidivism

来源：评论

学校读者我要写书评

暂无评论

Dynamic Visible Light Positioning Based on Enhanced Visual Target Tracking

引用

China Communications 2023年第10期20卷 276-291页

作者： Xiangyu Liu Jingyu Hao Lei Guo Song Song School of Information Science and Engineering Shenyang Ligong UniversityShenyang 110158China Department of Computer Science and Engineering Southern University of Science and TechnologyShenzhen 518055China School of Computer Science and Engineering Northeastern UniversityShenyang 110819China School of Communication and Information Engineering Chongqing University of Posts and TelecommunicationsChongqing 400065China Hangzhou Institute of Advanced Technology Hangzhou 310056China Institute of Intelligent Communications and Network Security Chongqing University of Posts and TelecommunicationsChongqing 400065China

In visible light positioning systems,some scholars have proposed target tracking algorithms to balance the relationship among positioning accuracy,real-time performance,and ***,there are still two problems:(1)When the captured LED disappears and the uncertain LED reappears,existing tracking algorithms may recognize the landmark in error;(2)The receiver is not always able to achieve positioning under various moving *** this paper,we propose an enhanced visual target tracking algorithm to solve the above ***,we design the lightweight recognition/demodulation mechanism,which combines Kalman filtering with simple image preprocessing to quickly track and accurately demodulate the ***,we use the Gaussian mixture model and the LED color feature to enable the system to achieve positioning,when the receiver is under various moving *** results show that our system can achieve high-precision dynamic positioning and improve the system’s comprehensive performance.

关键词： visible light positioning visual target tracking gaussian mixture model kalman filtering system performance

来源：评论

学校读者我要写书评

暂无评论

From Static to Dynamic: Adapting Landmark-Aware Image Models for Facial Expression Recognition in Videos

引用

IEEE Transactions on Affective computing 2025年第2期16卷 624-638页

作者： Chen, Yin Li, Jia Shan, Shiguang Wang, Meng Hong, Richang Hefei University of Technology School of Computer Science and Information Engineering Hefei230601 China Chinese Academy of Sciences Key Laboratory of Intelligent Information Processing Institute of Computing Technology Beijing100190 China University of Chinese Academy of Sciences Beijing100049 China

Dynamic facial expression recognition (DFER) in the wild is still hindered by data limitations, e.g., insufficient quantity and diversity of pose, occlusion and illumination, as well as the inherent ambiguity of facial expressions. In contrast, static facial expression recognition (SFER) currently shows much higher performance and can benefit from more abundant high-quality training data. Moreover, the appearance features and dynamic dependencies of DFER remain largely unexplored. Recognizing the potential in leveraging SFER knowledge for DFER, we introduce a novel Static-to-Dynamic model (S2D) that leverages existing SFER knowledge and dynamic information implicitly encoded in extracted facial landmark-aware features, thereby significantly improving DFER performance. First, we build and train an image model for SFER, which incorporates a standard Vision Transformer (ViT) and Multi-View Complementary Prompters (MCPs) only. Then, we obtain our video model (i.e., S2D), for DFER, by inserting Temporal-Modeling Adapters (TMAs) into the image model. MCPs enhance facial expression features with landmark-aware features inferred by an off-the-shelf facial landmark detector. And the TMAs capture and model the relationships of dynamic changes in facial expressions, effectively extending the pre-trained image model for videos. Notably, MCPs and TMAs only increase a fraction of trainable parameters (less than +10%) to the original image model. Moreover, we present a novel Emotion-Anchors (i.e., reference samples for each emotion category) based Self-Distillation Loss to reduce the detrimental influence of ambiguous emotion labels, further enhancing our S2D. Experiments conducted on popular SFER and DFER datasets show that we have achieved a new state of the art. 2010-2012 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

A Real-Time P2P Bot Host Detection in a Large-Scale Network Using Statistical Network Traffic Features and Apache Spark Streaming Platform 8

A Real-Time P2P Bot Host Detection in a Large-Scale Network ...

引用

8th IEEE International Conference for Convergence in technology, I2CT 2023

作者： Saravanan, S. Prakash, G. Uma Maheswari, B. Amrita School of Computing Bengaluru Amrita Vishwa Vidyapeetham Dept. of Computer Science and Engineering India Vellore Institute of Technology School of Computer Science and Engineering Vellore India

ISBN: (纸本)9798350334012

Nowadays, Peer-to-Peer (P2P) bots play a significant role in launching attacks such as phishing, distributed denial-of-service (DDoS), email spam, click fraud, cryptocurrency mining, etc. The analysis of statistical network traffic features of hosts is one of the commonly used methods to detect P2P bots. Modern P2P bot detection systems need to extract features from massive streaming network traffic as the size of the Internet keeps increasing every day. However, traditional detection systems have trouble detecting bots in real-time in large-scale networks as they are not implemented on big data streaming platforms. Hence, this work proposes a network flow-based P2P bot detection system implemented on Apache Spark Structured Streaming Platform to detect P2P bots in real time by analyzing massive streaming network traffic data generated from large-scale networks. Such detection of P2P bots is based on statistical network traffic features: destination diversity ratio, control packets ratio, and total source bytes sent in a flow. There are two components in the proposed system: the first component detects potential P2P hosts using the Destination Diversity Ratio (DDR), and the second component finds out P2P bot hosts from the P2P hosts identified by the first component. Furthermore, the performance of the detection components depends on the time window at which statistical features are extracted. Hence, this work also conducted experiments to study the effect of different time windows on detection components. The proposed system is evaluated using real-world datasets and achieves a True Positive Rate (TPR) of 99.87%. © 2023 IEEE.

关键词： Denial-of-service attack

来源：评论

学校读者我要写书评

暂无评论

A Group-Based Many-Task Collaborative Optimization Framework for Evolutionary Robots Design

引用

IEEE Transactions on Systems, Man, and Cybernetics: Systems 2025年第5期55卷 3492-3505页

作者： Hou, Yaqing Yu, Zhaoping Guo, Zheng Pei, Wenbin Wu, Yaoxin Ge, Hongwei Xue, Bing Zhang, Mengjie Dalian University of Technology College of Computer Science and Technology Dalian116024 China Avic Xi'an Aeronautics Computing Technique Research Institute 18th Research Department Xi'an710068 China Eindhoven University of Technology Department of Industrial Engineering and Innovation Sciences Eindhoven5612 AZ Netherlands Victoria University of Wellington School of Engineering and Computer Science Wellington6140 New Zealand

In evolutionary robotics (ER), the evolution of a robot's morphology (i.e., physical structure) or controller (i.e., control algorithm or instruction sequence) often entails tackling an extensive number of tasks. The use of evolutionary multitasking (EMT) in ER, which optimizes multiple tasks simultaneously by reusing potentially useful knowledge across diverse tasks, could improve the performance of problem-solving to each task. However, existing EMT methods do not fully use intertask correlations, limiting knowledge sharing. In view of this, this study introduces a novel framework, termed adaptive group-based collaborative optimization, tailored for handling optimization problems involving a large number of tasks within the ER domain simultaneously. The proposed framework divides tasks into groups according to their similarity and then proceeds through two principal stages, namely, intergroup knowledge separation and intragroup knowledge reunion. During intergroup knowledge separation stage, an adaptive method for selecting crossover operators enables source tasks to share useful knowledge to the target task across groups. During intragroup knowledge reunion stage, an adaptive knowledge combination strategy facilitates the target task in assimilating knowledge from multiple sources intragroup. We validated the efficacy of the proposed framework in both planar manipulators and hexapod robot experiments. The results indicate that our method outperforms existing state-of-the-art algorithms (i.e., MME, MMKT) on several metrics (e.g., mean fitness and quality diversity metrics). The proposed method can effectively improve the effectiveness and diversity of solutions in solving ER problems with a large number of tasks (e.g., 5 000 or 10 000), and has broad potential in practical ER applications. © 2025 IEEE.

关键词： Collaborative robots

来源：评论

学校读者我要写书评

暂无评论

Ensemble Learning Based Social engineering Fraud Detection Module for Cryptocurrency Transactions 9th

Ensemble Learning Based Social Engineering Fraud Detection M...

引用

9th International Conference on Mining Intelligence and Knowledge Exploration, MIKE 2023

作者： Pathak, Vishvesh Uma Maheswari, B. Geetha, S. Department of Computer Science and Engineering Amrita School of Computing Bengaluru Amrita Vishwa Vidyapeetham Bengaluru India School of Computer Science and Engineering Vellore Institute of Technology Chennai India

ISBN: (纸本)9783031440830

The practical applications of blockchains can far supersede the widely known trading and cryptocurrency realm. If any service provider is looking for a consistent, immutable, and multitenancy-supported ledger, then blockchain is the promising solution. Nowadays, social engineering attacks are prevalent. And the attackers deceive cryptocurrency traders. This work investigates various ensemble learning, neural network, and machine learning algorithms for fraud detection and identifies the best decision-making algorithm. It is observed that Adaptive Boosting (AdaBoost) algorithm outperforms with an accuracy of 98.92%. Further, the fraud detection module is integrated with an application developed for cryptocurrency transactions. Before a new transaction is committed to blockchain, The fraud detection module intervenes and alerts the user. We have also designed a test bed of deployable Peer-to-Peer (P2P) network to simulate cryptocurrency transaction. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Adaptive boosting

来源：评论

学校读者我要写书评

暂无评论

Mitigating Malware Propagation in Social Internet of Things Using an Exact Markov Chain-Based Epidemic Method

引用

IEEE Internet of Things Journal 2025年第13期12卷 24104-24118页

作者： Zhang, Hong Hu, Xinyi Shen, Yizhou Xu, Huibin Shen, Shigen Li, Ruidong Donghua University School of Computer Science and Technology Shanghai201620 China Newcastle University School of Computing Newcastle upon TyneNE1 7RU United Kingdom Huzhou University School of Information Engineering Huzhou313000 China Kanazawa University Institute of Science and Engineering Kakuma Kanazawa920-1192 Japan

In the Social Internet of Things (SIoT) environment, malware propagation is attracting more and more attention due to increasing damages. Markov chain models have been used to predict epidemic behavior qualitatively and quantitatively, but most of them model random propagation as a basic multiplicative factor. In this paper, we propose an epidemic model Susceptible-Infected without command-Infected with command (SII’), and derive an exact Markov chain for SIoT malware propagation. We also employ a Markov chain for an SIoT malware mitigation system that groups random devices alongside those with detected infections during the malware eradication process. This mitigation mechanism operates at the network scale, addressing the risks associated with large-scale SIoT deployments through a strategic, yet assertive, approach of widespread disconnections. Such a system effectively drives down the basic reproduction number to less than 1, preventing malware from gaining dominance over the network–all accomplished without modifying the recovery rate. We conducted experimental simulations of the proposed model’s dynamic predictions, and the experimental results show that the use of an exact Markov chain model better matches the benchmark results of our proposed model and also verifies the different effects of group-based mitigation in different SIoT contexts. © 2014 IEEE.

关键词： Markov chains

来源：评论

学校读者我要写书评

暂无评论

A Cost-Efficient Anonymous Authenticated and Key Agreement Scheme for V2I-Based Vehicular Ad-Hoc Networks

引用

IEEE Transactions on Intelligent Transportation Systems 2024年第9期25卷 12621-12630页

作者： Saleem, Muhammad Asad Li, Xiong Mahmood, Khalid Shamshad, Salman Alenazi, Mohammed J. F. Das, Ashok Kumar University of Electronic Science and Technology of China School of Computer Science and Engineering Chengdu611731 China Shenzhen Institute for Advanced Study University of Electronic Science and Technology of China Shenzhen518000 China National Yunlin University of Science and Technology Graduate School of Intelligent Data Science Yunlin Douliu64002 Taiwan The University of Lahore Department of Software Engineering Lahore54590 Pakistan King Saud University College of Computer and Information Sciences Department of Computer Engineering Riyadh11451 Saudi Arabia International Institute of Information Technology Center for Security Theory and Algorithmic Research Hyderabad500032 India

The rise of smart cities is directly connected to the increasing use of vehicles. The growing vehicle utilization has driven the emergence of Vehicular Ad-hoc Networks (VANETs), facilitating instant information exchange among vehicles. The system provides essential information regarding road conditions, traffic patterns, and more relevant data. VANETs encompass two fundamental categories of communication exchanges, namely Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I). V2I technology facilitates the integration of cars and transportation infrastructure, enabling effective communication between vehicles and infrastructure. Nevertheless, the potential of V2I communication has various security concerns arising from prevalent security threats. Current authentication techniques encounter challenges regarding complexity, security, and privacy considerations. We designed a hash-based lightweight and anonymous authentication scheme to address the aforementioned restrictions and enhance the effectiveness of authentication in V2I architecture. This scheme effectively combines identity, password, and bio-metric to enhance resistance against impersonation, denial of service, and privileged insider attacks. The devised scheme distinguishes itself by a comparative analysis and security proofs, highlighting its superior capability in guaranteeing secure authentication in V2I communication. The comprehensive security analysis conducted formally and informally showcases the robustness of the proposed solution against several threats. The performance evaluation results show that our scheme demonstrates a decrease in the computational cost of 51.40% approximately and a reduction in communication overhead of around 22.57%. These results establish the efficiency and scalability of the proposed scheme as a viable solution for V2I architecture. © 2000-2011 IEEE.

关键词： Vehicular ad hoc networks

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：