检索结果-内蒙古大学图书馆

Hierarchical adaptive cuckoo search algorithm for global optimization

Cluster computing 2025年第5期28卷 1-36页

作者： Ouyang, Chengtian Liu, Xin Zhu, Donglin Li, Yehong Mao, Jihong Zhou, Changjun Xue, Jiankai School of Information Engineering Jiangxi University of Science and Technology Ganzhou341000 China School of Computer Science and Technology Zhejiang Normal University Jinhua321004 China School of Computer Science and Technology Anhui University of Technology Anhui Ma’anshan243032 China School of Computing and Artificial Intelligence Southwestern University of Finance and Economics Chengdu611130 China

The Cuckoo Search algorithm (CS) still has some shortcomings, such as insufficient global search capability, challenges in exploring values beyond local optima, and underutilization of effective information within populations. In order to address these existing shortcomings and enhance the search performance of the Cuckoo Search algorithm, a proposed solution is the introduction of the Hierarchical Adaptive Cuckoo Search algorithm for Global Optimization (HAGCS). This algorithm integrates the concept of distribution entropy and a weighted fitness-value-distance-balanced ranking selection method. These additions aim to enhance information exchange within populations and strengthen the local search abilities of CS algorithms. Furthermore, the algorithm includes an adaptive global search strategy to improve global search capabilities, enhance population diversity, and aid in the discovery of global optimal solutions. Additionally, a hierarchical nest replacement strategy is introduced to refine position updates, reducing the risk of getting trapped in local extremes. The incorporation of a sawtooth linear function population adaptive strategy accelerates algorithm convergence, while a shifting population enhances the algorithm’s capacity to navigate suboptimal scenarios. The experiment assesses the impact of the proposed strategy on step size, compares it with variant algorithms and the TOP algorithm on the CEC 2022 test set, and applies the algorithm to three practical engineering optimization problems. The results demonstrate that HAGCS exhibits good practicality and feasibility. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Learning algorithms

来源：评论

学校读者我要写书评

暂无评论

Model-Heterogeneous Federated Graph Learning With Prototype Propagation

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年第3期6卷 676-689页

作者： Liu, Zhi Zhou, Hanlin He, Xiaohua Yuan, Haopeng Du, Jiaxin Wang, Mengmeng Shen, Guojiang Kong, Xiangjie Xia, Feng Zhejiang University of Technology College of Computer Science & Technology Hangzhou310023 China RMIT University School of Computing Technologies MelbourneVIC3000 Australia

Federated graph learning (FGL) enables clients to collaboratively train a robust graph neural network (GNN) while ensuring their private graph data never leaves the local. However, existing FGL frameworks require all clients to train the identical GNN model, which limits their real-world applicability. Although many model-heterogenous frameworks have been proposed for traditional nongraph federated learning settings, directly transferring them to the FGL setting typically results in suboptimal performance. To fill the gap, this article presents federated prototype propagation network (FedPPN), a lightweight FGL framework that supports clients to train fully customized models. FedPPN only transmits prototypes between clients and the server for knowledge sharing. The core idea is propagating global prototypes on each client's local graph, generating prototype-based node representations and predictions. The prototype-based prediction can then be ensembled with the prediction of local GNN, allowing clients to achieve accurate prediction. We evaluate our FedPPN on six benchmark datasets with different heterogeneous model setups. Experimental results show that our FedPPN outperforms advanced baselines in model accuracy without adding any trainable parameters on clients or the server. Besides, FedPPN's communication cost is significantly lower than methods that rely on model parameter exchange. © 2024 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Cross-graph Knowledge Exchange for Personalized Response Generation in Dialogue Systems

引用

IEEE Internet of Things Journal 2025年第10期12卷 13769-13789页

作者： Dong, Yuezhou Qin, Ke Ke, Pei Liang, Shuang Luo, Guangchun National Key Laboratory of Intelligent Collaborative Computing China University of Electronic Science and Technology of China School of Computer Science and Engineering Chengdu China University of Electronic Science and Technology of China National Key Laboratory of Intelligent Collaborative Computing Chengdu China University of Electronic Science and Technology of China School of Information and Software Engineering Chengdu China

Recent advancements in language models have greatly improved dialogue systems, but they still face challenges in generating personalized responses that are consistent with the user's persona and dialogue context. Existing approaches typically model dialogue context and persona information together in a unified manner, but they lack fine-grained differentiation between the two, leading to inaccurate user modeling. This misalignment hinders the ability of dialogue systems to produce personalized responses. In this work, we propose CKE (Cross-Graph Knowledge Exchange), a novel algorithm designed to enhance personalized response generation in dialogue systems. CKE constructs separate dialogue user graphs for each party in the dialogue, representing both their dialogue context and persona information. These graphs are then utilized to perform cross-graph structured knowledge aggregation, where the aggregation is under supervised by both its own and the other party's persona and dialogue context, providing richer, more accurate representations. Furthermore, CKE introduces a hybrid prompt template that combines both discrete and continuous elements, improving the language model's ability to leverage graph-structured information. The experimental results demonstrate that CKE significantly outperforms existing baseline methods in generating more coherent, contextually appropriate, and personalized responses in dialogue systems. © 2014 IEEE.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Identification and Classification of Medicinal Plants using Machine Learning Algorithms 3

Identification and Classification of Medicinal Plants using ...

引用

3rd International Conference on Electronics and Renewable Systems, ICEARS 2025

作者： Bhargavi, G. Jayaraman, Venkatesh Jabez, J. Deepa, R. Surendran, R. Srm Institute of Science and Technology Department of Computing Technologies Chennai India Chennai Institute of Technology Department of Computer Science and Engineering Chennai600069 India Sathyabama Institute of Science and Technology School of Computing Department of Computer Science and Engineering Chennai India Srm Institute of Science and Technology School of Computing Department of Computing Technologies Kattankulathur Tamilnadu Chennai603203 India Saveetha Institute of Medical and Technical Sciences Saveetha School of Engineering Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798331509675

The increaing significance of plant life and botanical expertise extends beyond mere visual appreciation. With the growing interest in sustainable living and alternative remedies, there is a pressing demand for easily accessible and credible information on plant categorization, particularly focusing on medicinal plants. This innovative project synergizes machine learning capabilities, botanical proficiency, and user-friendly technology to give users the insights required to harness the benefits of medicinal plants responsibly and sustainably. The platform facilitates precise categorization and a comprehensive understanding of the therapeutic attributes of plants. © 2025 IEEE.

关键词： Medicinal chemistry

来源：评论

学校读者我要写书评

暂无评论

Enhancing sarcasm detection through grasshopper optimization with deep learning based sentiment analysis on social media

引用

International Journal of Information Technology (Singapore) 2025年第3期17卷 1785-1791页

作者： Srinu, Nidamanuri Sivaraman, K. Sriram, M. School of Computing Department of Computer Science and Engineering Bharath Institute of Higher Education and Research Chennai 600073 India

Detecting sarcasm in social media presents challenges in natural language processing (NLP) due to the informal language, contextual complexities, and nuanced expression of sentiment. Integrating sentiment analysis (SA) with sarcasm detection enhances the understanding of text meaning. Deep learning (DL), utilizing neural networks to grasp lexical and contextual features, offers a method for sarcasm detection. However, current DL-based sarcasm detection methods often overlook sentiment semantics, a crucial aspect for improving detection outcomes. Therefore, this study develops a new sarcasm detection using grasshopper optimization algorithm with DL (SD-GOADL) technique. The SD-GOADL technique aims to explore the patterns that exist in social media data and detect sarcasm. To obtain this, the SD-GOADL technique undergoes data pre-processing and Glove based word embedding technique. Next, the classification of sarcasm takes place using deep belief network (DBN) system. For enhancing the detection results of the DBN approach, the SD-GOADL technique uses GOA for hyperparameter selection process. The stimulation outcome of the SD-GOADL technique is tested on a sarcasm dataset and the results highlight the significant performance of the SD-GOADL technique compared to recent models. © Bharati Vidyapeeth's Institute of computer Applications and Management 2024.

关键词： Deep learning Grasshopper optimization algorithm Natural language processing Sarcasm detection Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

NL2Code: A Hybrid NLP and Model-Driven Framework for Automated Code Generation from Natural Language and UML

NL2Code: A Hybrid NLP and Model-Driven Framework for Automat...

引用

2025 IEEE International Students' Conference on Electrical, Electronics and computer science, SCEECS 2025

作者： Nair, Reshma P Thushara, M.G. Amrita School of Computing Amrita Department of Computer Science & Engineering Kerala Vishwa Vidyapeetham Amritapuri Kollam India Amrita School of Computing Amrita Department of Computer Science & Applications Kerala Vishwa Vidyapeetham Amritapuri Kollam India

ISBN: (纸本)9798331529833

The process of converting natural language requirements and visual models into executable software code remains an ongoing challenge in software engineering. We developed an intelligent system that adopts Natural Language Processing (NLP) and Model-Driven Engineering (MDE) to generate code from natural language requirements and UML diagrams. It then applies advanced NLP techniques for natural language specification, such as tokenization, POS tagging, semantic analysis, and information extraction. This information can be transformed into computer language construct and design patterns with the help of MDE concepts. It generates code templates, skeleton code, and real code and incorporates key principles of software engineering and technological specializations. Case studies explain how the system can improve the software development lifecycle for a development company by automatically translating the natural language and the visual models into code. The system synthesized code with 92-95 % accuracy and 92-93 % completeness in three particular cases of use, proving the efficiency of the translation process of natural language requirements and UML diagrams into executable code. The present work enriches software engineering, NLP, as well as the integration of MDE, and code generation, reducing time consumption, manual efforts, and fast communication in the Software Development Process. © 2025 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Lower Bounds for Approximate (& Exact) k-Disjoint-Shortest-Paths 22nd

Lower Bounds for Approximate (& Exact) k-Disjoint-Shortest-...

引用

22nd International Workshop on Approximation and Online Algorithms, WAOA 2024

作者： Chitnis, Rajesh Thomas, Samuel Wirth, Anthony School of Computer Science The University of Birmingham Birmingham United Kingdom School of Computing and Information Systems The University of Melbourne Melbourne Australia School of Computer Science The University of Sydney Sydney Australia

ISBN: (纸本)9783031813955

Given a graph G=(V,E) and a set T={(si,ti):1≤i≤k}⊆V×V of k pairs, the k-Vertex-Disjoint-Paths (resp. k-Edge-Disjoint-Paths) problem asks to determine whether there exist k pairwise vertex-disjoint (resp. edge-disjoint) paths P1,P2,…,Pk in G such that, for each 1≤i≤k, Pi connects si to ti. Both the edge-disjoint and vertex-disjoint versions in undirected graphs are famously known to be FPT (parameterized by k) due to the Graph Minor Theory of Robertson and Seymour. Eilam-Tzoreff [DAM ‘98] introduced a variant, known as the k-Disjoint-Shortest-Paths problem, where each path is further required to be a shortest path connecting its pair. They showed that the k-Disjoint-Shortest-Paths problem is NP-complete on both directed and undirected graphs;this holds even if the graphs are planar and have unit edge lengths. We focus on four versions of the problem, corresponding to considering edge/vertex disjointness, and to considering directed/undirected graphs. Building on the reduction of Chitnis [SIDMA ’23] for k-Edge-Disjoint-Paths on planar DAGs, we obtain the following inapproximability lower bound for each of the four versions of k-Disjoint-Shortest-Paths on n-vertex graphs:Under the gap version of the Exponential Time Hypothesis (Gap-ETH), there exists a constant δ>0 such that for any constant 0δ·k time. Under the gap version of the Exponential Time Hypothesis (Gap-ETH), there exists a constant δ>0 such that for any constant 0δ·k time. We provide a single, unified framework to obtain lower bounds for each of the four versions of k-Disjoint-Shortest-Paths. We are able to further strengthen our results by restricting the structure of the input graphs in the lower bound constructions as follows:Directed: The inapproximability lower bound for edge-disjoint (resp. vertex-disjoint) paths holds even if the input graph is a planar (resp. 1-planar) DAG with max in-degree and max out-degree at most ***: The inapproximability lower bound for edge-disjoint (resp. vertex-dis

关键词： Undirected graphs

来源：评论

学校读者我要写书评

暂无评论

Integration of Traffic Shaping with Time-Sensitive Networking for Effective Traffic Management 3

Integration of Traffic Shaping with Time-Sensitive Networkin...

引用

3rd International Conference on Intelligent Data Communication Technologies and Internet of Things, IDCIoT 2025

作者： Penumada, Archana Reddy Gundapaneni, Divya Jyothi Chennupati, Akshitha Rajagopal, Shinu M. Department of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Bengaluru India

ISBN: (纸本)9798331527549

Time-Sensitive Networking (TSN) has recently become an important standard to support time-sensitive reliable low-latency data transmission in industrial and real-time applications. This work explores the use of different traffic shaping algorithms, namely, credit-based shaping (CBS) and time-aware shaping (TAS) alongside frame preemption—a TSN instance which gives precedence to high-importance frames by means of lower-priority traffic interruption—to extend TSN’s traffic management capabilities. By controlling data flow and allowing for high priority packets to pass through unimpeded, traffic shaping algorithms serve an essential function in making the network more efficient and reducing congestion rates. The work compares these algorithms in isolation and also in combination over a network model, comprising of four switches and three end users for each switch, using the OMNeT++ simulation environment. This study aims to contribute insights into optimal shaping techniques for TSN, enhancing its suitability for critical network applications. © 2025 IEEE.

关键词： Traffic congestion

来源：评论

学校读者我要写书评

暂无评论

Enhancing Underwater Image Captioning Using Transformer Models and Augmented Terrestrial Datasets

Enhancing Underwater Image Captioning Using Transformer Mode...

引用

2025 International Conference on Pervasive Computational Technologies, ICPCT 2025

作者： Gamidi, Rohan Hemasri, M. Muppala, Tejaswi Vinitha Chowdary, A. Aiswariya Milan, K. Palaniswamy, Suja Amrita School of Computing Department of Computer Science and Engineering Amrita Vishwa Vidyapeetham Bengaluru India

ISBN: (纸本)9798331508685

Underwater environments present unique challenges for image analysis due to light attenuation, scattering, and color distortion, which significantly degrade image quality. These facts make simple operations such as image captioning, which is crucial in marine sciences, search, and conservation, challenging. Current datasets such as Flickr8K contain mostly terrestrial images and do not possess features distinctive to underwater environments therefore direct training of models is not possible in this context. To cover this gap, the current work combines a Transformer-based image captioning model with an enhanced underwater-specific image augmentation set. These augmentations mimic real-life underwater consequences such as RGB attenuation, Gaussian blur, muddy particles, noise, and gradient-based illumination distortions. The implemented model, learned and recursively tuned on the Flickr8k dataset, proves the ability to generate reasonable captions for underwater images even though the dataset belongs to Flickr8k providing a terrestrial environment. This is proven by the enhanced BLEU scores as well as from the results of the experiment which trained the translation model in less than ten percent of the dataset to show its flexibility on underwater scenarios but without necessitating the use of an underwater dataset. © 2025 IEEE.

关键词： Underwater imaging

来源：评论

学校读者我要写书评

暂无评论

Comprehensive Web Analysis of Automobile Websites: Performance, SEO, and Accessibility Insights

Comprehensive Web Analysis of Automobile Websites: Performan...

引用

2025 IEEE International Conference on Computational, Communication and Information Technology, ICCCIT 2025

作者： Busetty, Abhishek Balam, Ruchith Chillakuru, Hari Radha, R. Saranya Devi, B. Amrita School of Computing Amrita Vishwa Vidyapeetham Department of Computer Science and Engineering Bengaluru India

ISBN: (纸本)9798331512965

The current research work proposes a difficultydriven comparison of three most visited automobile websites namely the Automobile Site1, Automobile Site2, Automobile Site3 on parameters including performance, Search engine optimization (SEO), accessibility and traffic. The research involves tool such as Yellow Labs, Google Page Speed Insight and Semrush with regard to the website's weight, loading time, user's interactions and search engine optimization. The findings reveal distinct strengths across the sites: From the three Automobile Site1 is the most optimized and accessibly friendly site, the factors of search engine friendliness is as follows. Looking at the traffic and users of Automobile Site2, it comfortably outclasses the other sites in terms of page views as well as the audience loyalty. At the same time, Automobile Site3 has relatively low first load time which indicates fast website browsing while it is still relatively weak in terms of its SEO and the usability of the site has to be improved in terms of theoretical potential. In conclusion, the realized comparison exposes the strengths and weaknesses of each site and is informative to future developments of the sites with an aim at better user interaction levels and competitive advantage. © 2025 IEEE.

关键词： Automobiles

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：