检索结果-内蒙古大学图书馆

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Chen, Xinping Ke, Xiao Guo, Wenzhong Fujian Provincial Key Laboratory of Networking Computing and Intelligent Information Processing College of Computer and Data Science Fuzhou University Fuzhou350116 China Engineering Research Center of Big Data Intelligence Ministry of Education Fuzhou350116 China

Few-shot font generation (FFG) aims to learn the target style from a limited number of reference glyphs and generate the remaining glyphs in the target font. Previous works focus on disentangling the content and style features of glyphs, combining the content features of the source glyph with the style features of the reference glyph to generate new glyphs. However, the disentanglement is challenging due to the complexity of glyphs, often resulting in glyphs that are influenced by the style of the source glyph and prone to artifacts. We propose IF-Font, a novel paradigm which incorporates Ideographic Description Sequence (IDS) instead of the source glyph to control the semantics of generated glyphs. To achieve this, we quantize the reference glyphs into tokens, and model the token distribution of target glyphs using corresponding IDS and reference tokens. The proposed method excels in synthesizing glyphs with neat and correct strokes, and enables the creation of new glyphs based on provided IDS. Extensive experiments demonstrate that our method greatly outperforms state-of-the-art methods in both one-shot and few-shot settings, particularly when the target styles differ significantly from the training font styles. The code is available at https://***/Stareven233/IF-Font. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Influence-Aware Group Recommendation for Social Media Propagation 24

Influence-Aware Group Recommendation for Social Media Propag...

引用

24th IEEE International Conference on data Mining, ICDM 2024

作者： He, Chengkun Zhou, Xiangmin Wang, Chen Cao, Longbing Shao, Jie Tari, Zahir School of Computing Technologies RMIT University Melbourne Australia Data61 CSIRO Australia Macquarie University Australia University of Electronic Science and Technology of China China

ISBN: (纸本)9798331506681

Group recommendation over social media streams has attracted attention due to its wide applications such as e-commerce, entertainment and online news broadcasting. However, existing stream group recommendation techniques ignore the influence of user groups, which are not effective for item propagation over social networks. To address this problem, we propose a framework for Influence-aware Group Recommendation (IGR) over high-speed social streams. Specifically, we first propose a novel GroupGCN model to capture the dynamics of user attributes and interactions which maps groups and items to their embeddings. A Temporal GroupGCN-RNN-Autoencoder (TGGCN-RA) model is designed to extend GroupGCN for sequence-based tasks, enabling the prediction of group interests over time. Then, we adopt an Independent Cascade (IC) model to predict the influence propagation of social items over user groups. Extensive experiments prove the high effectiveness and efficiency of IGR. © 2024 IEEE.

关键词： Social networking (online)

来源：评论

学校读者我要写书评

暂无评论

Time Matters: An End-to-End Solution for Temporal Claim Verification

Time Matters: An End-to-End Solution for Temporal Claim Veri...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Barik, Anab Maulana Hsu, Wynne Lee, Mong Li School of Computing National University of Singapore Singapore Institute of Data Science National University of Singapore Singapore Centre for Trusted Internet & Community National University of Singapore Singapore

ISBN: (纸本)9798891761667

Automated claim verification plays an essential role in fostering trust in the digital space. Temporal claim verification brings new challenges where cues of the temporal information need to be extracted, and temporal reasoning involving various temporal aspects of the text must be applied. In this work, we describe an end-to-end solution for temporal claim verification that considers the temporal information in claims to obtain relevant evidence sentences and harnesses the power of a large language model for temporal reasoning. We curate two datasets comprising a diverse range of temporal claims to learn time-sensitive representations that encapsulate not only the semantic relationships among the events, but also their chronological proximity. Experiment results demonstrate that the proposed approach significantly enhances the accuracy of temporal claim verification, thereby advancing current state-of-the-art in automated claim verification. © 2024 Association for Computational Linguistics.

关键词： Spatio-temporal data

来源：评论

学校读者我要写书评

暂无评论

A Comparative Analysis of Cloud Load Balancing Algorithms Using CloudSim Simulations 1

引用

International Conference on Advanced Engineering, Technology and Applications, ICAETA 2024

作者： Abdullah, Fasih Razi, Muhammad Faraz Ud Din Aleem, Muhammad Jamil, Akhtar Hameed, Alaa Ali National University of Computer and Emerging Sciences Islamabad Pakistan Istinye University Istanbul Turkey School of Computing Mathematics and Data Science Coventry University Coventry United Kingdom

ISBN: (数字)9783031709241

ISBN: (纸本)9783031709234

Cloud computing provides a set of services that allows its users to lease digital infrastructure as software. This abstraction allows users to focus on development without managing low-level aspects and associated challenges. The infrastructure comprises heterogeneous VMs with varying performance parameters such as MIPS and bandwidth. Thus, user cloud jobs must be executed with great care and consideration of load balancing. If scheduling is not done efficiently, the latency that users experience increases, resulting in lower customer satisfaction and decreased revenue. Since scheduling problems are NP-Hard, there is a need for continuous research into improving existing algorithms. Ideally, the scheduling of cloud jobs should be fair and well-balanced, with minimal waiting time and minimal overhead. This study focuses on the comparison of nine such state-of-the-art, batch-based, non-preemptive scheduling schemes. All algorithms have been implemented in Java with the help of the CloudSim simulator and tested across seven performance metrics. We provide an in-depth performance analysis of Random Selection, Round Robin, Minimum Completion Time, Suffrage, Min-Min, Max-Min, Weighted Mean Time-Min, Resource Aware Load Balancing Algorithm, and QoS Guided Weighted Mean Time Min Heuristic across asymptotic time and space complexity, make-span, throughput, average resource utilization rate (ARUR), load balancing level, and batch scheduling time. Comparison of the mechanisms and their performance shall aid future work in comparing and contrasting the concepts deployed in existing schemes. It was observed that Suffrage and QWMTM performed the best across most of the performance metrics. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Customer satisfaction

来源：评论

学校读者我要写书评

暂无评论

Mixed-Based Approach for an Intelligent E-Commerce Recommender System 16

Mixed-Based Approach for an Intelligent E-Commerce Recommend...

引用

16th International Conference on Electronics, Computers and Artificial Intelligence, ECAI 2024

作者： Jusoh, Shaidah Ng, Ger Qin Xiamen University Malaysia School of Eletrical Engineering and Artificial Intelligence Selagor Sepang Malaysia Xiamen University Malaysia School of Computing and Data Science Selangor Sepang Malaysia

ISBN: (纸本)9798350371154

Content-based recommender systems and collaborative filtering recommender systems might be useful in recommending items that might be preferred by users based on their preferences from the past, these systems rely on users' preferences, items' ratings, and items' descriptions. Thus, these systems might be less useful for items for which users' preferences and ratings cannot be easily collected. On the other hand, a constraint-based recommender system can suggest items that fulfil users' requirements (or constraints) by relying on users' requirements and knowledge about the domain of the items for giving a recommendation. In the case where, no item can be found, the system will proceed with the constraints of a relaxation phase to search for possible modifications in the constraints, give advice to the user, and enable them to edit their requirements to obtain more results. In this study, we propose a new approach called a mixed-based approach. The approach is based on an integration of soft and hard relaxation-based approaches. The proposed approach is implemented as a notebook recommender system. We conducted experiments on the proposed approach using collected data from e-commerce websites. The results also show that the proposed mixed-based approach is able to control constraints, and gives better performance in giving recommendations to users. © 2024 IEEE.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

A Graph with Adaptive AdjacencyMatrix for Relation Extraction

引用

Computers, Materials & Continua 2024年第9期80卷 4129-4147页

作者： Run Yang YanpingChen Jiaxin Yan Yongbin Qin Engineering Research Center of Text Computing&Cognitive Intelligence Ministry of EducationGuizhou UniversityGuiyang550025China State Key Laboratory of Public Big Data Guizhou UniversityGuiyang550025China College of Computer Science and Technology Guizhou UniversityGuiyang550025China

The relation is a semantic expression relevant to two named entities in a *** a sentence usually contains several named entities,it is essential to learn a structured sentence representation that encodes dependency information specific to the two named *** related work,graph convolutional neural networks are widely adopted to learn semantic dependencies,where a dependency tree initializes the adjacency ***,this approach has two main ***,parsing a sentence heavily relies on external toolkits,which can be ***,the dependency tree only encodes the syntactical structure of a sentence,which may not align with the relational semantic *** this paper,we propose an automatic graph learningmethod to autonomously learn a sentence’s structural *** of using a fixed adjacency matrix initialized by a dependency tree,we introduce an Adaptive Adjacency Matrix to encode the semantic dependency between *** elements of thismatrix are dynamically learned during the training process and optimized by task-relevant learning objectives,enabling the construction of task-relevant semantic dependencies within a *** model demonstrates superior performance on the TACRED and SemEval 2010 datasets,surpassing previous works by 1.3%and 0.8%,*** experimental results show that our model excels in the relation extraction task,outperforming prior models.

关键词： Relation extraction graph convolutional neural network adaptive adjacency matrix

来源：评论

学校读者我要写书评

暂无评论

FedCal: Achieving Local and Global Calibration in Federated Learning via Aggregated Parameterized Scaler 41

FedCal: Achieving Local and Global Calibration in Federated ...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Peng, Hongyi Yu, Han Tang, Xiaoli Li, Xiaoxiao College of Computing and Data Science Nanyang Technological University Singapore Department of Electrical and Computer Engineering The University of British Columbia VancouverBC Canada Vector Institute Canada

Federated learning (FL) enables collaborative machine learning across distributed data owners. However, this approach poses a significant challenge for model calibration due to data heterogeneity. While prior work focused on improving accuracy for non-iid data, calibration remains under-explored. This study reveals existing FL aggregation approaches lead to sub-optimal calibration, and theoretical analysis shows despite constraining variance in clients' label distributions, global calibration error is still asymptotically lower bounded. To address this, we propose a novel Federated Calibration (FedCal) approach, emphasizing both local and global calibration. It leverages client-specific scalers for local calibration to effectively correct output misalignment without sacrificing prediction accuracy. These scalers are then aggregated via weight averaging to generate a global scaler, minimizing the global calibration error. Extensive experiments demonstrate that FedCal significantly outperforms the best-performing baseline, reducing global calibration error by 47.66% on average. Copyright 2024 by the author(s)

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Performance Evaluation of Cardiovascular Disease Prediction Using Supervised Machine Learning Algorithms 1

Performance Evaluation of Cardiovascular Disease Prediction ...

引用

1st International Conference on Innovations in Communications, Electrical and Computer Engineering, ICICEC 2024

作者： Archana, R. Sujatha, V. Abinaya, M. Vijay Anand, P. Ramya, M. Priya, S. SRM Institute of Science and Technology School of Computing Department of Networking and Communications Kattankulathur Chennai India Cmr Institute of Technology Department of Artificial Intelligence and Data Science Bengaluru560037 India SRM Institute of Science and Technology Department of Computing Technologies Kattankulathur Chennai India Vel Tech Rangarajan Dr.Sagunthala R and D Institute of Science and Technology School of Computing Department of CSE[Cyber Security Avadi Chennai India

ISBN: (数字)9798350376517

ISBN: (纸本)9798350376517

Cardiovascular disease has emerged as a very critical and intricate condition on a global scale. Healthcare practitioners have long faced substantial challenges in predicting and identifying heart disease. Medical and other healthcare facilities provide costly treatments and surgical procedures for cardiovascular conditions. People throughout the globe would benefit from early heart disease prediction since it would allow them to take action before the issue worsens. These days, heart disease is a major concern, and the primary causes of this illness include things like smoking, excessive alcohol use, and not getting enough exercise. Based on the extensive dataset that the healthcare industry has produced over time, machine learning has proven effective in decisions and forecasts. Logistic regression, Decision- Trees, Naive-Bayes, Random-Forest, and SVM are among the supervised machine learning algorithms that may anticipate the occurrence of heart illness. A synopsis of the algorithm's performance assessment is also included. A comparison of five models showed that the Decision Tree and Random Forest models were better at predicting heart disease than the others. Also, these models were very accurate, with a 98.5% success rate. © 2024 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Adjusting Education Amid Emergency—A University of Technology Case 1

引用

8th International Conference on Information and Communication Technology for Intelligent Systems, ICTIS 2024

作者： Moletsane, Ramadile Liebenberg, Janet Smit, Imelda Vaal University of Technology Private Bag x021 Vanderbijlpark South Africa Unit for Data Science and Computing North West University Potchefstroom South Africa

ISBN: (数字)9789819758104

ISBN: (纸本)9789819758098

The COVID-19 pandemic significantly affected different parts of society, and higher education is no special case. Schools, colleges, and universities overall have been confronted by extraordinary difficulties, requiring quick and creative reactions to guarantee the congruity of instruction while focusing on the well-being and security of students, personnel, and staff. In this study, the supporting strategy is a subjective contextual investigation in the form of a qualitative case study where information was assembled utilizing semi-structured interviews. The participants were identified through purposeful sampling. Content analysis was utilized to generate the study's themes. The study is expected to contribute to the literature by investigating the effect of e-learning on students during the COVID-19 pandemic, featuring its advantages and the limits set. The COVID-19 pandemic constrained educational instruction and forced educators to embrace e-learning as a fundamental device to guarantee instructive congruity. Regardless of its difficulties, e-learning showed its capability to cultivate development, adaptability, and commitment to teaching and learning. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Topological data Analysis in Natural Language Processing – A Tutorial 36th

Topological Data Analysis in Natural Language Processing – ...

引用

36th International Florida Artificial Intelligence Research Society Conference, FLAIRS-36 2023

作者： Zadrozny, Wlodek College of Computing University of North Carolina Charlotte United States School of Data Science University of North Carolina Charlotte United States

Topological data Analysis (TDA) introduces methods that capture the underlying structure of shapes in data. Within the last two decades, TDA has been mostly examined in unsupervised machine learning tasks. TDA has been often considered an alternative to the conventional algorithms due to its capability to deal with highdimensional data. in different tasks including but not limited to clustering, This tutorial will focus on applications of topological data analysis to text data. © 2023 by the authors. All rights reserved.

关键词： Natural language processing systems

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：