检索结果-内蒙古大学图书馆

2012 Cross Language Evaluation Forum Conference, CLEF 2012

作者： Chien, Been-Chian Chen, Guan-Bin Gaou, Li-Ji Ku, Chia-Wei Huang, Rong-Sing Wang, Siao-En Knowledge Information and Database System Laboratory Department of Computer Science and Information Engineering National University of Tainan Tainan Taiwan

The task of visual concept detection, annotation, and retrieval using Flickr photos at ImageCLEF 2012 was organized as two subtasks: concept annotation and concept retrieval. In this paper, we present the effort of KIDS lab for the two subtasks. The proposed approaches combine various visual and textual features, dimension reduction methods, the random forest classification models, and the semi-supervised learning strategy. For the concept annotation subtask, the annotation results show that combination of tags and visual features outperforms visual-only features while using the same classification model. The results also show that semi-supervised learning is not superior to supervised learning in this subtask. Further, it does not seem able to gain more advantage on F-measure when more different visual features were used. For the concept retrieval task, the results illustrate that the textual features contain much richer informatics than visual features in general retrieved concepts.

关键词： Image retrieval

来源：评论

学校读者我要写书评

暂无评论

A Novel Approach for Sign Language Video Generation Using Deep Networks 2

A Novel Approach for Sign Language Video Generation Using De...

引用

2nd IEEE International Conference on Data science and Network Security, ICDSNS 2024

作者： Kumar, Sachin Deepa, B. Kavitha, T. Tamilselvi, M. Sathiyapriya, V. Natarajan, B. Amrita School of Computing Amrita Vishwa Vidyapeetham Department of Computer Science and Engineering Chennai India Srmist Ramapuram Campus Department of Electroncis and Communication Engineering Ramapuram India MA.M College of Engineering and Technology Department of Computer Science and Engineering Trichy India Roever Engineering College Department of Computer Science and Engineering Perambalur India Knowledge Institute of Technology Department of Computer Science and Engineering Salem India School of Computer Science and Engineering Vellore Institue of Technology Chennai Campus Chennai India

ISBN: (纸本)9798350373110

Sign language enhances the communication capabilities of the deaf-mute community, allowing for a deeper understanding of their needs and emotions. These languages are highly structured and visual, using gestures and various upper body movements such as those of the hands, face, eyes, and gaze. Researchers face numerous challenges in recognizing and translating the diverse variations in sign movements, which requires specialized expertise in computer vision and artificial intelligence. Sign language recognition and translation research has garnered global attention. This research work introduces a novel methodology for generating sign gesture videos from text inputs by integrating various intelligent techniques. The proposed model employs an enhanced generative adversarial network (GAN) to create sign videos from input sentences. Experiments with the proposed VideoGAN model using diverse sign language datasets from multiple countries have demonstrated its effectiveness. The research outcomes highlight its contribution to high-quality video production, with improved evaluation metrics underscoring the model's superior performance. © 2024 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Design metamodels for domain-specific modelling methods using conceptual structures 11

Design metamodels for domain-specific modelling methods usin...

引用

2018 Doctoral Consortium Papers presented at the 11th IFIP WG 8.1 Working Conference on the Practice of Enterprise Modelling, PoEM-DC 2018

作者： Utz, Wilfrid Department of Knowledge Engineering Faculty of Computer Science University of Vienna Währingerstr. 29 Vienna1090 Austria

Enterprises nowadays operate in fast-changing environments and need to adapt dynamically to new circumstances. This impacts the way how enterprise information systems are analysed, designed and implemented. Conceptual modelling methods experience a constant evolution nowadays. The methods are re-constructed continuously to reflect the changing application/industrial domain. It is therefore required to provide additional design support to realise domain specific modelling methods and more precisely their underlying metamodel. The goal of this research project is to enable the knowledge representation of a metamodel to support its design process. Building upon the assumption that conceptual structures within a metamodel exist, a knowledge representation framework is proposed using conceptual graphs as the mathematical baseline. The proposal will be evaluated in a laboratory setting, applied on metamodel design challenges observed. Copyright 2018 for this paper by its authors.

关键词： knowledge representation

来源：评论

学校读者我要写书评

暂无评论

Optimizing SPARQL queries over disparate RDF data sources through distributed semi-joins

Optimizing SPARQL queries over disparate RDF data sources th...

引用

7th International Semantic Web Conference, ISWC 2008

作者： Zemánek, Jan Schenk, Simon Svátek, Vojtěch Department of Information and Knowledge Engineering University of Economics Prague Czech Republic Institute for Computer Science University of Koblenz Germany

With the ever-increasing amount of data on the Web available at SPARQL endpoints [1] the need for an integrated and transparent way of accessing the data has arisen. It is highly desirable to have a way of asking SPARQL queries that make use of data residing in disparate data sources served by multiple SPARQL endpoints. We aim at providing such a capability and thus enabling an integrated way of querying the whole Semantic Web at a time.

关键词： Resource Description Framework (RDF)

来源：评论

学校读者我要写书评

暂无评论

Lay Summarization of Biomedical Documents with Discourse Structure-Based Prompt Tuning 28th

Lay Summarization of Biomedical Documents with Discourse S...

引用

28th International Conference on Technologies and Applications of Artificial Intelligence, TAAI 2023

作者： Wu, Yu-Hsuan Chiu, Chi-Min Kao, Hung-Yu Intelligent Knowledge Management Lab Department of Computer Science and Information Engineering National Cheng Kung University Tainan Taiwan

ISBN: (纸本)9789819717101

Transforming complex biomedical texts into accessible lay summaries is a critical endeavor in Natural Language Generation (NLG). This study addresses the challenges associated with this task by employing a multi-aspect approach. Firstly, we undertake a comprehensive analysis of discourse structures within a diverse range of biomedical datasets and clarify underlying patterns and structures. Secondly, we designed the power of prompting strategies to integrate training on these varied datasets, thereby reducing the noise introduced by their diversity. This twofold strategy fine-tunes the model’s training and enriches it with the ability to generate coherent and simplified lay summaries of biomedical content. Our experimental results clearly demonstrate the effectiveness of our study, underscoring its potential to make complex medical information more accessible to general readers. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

R-AT: Regularized Adversarial Training for Natural Language Understanding

R-AT: Regularized Adversarial Training for Natural Language ...

引用

2022 Findings of the Association for Computational Linguistics: EMNLP 2022

作者： Ni, Shiwen Li, Jiawen Kao, Hung-Yu Intelligent Knowledge Management Lab Department of Computer Science and Information Engineering National Cheng Kung University Tainan Taiwan

Currently, adversarial training has become a popular and powerful regularization method in the natural language domain. In this paper, we Regularized Adversarial Training (R-AT) via dropout, which forces the output probability distributions of different sub-models generated by dropout to be consistent under the same adversarial samples. Specifically, we generate adversarial samples by perturbing the word embeddings. For each adversarial sample fed to the model, R-AT minimizes both the adversarial risk and the bidirectional KL-divergence between the adversarial output distributions of two sub-models sampled by dropout. Through extensive experiments on 13 public natural language understanding datasets, we found that RAT has improvements for many models (e.g., rnn-based, cnn-based, and transformer-based models). For the GLUE benchmark, when RAT is only applied to the fine-tuning stage, it is able to improve the overall test score of the BERT-base model from 78.3 to 79.6 and the RoBERTa-large model from 88.1 to 88.6. Theoretical analysis reveals that R-AT has potential gradient regularization during the training process. Furthermore, R-AT can reduce the inconsistency between training and testing of models with dropout. © 2022 Association for Computational Linguistics.

关键词： Rats

来源：评论

学校读者我要写书评

暂无评论

Billiards Optimization with Modified Deep Learning for Fault Detection in Wireless Sensor Network

引用

computer Systems science & engineering 2023年第11期47卷 1651-1664页

作者： Yousif Sufyan Jghef Mohammed Jasim Mohammed Jasim Subhi R.M.Zeebaree Rizgar R.Zebari Department of Computer Engineering College of EngineeringKnowledge UniversityErbil44001Iraq Engineering College Al-Kitab UniversityKirkukIraq Energy Eng.Department Technical College of EngineeringDuhok Polytechnic UniversityDuhokIraq Computer Science Department College of ScienceNawroz UniversityDuhokIraq

Wireless Sensor Networks(WSNs)gather data in physical environments,which is some *** ubiquitous sensors face several challenges responsible for corrupting them(mostly sensor failure and intrusions in external agents).WSNs were disposed to error,and effectual fault detection techniques are utilized for detecting faults from WSNs in a timely *** learning(ML)was extremely utilized for detecting faults in ***,this study proposes a billiards optimization algorithm with modified deep learning for fault detection(BIOMDL-FD)in *** BIOMDLFD technique mainly concentrates on identifying sensor faults to enhance network *** do so,the presented BIOMDL-FD technique uses the attention-based bidirectional long short-term memory(ABLSTM)method for fault *** the ABLSTM model,the attention mechanism enables us to learn the relationships between the inputs and modify the probability to give more attention to essential *** the same time,the BIO algorithm is employed for optimal hyperparameter tuning of the ABLSTM model,which is stimulated by billiard games,showing the novelty of the *** analyses are made to affirm the enhanced fault detection outcomes of the BIOMDL-FD *** simulation results demonstrate the improvement of the BIOMDL-FD technique over other models with a maximum classification accuracy of 99.37%.

关键词： Wireless sensor network fault detection reliability deep learning metaheuristics

来源：评论

学校读者我要写书评

暂无评论

A Comprehensive Study on Advancements in Text Mining and Natural Language Processing 2

A Comprehensive Study on Advancements in Text Mining and Nat...

引用

2nd IEEE International Conference on Emerging Research in Computational science, ICERCS 2024

作者： Mahire, Sarthak Balaji Saravanan, U.K. Murali, M. Prabakaran, N. Vellore Institute of Technology School of Computer Science and Engineering Vellore632014 India Knowledge Institute of Technology Department of Computer Science and Engineering Salem India Sona College of Technology Department of Information Technology Salem India

ISBN: (纸本)9798331534950

Text mining and Natural Language Processing (NLP) have witnessed significant advancements in recent years, driven by the increasing availability of unstructured data and the development of sophisticated machine learning models. This review explores the evolution of text mining and NLP, highlighting the transition from rule-based systems to modern deep learning approaches. This paper reviews various techniques and its impact across various domains. Challenges such as data bias and model interpretability still remain. The paper also discusses future directions, emphasizing the need for fair, interpretable and sustainable NLP systems. This paper aims to provide insights into the current state of text mining and NLP, the challenges faced and potential pathways for future research. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Monte Carlo Tree Search in simultaneous move games with applications to Goofspiel 1

引用

Workshop on computer Games, CGW 2013 held in Conjunction with the 23rd International Conference on Artificial Intelligence, IJCAI 2013

作者： Lanctot, Marc Lisý, Viliam Winands, Mark H.M. Department of Knowledge Engineering Maastricht University Maastricht Netherlands Department of Computer Science Czech Technical University in Prague Praha Czech Republic

ISBN: (数字)9783319054285

ISBN: (纸本)9783319054278

Monte Carlo Tree Search (MCTS) has become a widely popular sampled-based search algorithm for two-player games with perfect information. When actions are chosen simultaneously, players may need to mix between their strategies. In this paper, we discuss the adaptation of MCTS to simultaneous move games. We introduce a new algorithm, Online Outcome Sampling (OOS), that approaches a Nash equilibrium strategy over time. We compare both head-to-head performance and exploitability of several MCTS variants in Goofspiel. We show that regret matching and OOS perform best and that all variants produce less exploitable strategies than UCT. © Springer International Publishing Switzerland 2014.

关键词： Monte Carlo methods

来源：评论

学校读者我要写书评

暂无评论

Solar Radiation Prediction Using the Random Forest Regression Algorithm 4th

Solar Radiation Prediction Using the Random Forest Regressio...

引用

4th EAI International Conference on Big Data Innovation for Sustainable Cognitive Computing, BDCC 2021

作者： Thota, Sandeep Kumar Reddy Mala, C. Chandamuri, Praneetha Nooka, Chandana Department of Computer Science and Engineering Rajiv Gandhi University of Knowledge Technologies- RK Valley Campus Andhra Pradesh Kadapa India Department of Computer Science and Engineering National Institute of Technology Tamil Nadu Tiruchirappalli India

ISBN: (纸本)9783031076534

Solar radiation data is critical for solar energy research and development. It is extremely valuable for design and simulation of solar thermal technologies and solar photovoltaic applications because it offers essential data about the energy that impacts the earth. Because conventional energy sources are rapidly decreasing, sustainable energy sources such as solar, biomass, and wind are growing more prominent. Solar energy is the most generally acknowledged renewable energy source since it is abundant all around the planet. As a result, this project examines the model and methodologies utilized to forecast solar radiation. Random forest algorithm is used in solar radiation prediction;this algorithm combines the benefits of both techniques utilized, resulting in faster calculation and convergence to optimal results. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Forecasting

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：