检索结果-内蒙古大学图书馆

Efficient Parameterization for Knowledge Graph Embedding Using Hierarchical Attention Network

Computers, Materials & Continua 2025年第3期82卷 4287-4300页

作者： Zhen-Yu Chen Feng-Chi Liu Xin Wang Cheng-Hsiung Lee Ching-Sheng Lin Master Program of Digital Innovation Tunghai UniversityTaichung40704Taiwan Department of Statistics Feng Chia UniversityTaichung40724Taiwan College of Integrated Health Sciences and the AI Plus Institute The University at AlbanyState University of New York(SUNY)AlbanyNY 12222USA

In the domain of knowledge graph embedding,conventional approaches typically transform entities and relations into continuous vector ***,parameter efficiency becomes increasingly crucial when dealing with large-scale knowledge graphs that contain vast numbers of entities and *** particular,resource-intensive embeddings often lead to increased computational costs,and may limit scalability and adaptability in practical environ-ments,such as in low-resource settings or real-world *** paper explores an approach to knowledge graph representation learning that leverages small,reserved entities and relation sets for parameter-efficient *** introduce a hierarchical attention network designed to refine and maximize the representational quality of embeddings by selectively focusing on these reserved sets,thereby reducing model *** assessments validate that our model achieves high performance on the benchmark dataset with fewer parameters and smaller embedding *** ablation studies further highlight the impact and contribution of each component in the proposed hierarchical attention structure.

关键词： Knowledge graph embedding parameter efficiency representation learning reserved entity and relation sets hierarchical attention network

来源：评论

学校读者我要写书评

暂无评论

Three generations of epigenetic clocks in mediating the adverse effect of smoking on metabolic health

引用

Epigenomics 2025年第7期17卷 453-461页

作者： Yeh, Chen-Yu Lin, Wan-Yu Institute of Epidemiology and Preventive Medicine College of Public Health National Taiwan University Taipei Taiwan Institute of Health Data Analytics and Statistics College of Public Health National Taiwan University Taipei Taiwan Master of Public Health Program College of Public Health National Taiwan University Taipei Taiwan

Aims: Metabolic syndrome (MetS) is a composite disorder that includes abdominal obesity, impaired glucose levels, high blood pressure, and dyslipidemia. Smoking can alter epigenetic profiles and is a critical modifiable risk factor for MetS. We aim to explore the epigenetic age acceleration (EAA) that can mainly deliver smoking influences on metabolic health. Methods: We conducted a mediation analysis of 2,474 individuals with data in the Taiwan Biobank. Current and former smoking and the respective pack-years were included as four exposure factors. Seven markers of DNA methylation (DNAm) covering three generations of epigenetic clocks were included as mediators. Seven metabolic outcomes included MetS status (yes vs. no) and six related traits. Results: GrimEAA and DunedinPACE mediated the associations of the four smoking factors with MetS, fasting glucose, triglyceride, and high-density lipoprotein cholesterol levels (false discovery rate < 0.05). GrimEAA and DunedinPACE respectively mediated 48.2% and 24.2% of current smoking’s effect on MetS and 60.9% and 26.1% of current smoking pack-year’s effect on MetS risk. The DNAm plasminogen activator inhibitor 1 level mediated the adverse effects of current smoking status and pack-years on all seven metabolic outcomes. Conclusion: The GrimEAA-mediated proportions were approximately two times greater than the DunedinPACE-mediated proportions. © 2025 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.

关键词： Biological age DNA methylation epigenetic age acceleration insulin resistance Taiwan Biobank

来源：评论

学校读者我要写书评

暂无评论

Uncovering Self-Similar Patterns in Infectious Disease Transmission Trees: Development of a Bayesian Connectivity Algorithm

SSRN

引用

SSRN 2025年

作者： Wang, Cheng-Hua Wen, Tzai-Hung Master Program in Statistics National Taiwan University No. 1 Sec. 4 Roosevelt Rd. Taipei Taiwan Department of Geography National Taiwan University No. 1 Sec. 4 Roosevelt Rd. Taipei City Taiwan

Epidemic dynamics are effectively modeled using tree structures that represent the transmission pathways of infectious diseases. Analyzing these transmission trees offers key insights into the mechanisms driving power-law growth, especially in the early stages of an outbreak. However, the degree of self-similarity within these trees remains insufficiently understood. This study introduces a Bayesian Connectivity Algorithm (BCA) designed to detect Conditional Infectious Connectivity Structures (CICSs) that exhibit self-similar properties. Simulation results demonstrate that the BCA accurately identifies self-similar patterns, except in cases of high homogeneity in transmission abilities or shallow tree generations. Empirical analysis of Canadian COVID-19 data further confirms the robustness of these self-similar structures, even in the presence of disruptions in transmission chains, aligning with observed power-law growth. In summary, the BCA detects self-similarity in early epidemic transmission stages by identifying CICSs within transmission trees, providing critical insights for forecasting epidemic trajectories and characterizing early transmission dynamics. © 2025, The Authors. All rights reserved.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

Media Discourse, Influence, and Reflection: Content Analysis and Text-Mining Study of Suicides and Homicides in Long-Term Care

引用

Journal of Medical Internet Research 2025年 27卷

作者： Wang, Charlotte Fang, Hsiu-Ju Lu, Hsin-Yang Chen, Chen-Fen Institute of Health Data Analytics and Statistics College of Public Health National Taiwan University Taipei Taiwan Master of Public Health Program College of Public Health National Taiwan University Taipei Taiwan Department of Long-Term Care National Taipei University of Nursing and Health Sciences Taipei Taiwan

Background: As populations age, the demand for long-term care services steadily increases. The effectiveness of government-promoted long-term care policies and the public’s access to relevant service information are demonstrably influenced by media representation. In addition, prior research has suggested that news framing can mitigate the negative influence (the Werther effect) with a more hopeful framing (the Papageno effect), thereby reducing the public’s susceptibility to negative news. Objective: This study investigates the phenomenon of suicides and homicides in long-term care reported in the news, in which family caregivers or care receivers died by suicide or homicide. We examined changes in the media’s reporting framework before and after the implementation of Taiwan’s Long-Term Care Plan 2.0 in 2017. We further examined the consistency between the content of news reports and the information provided by the media on long-term care services and suicide prevention (eg, hotlines). Methods: Content analysis and text-mining techniques were used to analyze 433 news reports covering 95 cases of suicides and homicides in long-term care in Taiwan from 2009 to 2021. A random-effects model was applied to examine term frequency transition post implementation. Results: The majority (>60%) of the cases involved family caregivers’ homicide-suicide. The term “family moral tragedy” has been replaced by “long-term care tragedy” in recent discourse. This shift is evident in a decline in the frequency of “family moral tragedy” since 2017, with usage decreasing by 32.4% in headlines and by 24% in news content. The term frequency of “care burden” has significantly increased from 0.0006 (SD 0.0008) to 0.017 (SD 0.0461;t₃₃₇=3.006;P=.003). While linguistic characteristics of the content have remained consistent, there were statistically significant differences in medical and ethics-related terms. The media tends to provide more suicide prevention information (eg, hotline

关键词： family caregiver homicide news framing suicide text mining

来源：评论

学校读者我要写书评

暂无评论

Correction: Motif clustering and digital biomarker extraction for free-living physical activity analysis

引用

BioData mining 2025年第1期18卷 34页

作者： Ya-Ting Liang Charlotte Wang Institute of Epidemiology and Preventive Medicine College of Public Health National Taiwan University Taipei Taiwan. Institute of Health Data Analytics and Statistics College of Public Health National Taiwan University No. 17 Xu-Zhou Road Taipei 100025 Taiwan. cywwang@ntu.edu.tw. Master of Public Health Program College of Public Health National Taiwan University Taipei Taiwan. cywwang@ntu.edu.tw.

来源：评论

学校读者我要写书评

暂无评论

Static and dynamic panel models: Which is better? (Case study: Poverty data in Indonesia 2012-2019) 1

Static and dynamic panel models: Which is better? (Case stud...

引用

1st Universitas Lampung International Conference on Science, Technology and Environment, ULICOSTE 2020

作者： Sihombing, P.R. Arsani, A.M. Departemen of Statistics IPB University Bogor Indonesia Applied Economics Master Program Padjadjaran University Bandung Indonesia Statistics Indonesia Indonesia

Indonesia's poverty rate continues to decline from year to year;however, the decline in the percentage of poor people is slowing down continuously, following the law of diminishing returns. Many factors influence poverty, including HDI, Gini ratio, and the open unemployment rate. This study model static and dynamic panel data on Indonesia's poverty cases in 2012-2019 to explain the relationship between HDI, Gini ratio, and the unemployment rate to poverty. It also compares poverty modeling with static and dynamic panel data regression. The results show that the Gini ratio has a significant and positive relationship to poverty. HDI has a significant and negative relationship to poverty. However, there is insufficient evidence to suggest that unemployment affects poverty. Using static panel data regression, the selected model is a fixed effect model but still experiences violations of the classic autocorrelation assumption. The dynamic panel data model gives better results than the static panel data regression model when viewed from the r-square value and the number of variables that significantly affect. From the results of this study, it can be implemented that an increase in economic growth without an increase in equity has no impact on poverty. So, it needs extra efforts by the government in economic equality for all regions of Indonesia. Also, there is a need for additional actions by the government and the community to improve education, health, purchasing power, and community competitiveness. © Published under licence by IOP Publishing Ltd.

关键词： Population statistics

来源：评论

学校读者我要写书评

暂无评论

Primarily research on the mathematical problem-solving abilities of grade X Vocational High School students for Computer and Network Engineering Competencies 4

Primarily research on the mathematical problem-solving abili...

引用

4th International Conference on Mathematics and Mathematics Education, ICM2E 2020

作者： Oktari, A. Permana, D. Master Program Universitas Negeri Padang Padang City Indonesia Department of Statistics Universitas Negeri Padang Padang City Indonesia

Computer and network engineering are one of the skill competencies in Vocational high school. Graduates of this major are expected to have problem-solving abilities to be able to compete in the working world. Mathematics is a subject that can develop mathematical problemsolving abilities. This study aimed to see, review and describe the mathematical problem-solving abilities of students and draw conclusions about these things as the phenomena found and to review the conditions of the learning devices used and developed by the teacher. This research used descriptive quantitative research methods. The data was collected through interviews, observations, and giving tests. Data analysis techniques used descriptive techniques for the results of observation and interview data, and scoring rubrics for mathematical problem-solving abilities for test result data. The results showed that the students' mathematical problem-solving abilities were still not relatively low. Some of the factors that cause this are 1) the learning was still using the lecturing method so that learning became meaningless, 2) students were accustomed to copying and memorizing formulas given by the teacher without understanding them, 3) the learning tools used were the same for each skill program, 4) a mathematics learning approach had not led students to be actively involved in the learning process and had not guided students to construct their knowledge in improving mathematical problem-solving abilities. Based on these results, further action is needed in the form of research on the development of constructivism-based learning tools to improve the mathematical problem-solving abilities of students in the Vocational High School for Computer and Network Engineering Competencies. © Published under licence by IOP Publishing Ltd.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Improved LSTM hyperparameters alongside sentiment walk-forward validation for time series prediction

引用

Journal of Open Innovation: Technology, Market, and Complexity 2025年第1期11卷

作者： Wahyuddin, Eko Putra Caraka, Rezzy Eko Kurniawan, Robert Caesarendra, Wahyu Gio, Prana Ugiana Pardamean, Bens Department of Statistical Computing Politeknik Statistika STIS Jakarta 13330 Indonesia Statistics Indonesia (BPS) Jl. Dr Sutomo 6-8 Jakarta Indonesia School of Economics and Business Telkom University Bandung 40257 Indonesia Research Center for Data and Information Sciences Research Organization for Electronics and Informatics National Research and Innovation Agency (BRIN) Bandung 40135 Indonesia Faculty of Integrated Technologies Universiti Brunei Darussalam Gadong BE1410 Brunei Darussalam Department of Mathematics Universitas Sumatera Utara Medan 20155 Indonesia Bioinformatics and Data Science Research Centre Bina Nusantara University DKI Jakarta 11480 Indonesia Computer Science Department BINUS Graduate Program Master of Computer Science Program Bina Nusantara University DKI Jakarta 11480 Indonesia

This study aims to address the common issue of biased estimation errors in time series modeling by analyzing the error in locating ideal hyperparameters and defining appropriate validation methods. Specifically, it focuses on predicting the stock price of Bank Rakyat Indonesia using a combination of historical stock prices, technical indicators, exchange rates, and news sentiment data, while determining the optimal variables for deep learning models. Employing a deep learning-based Long-Short Term Memory (LSTM) model, the study optimizes hyperparameters alongside walk-forward validation for time series prediction. It explores different combinations of variables and adapts the sliding window approach to the context of the data. The results highlight the importance of optimizing hyperparameters and utilizing walk-forward validation for accurate time series prediction. The model incorporating historical stock prices and sentiment scores outperforms others, achieving an RMSE of 96.61 and MAE of 86.97. Incorporating sentiment scores reduces RMSE by 39.55 % compared to models using only historical stock prices, while adding technical indicators does not yield improvement. This study contributes to the field by addressing the issue of biased estimation errors in time series modeling, offering insights into the optimization of hyperparameters and validation methods for accurate predictions. It also underscores the significance of incorporating sentiment analysis alongside historical stock prices for improved forecasting accuracy. © 2024 The Authors

关键词： LSTM Sentiment analysis Stock price prediction Time series modeling Validation methods

来源：评论

学校读者我要写书评

暂无评论

Survival Information System Using ReliefF Feature Selection and Backpropagation in Hepatocellular Carcinoma Disease 24

Survival Information System Using ReliefF Feature Selection ...

引用

24th International Seminar on Intelligent Technology and Its Applications, ISITIA 2023

作者： Wulandari, Umi Meganinditya Warsito, Budi Farikin, Farikhin Diponegoro University Master Program of Information Systems Semarang Indonesia Diponegoro University Department of Statistics Semarang Indonesia Diponegoro University Department of Mathematichs Semarang Indonesia

ISBN: (纸本)9798350313956

This study was conducted to classify patients with hepatocellular carcinoma by dividing the dataset features. The purpose of this study was to select the features in medical records with the influential features on hepatocellular carcinoma. In this paper, the proposed model is the use of ReliefF at the feature selection stage and Backpropagation at the classification stage. At this feature selection stage, there are two steps, namely weight calculation and feature reduction process. In the feature weight calculation step, each feature is given a weight and the resulting features will be processed in the feature reduction process. The results of the feature weight calculation will produce a ranking from the highest value to the lowest value which will then be reduced by the feature ranking. The best features that have been produced will be used as input for the second stage, namely the classification stage using Backpropagation. The results showed that the 10 features used were the best features out of 39 features. The best accuracy is produced by the ReliefF+BPNN method of 85%. The comparison results show that the ReliefF feature selection method has the best success rate among all. The results of this study indicate that the proposed method is successful. © 2023 IEEE.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Towards a legal compliance verification approach on the procurement process of IT solutions for the Brazilian Federal Public Administration

Towards a legal compliance verification approach on the proc...

引用

2014 IEEE 7th International Workshop on Requirements Engineering and Law, RELAW 2014

作者： Da Silva Barboza, Lamartine De A. Cysneiros Filho, Gilberto A. De Souza, Ricardo A. C. Master's Program in Applied Informatics Federal Rural University of Pernambuco Recife Brazil Department of Statistics and Informatics Federal Rural University of Pernambuco Recife Brazil

ISBN: (纸本)9781479963256

The Brazilian federal government regulates the process for procurement of Information Technology (IT) solutions through specific legislation named Regulatory Instruction-RI N° 04/2010. This process consists of three phases: procurement planning, supplier selection and contract management. This work helps (i) specify and validate an approach for traceability between legal requirements and documents created in the procurement process of IT solutions;(ii) reduce manual work for the verification of legal compliance in the set of artifacts produced;and (iii) support activities of auditing and inspection during and after the procurement of IT solutions by the Brazilian federal government. © 2014 IEEE.

关键词： Public administration

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：