检索结果-内蒙古大学图书馆

SSRN 2025年

作者： Hsu, Cheyu Cheng, Yuju L. Chang, Chunhao Quita, Reyna Chen, Rouyi Wang, Weichung Division of Medical Oncology Department of Oncology National Taiwan University Hospital Taipei100225 Taiwan Institute of Applied Mathematical Sciences National Taiwan University Taipei106319 Taiwan Data Science Master Degree Program National Taiwan University Academia Sinica Taipei106319 Taiwan National Taiwan University of Science and Technology Taiwan

Objectives: Accurate classification of hypopharyngeal MRI sequences and acquisition planes is crucial for improving clinical decision-making. This study evaluates classifiers based on machine learning (ML), deep learning (DL), and large language models (LLM), and their integration through voting inference methods to improve classification accuracy. Methods: This retrospective multicenter study analyzed DICOM-format datasets from three hospitals. The internal dataset included 567 hypopharyngeal cancer patients from National Taiwan University Hospital (NTUH) (2008–2022). The external datasets consisted of 109 patients from Chang Gung Memorial Hospital (CGMH) (2003–2010) and 51 patients from NTUH (Yunlin) (2009–2015). The study employed LightGBM for the ML model, ResNet-18 for DL, and ChatGPT for the LLM model. Model performance was assessed using accuracy and F1-score, while the chi-square test was used to evaluate the performance of individual models and the ensemble ***: The results demonstrated high accuracy for the ML and DL models, achieving scores of 0.9746 and 0.9771, respectively, on internal testing. Additionally, on external testing, the ML and DL models attained accuracies of 0.9950 and 0.9862, respectively. In contrast, the LLM model exhibited moderate performance, achieving accuracies of 0.7634 for internal testing and 0.8552 for external testing. The voting classifier achieved the highest accuracy, with 0.9796 and 0.9962 for internal and external testing, ***: These results suggest that combining ML, DL, and LLM models can enhance hypopharyngeal MRI classification. © 2025, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Accuracy of a one-dimensional reduction of dynamical systems on networks

arXiv

引用

arXiv 2021年

作者： Kundu, Prosenjit Kori, Hiroshi Masuda, Naoki Department of Mathematics State University of New York BuffaloNY14260-2900 United States Department of Complexity Science and Engineering The University of Tokyo Chiba277-8561 Japan Computational and Data-Enabled Science and Engineering Program State University of New York at Buffalo BuffaloNY14260-5030 United States

Resilience is an ability of a system with which the system can adjust its activity to maintain its functionality when it is perturbed. To study resilience of dynamics on networks, Gao et al. [Nature, 530, 307 (2016)] proposed a theoretical framework to reduce dynamical systems on networks, which are high dimensional in general, to one-dimensional dynamical systems. The accuracy of this one-dimensional reduction relies on several assumption in addition to the assumption that the network has a negligible degree correlation. In the present study, we analyze the accuracy of the one-dimensional reduction assuming networks without degree correlation. We do so mainly through examining the validity of the individual assumptions underlying the method. Across five dynamical system models, we find that the accuracy of the one-dimensional reduction hinges on the spread of the equilibrium value of the state variable across the nodes in most cases. Specifically, the one-dimensional reduction tends to be accurate when the dispersion of the node's state is small. We also find that the correlation between the node's state and the node's degree, which is common for various dynamical systems on networks, is unrelated to the accuracy of the one-dimensional reduction. © 2021, CC BY.

关键词： Dynamical systems

来源：评论

学校读者我要写书评

暂无评论

Improving wind speed forecasting at Adama wind farm II in Ethiopia through deep learning algorithms

引用

Case Studies in Chemical and Environmental Engineering 2024年 9卷

作者： Chaka, Mesfin Diro Semie, Addisu Gezahegn Mekonnen, Yedilfana Setarge Geffe, Chernet Amente Kebede, Hailemichael Mersha, Yonas Anose, Fikru Abiko Benti, Natei Ermias Computational Data Science Program College of Natural and Computational Sciences Addis Ababa University P. O. Box 1176 Addis Ababa Ethiopia Center for Environmental Science College of Natural and Computational Sciences Addis Ababa University P. O. Box 1176 Addis Ababa Ethiopia Department of Physics College of Natural and Computational Sciences Addis Ababa University P.O. Box 1176 Addis Ababa Ethiopia Ethiopian Institute of Water Resources (EIWR) Addis Ababa University (AAU) Addis Ababa Ethiopia Department of Physics College of Natural and Computational Sciences Wolkite University Ethiopia

Wind power plays a critical role in supporting Ethiopia's electricity generation, particularly during dry seasons when hydropower availability diminishes. This contribution becomes even more crucial for sustaining industrial growth. However, accurately estimating wind energy remains challenging due to its random and highly variable nature. Deep learning models for time series prediction have been employed to address this issue, and this study evaluates the effectiveness of long short-term memory (LSTM) and gated recurrent unit (GRU) models in predicting wind speed at Adama Wind Farm II in Ethiopia. Model performance was assessed using R², RMSE, and MAE. GRU significantly outperformed LSTM across all metrics and forecasting horizons. For daily forecasting, GRU achieved an R² of 0.9727, exceeding LSTM's 0.9062 by 7.34 %, with corresponding RMSE values of 0.0005 and 0.0018, respectively. Similarly, GRU surpassed LSTM's performance for weekly and monthly forecasting. In weekly forecasting, GRU achieved an R² of 0.9969 compared to LSTM's 0.9930, with RMSE values of 0.0001 for both models. For monthly forecasting, GRU achieved an R² of 0.9330 compared to LSTM's 0.9303, with corresponding RMSE values of 0.0019 for both models. MAE values also followed the same pattern, with GRU consistently demonstrating lower values than LSTM across all forecasting horizons. These findings advance renewable energy and wind power forecasting research, highlighting the adaptability and versatility of LSTM and GRU models for capturing wind behavior. Decision-makers can use these findings to optimize wind power production, enhance grid operation efficiency, and promote sustainable and cost-effective wind energy. LSTM and GRU models have emerged as powerful tools for precise wind speed and wind power forecasting, which are essential for modern wind energy planning systems. They lay the foundation for a more environmentally friendly and sustainable en

关键词： Adama wind farm II Deep learning Ethiopia LSTM and GRU models Renewable energy Wind speed forecasting

来源：评论

学校读者我要写书评

暂无评论

Gender Disparities in Parkinson's Disease: Exploring Quality of Life Among Mexican Men and Women (2011-2019)

引用

Parkinsonism and Related Disorders 2024年 122卷 106832-106832页

作者： Hernández-Medrano, A.J. Medina-Pérez, M.F. Romero-Terán, D.P. Abundes-Corona, A. Cervantes-Arriaga, A. Rodríguez-Violante, M. Instituto Nacional de Neurología y Neurocirugía Clinical Laboratory of Neurodegenerative Diseases Mexico City Mexico University of Maryland Baltimore County Master of Professional Studies (MPS) in Data Science Program Baltimore United States

来源：评论

学校读者我要写书评

暂无评论

State-transition dynamics of resting-state functional magnetic resonance imaging data: Model comparison and test-to-retest analysis

arXiv

引用

arXiv 2023年

作者： Islam, Saiful Khanra, Pitambar Nakuci, Johan Muldoon, Sarah F. Watanabe, Takamitsu Masuda, Naoki Department of Mathematics University at Buffalo State University of New York Buffalo United States Institute for Artificial Intelligence and Data Science University at Buffalo State University of New York Buffalo United States Neuroscience Program University at Buffalo State University of New York Buffalo United States School of Psychology Georgia Institute of Technology United States International Research Centre for Neurointelligence The University of Tokyo Institutes for Advanced Study Tokyo Japan

Electroencephalogram (EEG) microstate analysis entails finding dynamics of quasi-stable and generally recurrent discrete states in multichannel EEG time series data and relating properties of the estimated state-transition dynamics to observables such as cognition and behavior. While microstate analysis has been widely employed to analyze EEG data, its use remains less prevalent in functional magnetic resonance imaging (fMRI) data, largely due to the slower timescale of such data. In the present study, we extend various data clustering methods used in EEG microstate analysis to resting-state fMRI data from healthy humans to extract their state-transition dynamics. We show that the quality of clustering is on par with that for various microstate analyses of EEG data. We then develop a method for examining test-retest reliability of the discrete-state transition dynamics between fMRI sessions and show that the within-participant test-retest reliability is higher than between-participant test-retest reliability for different indices of state-transition dynamics, different networks, and different data sets. This result suggests that state-transition dynamics analysis of fMRI data could discriminate between different individuals and is a promising tool for performing fingerprinting analysis of *** Codes 92F05 © 2023, CC BY.

关键词： Dynamics

来源：评论

学校读者我要写书评

暂无评论

Prediction of Stock Price and Direction Using Neural Networks: datasets Hybrid Modeling Approach

Prediction of Stock Price and Direction Using Neural Network...

引用

data Analytics for Business and Industry: Way Towards a Sustainable Economy (ICDABI), International Conference on

作者： Mohamed Al Aradi Nabil Hewahi Big Data Science & Analytics Master Program Collage of Science Univestity of Bahain Sakhir Kingdom of Bahrain Collage of Information Technology Univestity of Bahain Sakhir Kingdom of Bahrain

ISBN: (数字)9781728196756

ISBN: (纸本)9781728196763

Stock prices prediction is one of the most daunting tasks to achieve for day traders, investors, and data scientists. They are complex functions of a wide array of contributing factors that affects the movement dynamics. Politics, social outlook, the company's sales performance, and socio-economic factors all have an influence of how the stock is perceived globally and thus affecting the price movement based on the supply/demand balance. In this work a Framework is proposed to predict the price and movement direction by using multiple datasets that includes news sentiment, social sentiment, company earnings announcement, and technical indicators. A case study was applied on the Apple Inc. stock using Long short-term Memory Neural Networks (LSTM) and Deep Neural Networks (DNN). Results showed superiority of the LSTM model in both predicting stock value [ 75.4 MSE, +-2.52 PE] and the direction with [70.1% classification accuracy] over the DNN model which suffered from prediction lag and dependency on trend indicators achieving [53.1%] in predicting the direction of the stock.

关键词： Predictive models Machine learning Companies Neural networks Market research data models Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Tweets Sentiment Analysis During COVID-19 Pandemic

Tweets Sentiment Analysis During COVID-19 Pandemic

引用

data Analytics for Business and Industry: Way Towards a Sustainable Economy (ICDABI), International Conference on

作者： Maha A. Alanezi Nabil M. Hewahi Big Data Science & Analytics Master Program Collage of Science University of Bahrain Sakhir Kingdom of Bahrain Collage of Information Technology University of Bahrain Sakhir Kingdom of Bahrain

ISBN: (数字)9781728196756

ISBN: (纸本)9781728196763

The aim of this paper is to investigate the impact of social distance on people during COVID-19 pandemic using twitter sentiment analysis through a comparison between the k-means clustering and Mini-Batch k-means clustering approaches. To find the most common frequent words, two datasets have been investigated (WHO and Bahrain ministry of health datasets) to be as data preparation and exploration. Another two datasets (English and Arabic datasets) are used in the clustering of k-means. In this paper, a comparison between k-means and Mini-Batch k-means is performed to find a pattern. The word frequency shows that there are several words related to the pandemic. The sentiment analysis result show that in USA, Australia, Nigeria, Canada, and England, most tweets are neutral. However, the majority of tweets are positive tweets from both Italy and India. In addition, the k-means cluster in the English dataset reveals several cluster trends where COVID-19 pandemic procedures are addressed in cluster 1, and health workers are encouraged in cluster 3.

关键词： COVID-19 Social networking (online) Pandemics Sentiment analysis Blogs Elbow Viruses (medical)

来源：评论

学校读者我要写书评

暂无评论

Graph Laplacian Learning with Exponential Family Noise

引用

IEEE Transactions on Signal and Information Processing over Networks 2025年

作者： Shi, Changhao Mishne, Gal UC San Diego Electrical and Computer Engineering Department CA92093 United States UC San Diego Hal.c.o.glu Data Science Institute and the Neurosciences Graduate Program CA92093 United States

Graph signal processing (GSP) is a prominent framework for analyzing signals on non-Euclidean domains. The graph Fourier transform (GFT) uses the combinatorial graph Laplacian matrix to reveal the spectral decomposition of signals in the graph frequency domain. However, a common challenge in applying GSP methods is that in many scenarios the underlying graph of a system is unknown. A solution in such cases is to construct the unobserved graph from available data, which is commonly referred to as graph or network inference. Although different graph inference methods exist, they are restricted to learning from either smooth graph signals or simple additive Gaussian noise. Other types of noisy data, such as discrete counts or binary digits, are rather common in real-world applications, yet are underexplored in graph inference. In this paper, we propose a versatile graph inference framework for learning from graph signals corrupted by exponential family noise. Our framework generalizes previous methods from continuous smooth graph signals to various data types. We propose an alternating algorithm that jointly estimates the graph Laplacian and the unobserved smooth representation from the noisy signals. We also extend our approach to include an offset variable which models different levels of variation of the nodes. Since real-world graph signals are frequently non-independent and temporally correlated, we further adapt our original setting to a timevertex formulation. We demonstrate on synthetic and real-world data that our new algorithms outperform competing Laplacian estimation methods that suffer from noise model mismatch. © 2025 IEEE.

关键词： Laplace transforms

来源：评论

学校读者我要写书评

暂无评论

Trends and sex-specific incidence of upper urinary tract cancer in Taiwan: A birth cohort study

引用

Cancer medicine 2023年第14期12卷 15350-15357页

作者： Yu-Hsuan Chang Wan-Lun Hsu Yu-Kwang Lee Chun-Ju Chiang Ya-Wen Yang San-Lin You Yong-Chen Chen Tai-Shuan Lai School of Medicine Fu-Jen Catholic University New Taipei City Taiwan. Data Science Center College of Medicine Fu-Jen Catholic University New Taipei City Taiwan. Master Program of Big Data Analysis in Biomedicine College of Medicine Fu-Jen Catholic University New Taipei City Taiwan. Department of Surgery Division of General Surgery National Taiwan University Hospital Taipei Taiwan. Graduate Institute of Epidemiology and Preventive Medicine College of Public Health National Taiwan University Taipei Taiwan. Department of Internal Medicine National Taiwan University Hospital Taipei Taiwan. Department of Internal Medicine College of Medicine National Taiwan University Taipei Taiwan.

BACKGROUND:Taiwan has one of the highest incidences of upper tract urothelial cancer (UTUC) worldwide, especially in women; however, no nationwide, long-term follow-up study has evaluated this. METHODS:We investigated the incidence of UTUC in Taiwan using data from the national population-based Taiwan Cancer Registry database (1985-2019). We divided the birth cohort into nine 5-year age groups and calculated the age-specific incidence for these groups according to the corresponding birth years. RESULTS:The average annual percent change in the incidence of renal pelvis cancer from 1985 to 2019 showed sex-specific differences, with 3.5% and 5.3% increases in the incidences in men and women, respectively. The age-specific incidence rate for renal pelvis cancer among women showed a gradual increase in the group with older women as well as an increase over time in each age group. The results of a birth cohort analysis revealed that younger cohorts had higher incidence rates of renal pelvis cancer than older cohorts did. CONCLUSION:We demonstrated that the incidence of UTUC is unusually high among older Taiwanese women and that younger cohorts have a high risk of UTUC than older cohorts.

关键词： age period cohort incidence sex upper tract urothelial cancer

来源：评论

学校读者我要写书评

暂无评论

The development of video analysis instrument to determine teacher's character

引用

AIP Conference Proceedings 2023年第1期2594卷

作者： Y. B. Gromang M. A. Rudhito D. Sudigyo T. Suparyanto B. Pardamean Department of Mathematics Education Universitas Sanata Dharma Paingan Maguwoharjo Depok Sleman Yogyakarta Indonesia 55282 Bioinformatics & Data Science Research Center Bina Nusantara University Jakarta Indonesia 11480 Computer Science Department BINUS Graduate Program - Master of Computer Science Bina Nusantara University Jakarta Indonesia 11480

This study aims to develop a valid and reliable video analysis instrument to determine teachers’ character in the opening lesson in terms of its utterances and produce video analysis on micro-teaching that will be used in training related to Artificial Intelligence (AI). The type of research used is research and development. The subject of this research is the micro-teaching video of class E students’ batch 2019, Mathematics Education Study program at Sanata Dharma University. The product of this research is a video analysis instrument to find out the character of the prospective teachers, such as confidence, enthusiasm, and happiness in terms of voice utterances. This study uses the ADDIE research model. The results show that the video analysis instrument has a validity value of 4.80 which means it's very valid. Video analysis instrument to find out the enthusiastic character has a value of validity of 4.70, which means it's very valid. Then, the video analysis instrument to find out the character of happiness has a validity value of 4.60 which means it's very valid. The practicality of the developed video analysis instrument meets the required criteria with a practicality value of 4.31. Overall, the video analysis instrument developed belongs to an effective category.

关键词：

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：