Article
Biochemical Research Methods
James W. Webber, Kevin M. Elias
Summary: This study presents an efficient data imputation method based on constrained least squares and algorithms from the inverse problems literature, with applications in miRNA expression analysis. The proposed method shows significantly faster imputation speed and equal or higher accuracy compared to similar methods. This is important for improving cancer prediction accuracy in the presence of missing data.
BMC BIOINFORMATICS
(2022)
Article
Computer Science, Interdisciplinary Applications
Ji Hee Kim, Naeun Choi, Seongmin Heo
Summary: This work proposes a novel iterative least squares method to approximate nonlinear functions using constrained least squares to ensure continuity. The method improves upon the existing continuous piecewise linear (CPWL) method by modifying the main steps and employing partitioned least squares and constrained least squares to reduce computational complexity. An iterative procedure with gradient descent using momentum is used for breakpoint updates to improve convergence characteristics.
COMPUTERS & CHEMICAL ENGINEERING
(2022)
Article
Computer Science, Artificial Intelligence
Nicolas Nadisic, Jeremy E. Cohen, Arnaud Vandaele, Nicolas Gillis
Summary: This paper introduces a new form of sparse MNNLS problem and a two-step algorithm to solve it. By dividing the problem into subproblems and selecting Pareto front solutions, a matrix that satisfies the sparsity constraint is constructed. Experimental results show that this method is more accurate than existing heuristic algorithms.
Article
Computer Science, Information Systems
Xiao He, Ye Li, Jian Tan, Bin Wu, Feifei Li
Summary: Seasonal-trend decomposition is a fundamental concept in time series analysis, but existing methods are not efficient for real-time analysis. In this paper, we propose OneShotSTL, an algorithm that can decompose time series online with high efficiency and accuracy. OneShotSTL is more than 1,000 times faster than batch methods and achieves comparable or even better accuracy. Experimental results on benchmark datasets demonstrate its superiority.
PROCEEDINGS OF THE VLDB ENDOWMENT
(2023)
Article
Engineering, Multidisciplinary
Fei Xie, Wallace W. L. Lai, Xavier Derobert
Summary: The development of ground penetrating radar (GPR) as a near-surface geophysical detection method has led to an accurate survey equipment requiring an understanding of measurement errors and uncertainties. By using a constrained least squares algorithm, error sources affecting depth measurement of buried objects were modeled and uncertainty analysis was performed. Experimental validation showed that with a 95% confidence level, a centimetre-order of uncertainty can be achieved for depth estimation of objects at several meters deep, with errors in GPR center frequencies dominating the evaluation of uncertainty.
Article
Engineering, Industrial
Bin Liu, Yimin Shi, Hon Keung Tony Ng, Xiangwen Shang
Summary: This paper proposes a nonparametric Bayesian analysis method for system failure modes with dependence, constructing a dependence structure using copula function and deriving component reliability estimators from subsurvival function estimations. The study highlights the impact of dependence on estimation performance, emphasizing the importance of considering dependence between failure modes.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2021)
Article
Mathematics, Applied
Feng Ding
Summary: Least squares is an important method used for solving linear fitting and quadratic optimization problems. This paper explores the properties of least squares methods and multi-innovation least squares methods, and demonstrates important contributions in the area of system identification such as auxiliary model identification, multi-innovation identification theory, hierarchical identification principle, coupling identification concept, and filtering identification idea. The results of least squares and multi-innovation least squares algorithms for linear regressive systems with white noises can be extended to systems with colored noises.
JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS
(2023)
Article
Engineering, Industrial
Adel Ahmadi Nadi, Robab Afshari, Bahram Sadeghpour Gildeh
Summary: Competing risks data frequently appear in various real-world operations. This paper introduces two Shewhart-type control charts for monitoring the relative risk rate of independent Weibull random variables. The calculation and evaluation of the control charts are based on Monte Carlo simulations. The performance of the control charts is examined using the average run length metric, and an illustrative example is provided.
QUALITY TECHNOLOGY AND QUANTITATIVE MANAGEMENT
(2023)
Article
Chemistry, Medicinal
Zhenqiu Shu, Qinghan Long, Luping Zhang, Zhengtao Yu, Xiao-Jun Wu
Summary: The progress in single-cell RNA sequencing (ScRNA-seq) technology allows for the accurate discovery of cell heterogeneity and diversity. Clustering is a crucial step in ScRNA-seq data analysis, but it faces challenges due to the high dimensionality and noise of the data. To overcome these challenges, we propose a novel ScRNA-seq data clustering model, RGNMF-DS, which incorporates similarity and dissimilarity regularizers for matrix decomposition and utilizes a graph regularizer to uncover the local geometric structure in the data. Experimental results demonstrate that our proposed model outperforms other state-of-the-art methods in clustering ScRNA-seq datasets.
JOURNAL OF CHEMICAL INFORMATION AND MODELING
(2022)
Article
Chemistry, Analytical
Weiran Song, Zongyu Hou, Muhammad Sher Afgan, Weilun Gu, Hui Wang, Jiacheng Cui, Zhe Wang, Yun Wang
Summary: This study introduces a method for improving the predictive ability and computational efficiency of multivariate models in coal property analysis through ensemble learning. Experimental results show that this method outperforms benchmark algorithms in variable selection and performs better in six out of seven tasks. This approach can serve as a reliable choice when users are unsure which variables to select.
JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY
(2021)
Article
Chemistry, Analytical
Jie Huang, Xiaojing Chen, Zhonghao Xie, Shujat Ali, Xi Chen, Leiming Yuan, Chengxi Jiang, Guangzao Huang, Wen Shi
Summary: This study explores the application of laser-induced breakdown spectroscopy (LIBS) combined with partial least squares (PLS) model for predicting copper concentrations in T. granosa. The study uses the robust partial least squares (RSIMPLS) method to process high-dimensional LIBS data and successfully improves the prediction accuracy of copper concentrations.
ANALYTICAL METHODS
(2023)
Article
Environmental Sciences
Eva Gorrochategui, Isabel Hernandez, Roma Tauler
Summary: A powerful methodology based on the MCR-ALS method is proposed to handle complex and incomplete atmospheric data sets, providing concise results. The methodology is used to investigate changes in air quality and evaluate correlations among pollutants in Barcelona and other parts of Catalonia.
ATMOSPHERIC CHEMISTRY AND PHYSICS
(2022)
Article
Mathematics
Yosra Yousif, Faiz Elfaki, Meftah Hrairi, Oyelola Adegboye
Summary: This paper discusses masked issues in dealing with competing risk data and proposes a Bayesian analysis method for assessing the impact of explanatory variables on the cumulative incidence function. The effectiveness of the method is tested through numerical studies with simulated and real data sets.
Article
Multidisciplinary Sciences
Freeh N. Alenezi
Summary: The study introduces a method for variable selection in high dimensional data modeling, using majority scoring with backward elimination in PLS to improve prediction accuracy. The method performs well in predicting corn and diesel contents, while also examining the impact of data properties on prediction behavior.
SCIENTIFIC REPORTS
(2021)
Article
Engineering, Mechanical
Jing Chen, Yawen Mao, Manfeng Hu, Liuxiao Guo, Quanmin Zhu
Summary: This study proposes a decomposition optimization-based expectation maximization algorithm for switching models. The identities of each sub-model are estimated in the expectation step, while the parameters are updated using the decomposition optimization method in the maximization step. Compared with the traditional expectation maximization algorithm and the gradient descent expectation maximization algorithm, the decomposition optimization-based expectation maximization algorithm avoids the matrix inversion and eigenvalue calculation; thus, it can be extended to complex nonlinear models and large-scale models. Convergence analysis and simulation examples are given to show the effectiveness of the proposed algorithm.
NONLINEAR DYNAMICS
(2023)
Article
Engineering, Industrial
M. H. Ling, H. K. T. Ng, K. L. Tsui
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2019)
Article
Statistics & Probability
M. H. Ling, P. S. Chan, H. K. T. Ng, N. Balakrishnan
Summary: This study focused on two Archimedean copula models for analyzing data from one-shot devices with two correlated failure modes. Initial values for dependence parameters and maximum likelihood estimates of model parameters were provided, along with an examination of how stress levels impact the correlation between failure modes. Real data from a survival experiment were re-analyzed to demonstrate the proposed methods.
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS
(2021)
Article
Engineering, Industrial
M. H. Ling, X. W. Hu
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2020)
Article
Statistics & Probability
Chien-Tai Lin, Narayanaswamy Balakrishnan, Man Ho Ling
Summary: This study investigates the exact null distributions of test statistics for testing upper outliers in a two-parameter Laplace sample. Two types of test statistics are considered, and the critical values are obtained using mathematical methods. Some examples are provided for illustration.
COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION
(2022)
Article
Energy & Fuels
Chun-Pang Lin, Javier Cabrera, Fangfang Yang, Man-Ho Ling, Kwok-Leung Tsui, Suk-Joo Bae
Article
Engineering, Multidisciplinary
Narayanaswamy Balakrishnan, Elena Castilla, Man Ho Ling
Summary: This paper investigates the effect of model misspecification on the optimal design of constant-stress accelerated life-tests, showing that assuming Weibull or lognormal lifetime distributions is more robust while assuming gamma lifetime distribution is the most non-robust.
QUALITY AND RELIABILITY ENGINEERING INTERNATIONAL
(2022)
Article
Engineering, Industrial
Chun Pang Lin, Man Ho Ling, Javier Cabrera, Fangfang Yang, Denis Yau Wai Yu, Kwok Leung Tsui
Summary: This paper introduces a two-phase gamma process model to accurately simulate the cycle aging of batteries, providing better estimation of parameters such as state of charge, remaining useful discharge time, state of life, and remaining useful life. The proposed model outperforms the conventional two-term exponential model in generating more accurate predictions, highlighting the advantages of voltage-based modeling over capacity-based modeling. Additionally, an analytical expression for mean useful discharge time in a cycle is developed using a Taylor expansion and the Birnbaum-Saunders distribution, showing good agreement with the true mean of a gamma process.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2021)
Article
Mathematics
Man-Ho Ling, Narayanaswamy Balakrishnan, Chenxi Yu, Hon Yiu So
Summary: The paper presents an efficient expectation-maximization algorithm for one-shot device test data with multiple failure modes, utilizing exponential lifetime distributions of dependent components under constant-stress accelerated life-test. It discusses the maximum likelihood estimate and confidence intervals for the mean lifetime of one-shot devices under normal operating conditions, and evaluates the performance of the proposed inferential methods through Monte Carlo simulations. Three examples are used to illustrate the proposed methods, including Class-H failure modes data, mice data from an experiment, and simulated data with four failure modes.
Article
Mathematics
Man-Ho Ling
Summary: This study develops a general framework to obtain optimal constant-stress accelerated life test plans for one-shot devices with dependent components, subject to time and budget constraints. The optimal accelerated test plan considers an economical approach and determines the inspection time and sample size by minimizing the asymptotic variance of the maximum likelihood estimator for mean lifetime.
Article
Engineering, Multidisciplinary
Deepak Prajapati, Man Ho Ling, Ping Shing Chan, Debasis Kundu
Summary: This paper investigates the application of copula models in lifetime data analysis, focusing on the estimation of reliability for one-shot devices. The effects of misspecification on estimation are examined, and the Akaike information criterion is used for model validation. Simulation results demonstrate the impact of misspecification on estimation under different copula models and marginal distributions.
PROCEEDINGS OF THE INSTITUTION OF MECHANICAL ENGINEERS PART O-JOURNAL OF RISK AND RELIABILITY
(2023)
Article
Engineering, Industrial
Xiangwen Shang, Hon Keung Tony Ng, Man Ho Ling
Summary: One-shot device test data with defects induced in a manufacturing process are studied, and a method for estimating mean-time-to-failure is proposed. The impacts of masking effect on estimation are evaluated through a Monte Carlo simulation study, and practical guidelines and recommendations are provided.
QUALITY ENGINEERING
(2023)
Article
Public, Environmental & Occupational Health
Chun-Pang Lin, Ilaria Dorigatti, Kwok-Leung Tsui, Min Xie, Man-Ho Ling, Hsiang-Yu Yuan
Summary: The study found that the effective reproduction number of seasonal influenza in Hong Kong significantly decreased after the community spread of COVID-19 began. The results suggest that wearing face masks and avoiding crowded places potentially have significant suppressive impacts on influenza transmission.
FRONTIERS IN PUBLIC HEALTH
(2022)
Article
Engineering, Multidisciplinary
Man Ho Ling, Suk Joo Bae
Summary: An accelerated degradation test (ADT) is proposed to hasten the degradation mechanisms of products. A random-effect gamma process model is used for the reliability analysis of ADDT data, taking into account random initial degradation levels. Maximum likelihood estimates (MLEs) of model parameters are derived and an inferential procedure is constructed using asymptotic properties of the MLEs. The performance of the proposed methods is validated through Monte Carlo simulations.
QUALITY AND RELIABILITY ENGINEERING INTERNATIONAL
(2023)
Article
Statistics & Probability
Chien-Tai Lin, Yu-Nan Wu, N. Balakrishnan, Man Ho Ling
Summary: OTL is an online tool developed to perform outlier analysis in various samples using Shiny and R, aiding academics, policy-makers, and executives to compare and understand the results of outlier testing procedures more efficiently.
COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION
(2022)
Article
Nursing
Mei Hua Kerry Hsu, Qian Hong Ye, Man Ho Ling
Summary: This study explored the career preferences and related factors among nursing students in Macao, finding that senior students prefer community health nursing, while junior and female students prefer pediatric and obstetric & gynecological nursing.
