Article
Multidisciplinary Sciences
Sandra Plancade, Magali Berland, Melisande Blein-Nicolas, Olivier Langella, Ariane Bassignani, Catherine Juste
Summary: A univariate selection method was proposed to handle missing values in metaproteomics data, which combines a test of association between missingness and classes, and a test for difference of observed intensities between classes. The method showed good performance in comparison to imputation-based methods, especially in simulated scenarios.
Article
Computer Science, Information Systems
Zeliha Ergul Aydin, Zehra Kamisli Ozturk
Summary: Medical prediction models have become increasingly prevalent due to their potential benefits in improving patient outcomes and healthcare efficiency. This study analyzes the impact of combining missing data imputation and filter-based feature selection methods on medical prediction models. Various combinations of imputation methods, feature selection methods, and classifiers were tested on medical datasets. The results show that the choice of imputation and feature selection methods had no significant effect on the performance of prediction models using logistic regression and support vector machine. However, certain combinations were found to perform better for the K-nearest neighbor classifier. Additionally, the experiments revealed differences in the runtime of feature selection methods and demonstrated the effectiveness of feature selection in improving prediction success rate.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Review
Biochemical Research Methods
Zilong Zhang, Feifei Cui, Chunyu Wang, Lingling Zhao, Quan Zou
Summary: Single-cell RNA sequencing has revolutionized the study of gene expression at a cellular level, but the noise and dimensionality of the data pose challenges for statistical analysis. While there are many tools available for scRNA-seq data analysis, a universal gold standard pipeline is still lacking. Understanding bioinformatics and computational issues can help in selecting the appropriate tools for data analysis.
BRIEFINGS IN BIOINFORMATICS
(2021)
Article
Computer Science, Information Systems
Masurah Mohamad, Ali Selamat, Ondrej Krejcar, Ruben Gonzalez Crespo, Enrique Herrera-Viedma, Hamido Fujita
Summary: This study proposes an alternate data extraction method, named CFS-DRSA, to enhance classifier performance and address challenges in large and problematic datasets. The method includes two crucial feature extraction tasks and has been experimentally validated to have high accuracy and credibility.
Article
Anatomy & Morphology
Piero Antonio Zecca, Andrea Brambilla, Marcella Reguzzoni, Marina Protasoni, Mario Raspanti
Summary: This study developed a LEGO-based sample positioning system for SEM analysis, which consistently identified and aligned features, maintained accurate positioning, and provided repeatable and reliable results. The system demonstrated high precision and accuracy in sample repositioning, with reliable replication of results. Further research is needed to optimize the system's design and evaluate its performance in different SEM applications.
MICROSCOPY RESEARCH AND TECHNIQUE
(2023)
Article
Neurosciences
Jun Liu, Lechan Sun, Jun Liu, Min Huang, Yichen Xu, Rihui Li
Summary: This study utilized a deep learning neural network to automatically recognize music-evoked emotions by combining region-specific information and dynamic functional connectivity of EEG signals. The results showed that the highest classification accuracy was achieved with the best channel combination located in the frontal region. Additionally, longer temporal functional networks of the frontal cortex improved the performance of emotion recognition.
FRONTIERS IN NEUROSCIENCE
(2022)
Article
Health Care Sciences & Services
Mingxuan Fan, Xiaoling Peng, Xiaoyu Niu, Tao Cui, Qiaolin He
Summary: Data loss is common in clinical data collection, and appropriate data imputation methods can improve medical diagnosis accuracy. This study compared five popular imputation methods and utilized three variable selection techniques to achieve accurate diagnosis. The GAIN imputation method and BAR variable selection method were found to be particularly effective for enhancing prediction accuracy in medical diagnosis.
BMC MEDICAL RESEARCH METHODOLOGY
(2023)
Article
Mathematics, Applied
Jian Dang
Summary: This article introduces the construction of an intelligent sports health management system based on big data analysis and the Internet of things (IoT). The system consists of the user, IoT, cloud, system analysis, evaluation, and data layers. A new multilabel feature selection algorithm is proposed and its effectiveness is demonstrated through experiments. The feasibility of the proposed intelligent sports management system is also analyzed.
JOURNAL OF FUNCTION SPACES
(2022)
Article
Biochemical Research Methods
Fahimeh Motamedi, Horacio Perez-Sanchez, Alireza Mehridehnavi, Afshin Fassihi, Fahimeh Ghasemi
Summary: This article discusses two approaches for quantitative structure-activity prediction studies, focusing on identifying appropriate molecular descriptors and predicting the biological activities of designed compounds. The use of LASSO-random forest algorithm is shown to significantly improve output correlation, reduce implementation time and model complexity, while maintaining prediction accuracy.
Article
Biochemistry & Molecular Biology
Heejin Jin, Surin Jung, Sungho Won
Summary: This study aimed to further develop the missForest algorithm by combining a binary particle swarm optimization (BPSO)-based feature-selection strategy. The improved method, BPSOmf, showed better imputation accuracy for continuous variables. BPSOmf is a suitable method for imputing target data consisting mainly of continuous variables.
Article
Agronomy
Ruihan Mao, Lei Zhou, Zhaojun Wang, Jianliang Wu, Jianfeng Liu
Summary: A new strategy called the block-free method is proposed in this study to select a subset of SNPs from a high-density chip to form a low-density panel. The strategy utilizes Feature Selection using a Feature Similarity (FSFS) algorithm and a Multiple-Objective, Local-Optimization (MOLO) algorithm. Compared to the uniform method and the block-based method, our strategy shows higher accuracy in genotype imputation and genomic prediction.
Article
Multidisciplinary Sciences
Xingjian Chen, Zifan Zhu, Weitong Zhang, Yuchen Wang, Fuzhou Wang, Jianyi Yang, Ka-Chun Wong
Summary: Predicting human diseases from microbiome data is important in medical applications. Existing methods often overlook the abundance profiles of known and unknown microbial organisms, as well as the taxonomic relationships among them, resulting in information loss. To address these issues, we developed a comprehensive machine learning framework called MetaDR that combines deep learning and various information sources to predict human diseases.
Article
Education & Educational Research
Roberto Bertolini, Stephen J. Finch, Ross H. Nehm
Summary: Educators aim to improve student performance by identifying relevant features and using efficient feature selection techniques in predictive modeling pipelines. The study found that methods like Correlation Attribute Evaluation and Fisher's Scoring Algorithm showed better performance in predicting student success, while Relief Attribute Evaluation was unstable and less effective. Factors such as grade point average, number of credits taken, and performance on concept inventory assessments were identified as key predictors of student performance.
INTERNATIONAL JOURNAL OF EDUCATIONAL TECHNOLOGY IN HIGHER EDUCATION
(2021)
Article
Computer Science, Artificial Intelligence
Pankhuri Jain, Anoop Tiwari, Tanmoy Som
Summary: This paper introduces a technique for missing value imputation and feature selection using fuzzy rough set-based approaches. The experimental results demonstrate its high applicability and robustness, as well as its ability to significantly reduce data dimensionality while maintaining high performance.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Engineering, Electrical & Electronic
Rui Lima, Luis Barreto, Antonio Amaral, Sara Paiva
Summary: This article proposes a smartphone-based solution using visual recognition to help visually impaired individuals navigate and position themselves, providing them with information about nearby places and improving their quality of life.
IEEE SENSORS JOURNAL
(2023)
Article
Engineering, Civil
Minh Hieu Nguyen
Summary: During the COVID-19 era, factors such as company closure policies and prior frequency of remote work played a significant role in determining the extent of telework. Fear of COVID-19 and difficulties in work concentration were identified as influential factors in the perception of telework. Additionally, having multiple children at home negatively impacted the perception of telework but positively influenced the willingness to adopt a hybrid work model.
Article
Green & Sustainable Science & Technology
Minh Hieu Nguyen, Jimmy Armoogum
Summary: The study found that there are differences between men and women in their perceptions and preferences towards remote work, with women showing a more positive attitude towards HBT and a desire to do more telework post-COVID-19. Men's perceptions were influenced by work-related factors, while women's perceptions were influenced by family-related factors.
Article
Green & Sustainable Science & Technology
Minh Hieu Nguyen, Jimmy Armoogum, Binh Nguyen Thi
Summary: This study investigates the determinants of change in e-shopping behaviors during the COVID-19 pandemic in Hanoi, Vietnam, finding that females are more likely to engage in e-shopping and that in-store shopping enjoyment and decreased income can respectively facilitate and deter e-shopping. Specifically, working from home is positively associated with more frequent e-purchasing for electronics, while fear of disease encourages higher frequencies of e-shopping for food and medical products.
Article
Engineering, Civil
Minh Hieu Nguyen, Dorina Pojani
Summary: This study surveyed over 800 students and found that the decision of university students to ride buses is influenced by various factors, including socio-demographic variables, environmental variables, and psychological variables. The fear of Covid-19 infection had minimal impact, while the introduction of 'clean and green' electric buses showed potential to attract students back to public transport.
Article
Economics
Duy Quy Nguyen-Phuoc, Diep Ngoc Su, Minh Hieu Nguyen, Nguyen S. Vo, Oscar Oviedo-Trespalacios
Summary: The present study investigates the impact of psychological factors on passengers' intention to use on-demand shared ride-hailing (OSR) services. The results confirm the relevance of attitudes, perceived behavioral control, and social norms in determining usage intention. Additionally, perceived risk negatively influences usage intention, while price sensitivity and perceived green value influence usage intention through attitudes. The findings also suggest that the impact of perceived green value on usage intention is higher among females and varies among income groups.
JOURNAL OF TRANSPORT GEOGRAPHY
(2022)
Article
Economics
Minh Hieu Nguyen, Dorina Pojani, Thanh Chuong Nguyen, Thanh Tung Ha
Summary: The study reveals that the Covid-19 pandemic led to a decrease in active school travel rates among students, with parents' concerns, urbanization levels, and home-school distances influencing students' transportation modes, especially during the pandemic.
JOURNAL OF TRANSPORT GEOGRAPHY
(2021)
Article
Public, Environmental & Occupational Health
Nguyen Anh Thuy Tran, Ha Lan Anh Nguyen, Thi Bich Ha Nguyen, Quang Huy Nguyen, Thi Ngoc Lan Huynh, Dorina Pojani, Binh Nguyen Thi, Minh Hieu Nguyen
Summary: This study explores the health and safety issues facing delivery riders in Ho Chi Minh City, Vietnam, during the Covid-19 pandemic. The findings suggest that some riders are less consistent in adopting health and safety measures, which is associated with their gender, age, education level, vaccination status, financial pressure, and income loss. Job pressure, long working hours, and financial burdens contribute to risky traffic behaviors among riders. On the other hand, support from the company and co-workers helps improve riders' adherence to health prevention measures.
JOURNAL OF TRANSPORT & HEALTH
(2022)
Article
Economics
Duy Quy Nguyen-Phuoc, Nguyen An Ngoc Nguyen, Minh Hieu Nguyen, Ly Ngoc Thi Nguyen, Oscar Oviedo-Trespalacios
Summary: This study investigated the impact of job demands and resources on food delivery riders' compliance with road safety regulations. The results showed that job demands and resources have direct and indirect effects on compliance, and control variables such as age, gender, and income also influence road safety compliance.
TRANSPORTATION RESEARCH PART A-POLICY AND PRACTICE
(2022)
Article
Business
Binh Nguyen Thi, Thi Lan Anh Tran, Thi Thu Hien Tran, Thanh Thao Le, Phan Nhat Hang Tran, Minh Hieu Nguyen
Summary: This study investigates the factors influencing the online shopping continuance intention of Generation Y and Z during the new normal. The results reveal that perceived usefulness, perceived ease of use, satisfaction, and environmental awareness are facilitators of repurchase intention, while perceived risks of online shopping act as a barrier. Personalization has indirect effects through perceived usefulness, perceived ease of use, and satisfaction. The risk of COVID-19 is not a predictor of online repurchase intention.
COGENT BUSINESS & MANAGEMENT
(2022)
Article
Engineering, Industrial
Quy Nguyen-Phuoc Duy, Ngoc Thi Nguyen Ly, Ngoc Su Diep, Nguyen Minh Hieu, Oscar Oviedo-Trespalacios
Summary: Food delivery riders are at a higher risk of being involved in road crashes, likely due to the working conditions in the gig economy. Research is necessary to understand the safety issues faced by this vulnerable group and identify ways to address them.
Article
Green & Sustainable Science & Technology
Thanh Tung Ha, Thanh Chuong Nguyen, Sy Sua Tu, Minh Hieu Nguyen
Summary: This study aims to investigate the factors influencing motorcyclists' intention to use electric vehicles in motorcycle-dependent countries like Vietnam. The results show that safety and environmental concerns are significant push factors, while perceived usefulness, perceived ease of use, and financial incentives are significant pull factors. Both push and pull factors significantly contribute to the adoption intention, with pull factors having a stronger effect. Knowledge does not directly affect the intention, but it moderates the relationship between the pull factor and the intention.
Article
Public, Environmental & Occupational Health
Minh Hieu Nguyen, Dorina Pojani, Duy Quy Nguyen-Phuoc
Summary: Illegal motorcycle riding is common among teenagers in many low and middle-income countries, leading to accidents and disastrous consequences. This study aims to understand the factors behind this behavior and why parents permit it. Findings reveal that 61% of teenagers engage in illegal motorcycle riding, enabled by parents who view motorcycles as useful and easy to use. Social acceptance and perception of weak road rule enforcement also contribute, while perceived crash risk acts as a deterrent.
JOURNAL OF TRANSPORT & HEALTH
(2023)
Article
Transportation
Minh Hieu Nguyen, Duy Quy Nguyen-Phuoc, Lester W. Johnson
Summary: This study explores the factors influencing parents' intention to permit their children to use an e-bike. It found that parents' perceived usefulness of e-bikes has a direct positive effect on their permission intention, while perceived risk has a direct negative effect. Gender, household income, and living area also moderate the relationship between antecedents and permission intention.
TRAVEL BEHAVIOUR AND SOCIETY
(2023)
Article
Public, Environmental & Occupational Health
Thanh Chuong Nguyen, Minh Hieu Nguyen, Jimmy Armoogum, Thanh Tung Ha
Summary: This study investigates the determinants of bus crash severity in developing countries using data from Hanoi, Vietnam. Factors such as vehicle type, weather, time, road conditions, and road characteristics were found to impact crash severity. Policy recommendations include improving road conditions, promoting pedestrian infrastructure, increasing driver awareness of high-risk situations, enhancing bus service quality, and recording bus-related crashes.