Article
Business
Byron Graham, Karen Bonner
Summary: Traditional regression-based approaches cannot fully capture the determinants of entrepreneurship, but using machine learning reveals that individual entrepreneurial self-efficacy and networks are the dominant factors. Factors such as cultural perceptions play a relatively unimportant role in determining entrepreneurship.
JOURNAL OF BUSINESS RESEARCH
(2022)
Article
Computer Science, Artificial Intelligence
Juan Manuel Davila Delgado, Lukumon Oyedele
Summary: The study demonstrates that using autoencoders for data augmentation can significantly improve cost prediction accuracy for underground and overhead power transmission projects, with an average improvement of 7.2% and 11.5% respectively. Variational autoencoders provide more robust results and better capture non-linear correlations in the datasets.
APPLIED SOFT COMPUTING
(2021)
Article
Multidisciplinary Sciences
Weiyi Xia, Masahiro Sakurai, Balamurugan Balasubramanian, Timothy Liao, Renhai Wang, Chao Zhang, Huaijun Sun, Kai-Ming Ho, James R. Chelikowsky, David J. Sellmyer, Cai-Zhuang Wang
Summary: Rare earth elements are widely used in magnets, but finding alternative magnetic materials without rare earth elements has been a challenge. In this study, a machine learning-guided approach integrating genetic algorithms, first-principles calculations, and experimental synthesis was used to discover and synthesize a rare earth-free magnetic compound, Fe3CoB2, which exhibits excellent magnetic properties suitable for permanent-magnet applications.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2022)
Article
Engineering, Electrical & Electronic
Theodoros Moschos, Stelios Simos, Christos Vagionas, Theonitsa Alexoudi, Nikos Pleros
Summary: In this paper, the experimental demonstration of a complete all-optical 2-bit Address Look-Up (AL) table is reported for the first time. The performance of the table is validated for different settings, and a fully integrated optical CAM cell is also demonstrated, indicating a possible integration roadmap for integrated optical ALs.
JOURNAL OF LIGHTWAVE TECHNOLOGY
(2022)
Article
Materials Science, Multidisciplinary
Davoud Hejazi, Neda Kari Rezapour, John Ferrier, Sarah Ostadabbas, Swastik Kar
Summary: This article introduces a device called A-Eye, which can accurately recognize and reproduce colors without the need for dispersion. By using multiple transmissive windows with unique spectral features, A-Eye can transmit and modify different colored lights, which are used to identify and reproduce various colors.
Article
Environmental Sciences
Fan Meng, Yichen Yao, Zhibin Wang, Shiqiu Peng, Danya Xu, Tao Song
Summary: This study proposes a machine learning approach for probabilistic forecasting of tropical cyclone intensity. Previous studies cannot directly characterize the uncertainty in TC forecasting and suffer from computational effort issues. This study introduces a new method of evaluating the forecast without this uncertainty through the forecast distribution. The model outperforms current operational models and provides reliable probabilistic forecasts critical for disaster warnings.
ENVIRONMENTAL RESEARCH LETTERS
(2023)
Article
Nanoscience & Nanotechnology
Bin Wang, Weijia Li, Qi Lu, Yueying Zhang, Hao Yu, Lingchu Huang, Tong Wang, Xishuang Liang, Fengmin Liu, Fangmeng Liu, Peng Sun, Geyu Lu
Summary: This study utilized machine learning to identify potential sensitive electrode materials for NO2 detection, with over 400 materials selected from 8000 candidates for further experimental verification and testing.
ACS APPLIED MATERIALS & INTERFACES
(2021)
Article
Engineering, Chemical
Arun Muthukkumaran, Shrayas Raghunathan, Arjun Ravichandran, Raghunathan Rengaswamy
Summary: With the increasing volume of scientific literature, a natural language processing model was used to select perovskite materials for electrocatalytic applications. Word embeddings for perovskite materials were obtained from the model and used to rank material suitability. A novel methodology was developed to generate embeddings for newly designed materials, improving the accuracy of regression models for predicting electrocatalytic activity.
Article
Geriatrics & Gerontology
Yurun Cai, Suzanne G. Leveille, Olga Andreeva, Ling Shi, Ping Chen, Tongjian You
Summary: This study aimed to describe fall circumstances among older adults by analyzing quantitative data and using qualitative analysis methods. The results showed that 64% of the participants experienced at least one fall during the 4-year follow-up, with more falls occurring indoors than outdoors. Common activities during the falls included walking, standing, and going down stairs, and the most commonly reported causes were slip or trip and inappropriate footwear.
JOURNALS OF GERONTOLOGY SERIES A-BIOLOGICAL SCIENCES AND MEDICAL SCIENCES
(2023)
Article
Chemistry, Physical
Zuqiang Qiao, Shengzhi Dong, Qing Li, Xiangming Lu, Renjie Chen, Shuai Guo, Aru Yan, Wei Li
Summary: By introducing composition and process features as input, we successfully built a sintered NdFeB performance prediction model. The model has good generalization capability, high accuracy, and sound interpretation compared to previous work. Using the Shapley additive interpretation method, we solved the unexplainable problem of ML models and evaluated the contribution of features to the results. Our work is expected to accelerate performance screening and material development of sintered NdFeB.
JOURNAL OF ALLOYS AND COMPOUNDS
(2023)
Review
Materials Science, Multidisciplinary
Yongfei Juan, Yongbing Dai, Yang Yang, Jiao Zhang
Summary: The use of machine learning methods in materials science has seen significant progress in recent years, bringing about a profound revolution in society and greatly advancing scientific development. This review provides an overview of the application of machine learning in materials science research, emphasizing the main ideas, basic procedures, and classification and comparison of commonly used algorithms.
JOURNAL OF MATERIALS SCIENCE & TECHNOLOGY
(2021)
Article
Psychology, Multidisciplinary
Patricio E. Ramirez-Correa, F. Javier Rondan-Cataluna, Jorge Arenas-Gaitan, Elizabeth E. Grandon, Jorge L. Alfaro-Perez, Muriel Ramirez-Santana
Summary: This study analyzes the acceptance of social network sites among elder people in Chile, identifying different predictors across age groups and genders. The high heterogeneity among older people suggests that they should not be treated as a uniform group of social network site users.
FRONTIERS IN PSYCHOLOGY
(2021)
Article
Business
Marica Valente
Summary: Using machine learning methods in a quasi-experimental setting, this study examines the heterogeneous effects of implementing waste prices on waste demands and municipal costs. The research finds that waste demands are nonlinear and demonstrate varying elasticities at different price levels. Moreover, after three years of adoption, the policy leads to significant reductions in total waste, resulting in decreased waste management costs for all municipalities.
JOURNAL OF ENVIRONMENTAL ECONOMICS AND MANAGEMENT
(2023)
Review
Materials Science, Multidisciplinary
Chen Li, Kun Zheng
Summary: As a tool for implementing data-intensive scientific research methods, machine learning (ML) can effectively reduce the research and development (R&D) cycle of new materials by half or more. ML shows great potential in combination with other scientific research technologies, particularly in the processing and classification of large amounts of material data from theoretical calculations and experimental characterizations. It is crucial to systematically understand the research ideas of material informatics to speed up the exploration of new materials.
Article
Engineering, Environmental
Wenhao Zhao, Jin Ma, Qiyuan Liu, Yajing Qu, Lei Dou, Huading Shi, Yi Sun, Haiyan Chen, Yuxin Tian, Fengchang Wu
Summary: In traditional soil heavy metal pollution assessment, spatial interpolation analysis is often carried out to evaluate the overall status of heavy metal pollution. However, machine learning algorithms often fail to capture the hierarchical heterogeneity of the study area. Therefore, this study introduced new spatial covariates based on interpolation techniques and demonstrated the improvement in prediction performance. The analysis also identified key factors affecting the spatial distribution of heavy metals and evaluated the pollution status in the Pearl River Delta region.
ENVIRONMENTAL SCIENCE & TECHNOLOGY
(2023)