Article
Computer Science, Artificial Intelligence
Luis Ferreira, Andre Pilastri, Filipe Romano, Paulo Cortez
Summary: This paper explores the application of automated machine learning (AutoML) models in predictive maintenance (PdM) and compares the performance of different AutoML technologies through benchmark experiments. A novel AutoML approach, AutoOneClass, is proposed based on one-class learning, employing three types of learners for model search. The results show that the predictive performance of different AutoML technologies is similar, with supervised AutoGluon achieving the best results. Furthermore, the best results obtained with supervised AutoML and AutoOneClass are competitive compared to traditional manual modeling approaches.
APPLIED SOFT COMPUTING
(2022)
Article
Energy & Fuels
Alexander Vargas-Almeida, Miguel Angel Olivares-Robles, Andres Alfonso Andrade-Vallejo
Summary: This paper contributes to the study of geometric optimization of thermoelectric generators (TEGs) by combining the technique of reduced variables and supervised machine learning. It determines the architecture of TEGs by calculating cross-sectional area and length of the legs and uses a supervised machine learning algorithm to predict thermoelectric properties and maximum electrical power. This method allows for approximate results and optimal values of leg parameters, providing more realistic TEG models.
Article
Computer Science, Artificial Intelligence
Lin Li, Turghun Tayir, Yifeng Han, Xiaohui Tao, Juan D. Velasquez
Summary: This article proposes an effective multimodality information fusion method for automated machine translation based on semi-supervised learning. The method combines multimodality information, texts, and images through a multimodal attention network, improving the accuracy of machine translation.
INFORMATION FUSION
(2023)
Article
Public, Environmental & Occupational Health
Corine S. Meppelink, Hanneke Hendriks, Damian Trilling, Julia C. M. van Weert, Anqi Shao, Eline S. Smit
Summary: The study found that supervised machine learning can effectively classify health-related webpages as 'reliable' or 'unreliable' in an automated way, providing a promising tool for people to evaluate the credibility of online information.
PATIENT EDUCATION AND COUNSELING
(2021)
Article
Biology
Xiaohui Guo, Michael R. Lin, Asma Azizi, Lucas P. Saldyt, Yun Kang, Theodore P. Pavlic, Jennifer H. Fewell
Summary: Alarm signal propagation in ant colonies provides insights for analyzing information flow in natural systems and other social animals. Researchers developed a method to track alarm spread in a group of harvester ants and used a random forest regression model to assess individual alarm behavior. This approach allows analysis of spatio-temporal patterns in alarm signal propagation and integration of individual and collective alarm response.
PROCEEDINGS OF THE ROYAL SOCIETY B-BIOLOGICAL SCIENCES
(2022)
Article
Computer Science, Information Systems
Yoram Timmerman, Antoon Bronselaer
Summary: This study proposes a new automated procedure for monitoring online news accuracy by observing changes made to online articles and detecting errors that are corrected. Classification models are built to detect objective, subjective, and linguistic errors in online news updates. The models are evaluated using a large dataset collected from six Flemish online newspapers. Results show that different learning algorithms and language models have slight performance differences, with F2-scores of 0.45, 0.25, and 0.80 obtained for objective, subjective, and linguistic error classification respectively.
INFORMATION PROCESSING & MANAGEMENT
(2022)
Article
Chemistry, Applied
Maryam Mousavizadegan, Morteza Hosseini, Mahsa N. Sheikholeslami, Yalda Hamidipanah, Mohammad Reza Ganjali
Summary: This study presents a rapid detection method using machine learning algorithms for the analysis of tetracycline in water and milk samples. The developed models achieved high accuracy in identifying the presence of tetracycline, demonstrating the potential of machine learning in the development of rapid sensors for food analytes.
Article
Computer Science, Information Systems
Mohamed Omar, Mohamed H. Bakr, Ali Emadi
Summary: This research introduces an advanced optimization method for the geometry optimization of switched reluctance motor (SRM) using a supervised learning algorithm. The method considers both the static and dynamic characteristics of the machine. Back-propagation neural network (BPNN) is used to model the SRM, and finite element analysis and MATLAB simulations are used to verify the effectiveness of the proposed designs.
Article
Ecology
Xinghu Qin, Thomas Ryan Lock, Robert L. Kallenbach
Summary: Different supervised learning approaches (DAPC, LDAKPC, LFDA, LFDAKPC, and KLFDA) were tested for population genetic structure inference, with KLFDA showing the best performance in discriminating population structures. These methods successfully preserved the global genetic structure of populations while capturing continuous genetic gradients within populations.
METHODS IN ECOLOGY AND EVOLUTION
(2022)
Review
Engineering, Aerospace
Maksim Shirobokov, Sergey Trofimov, Mikhail Ovchinnikov
Summary: This paper presents a survey on the application of machine learning techniques in spacecraft control design, covering various areas such as trajectory design and controller synthesis. The works are categorized into supervised learning and reinforcement learning, with the latter further divided into direct and value-based approaches.
Review
Computer Science, Interdisciplinary Applications
Veenu Rani, Syed Tufael Nabi, Munish Kumar, Ajay Mittal, Krishan Kumar
Summary: Machine learning has made significant advances in image processing. Supervised learning relies on labeled data, while unsupervised learning learns from unlabeled data. Self-supervised learning is a type of unsupervised learning that enhances computer vision tasks. This review article provides an in-depth exploration of self-supervised learning and its applications, discussing terms, learning types, and challenges encountered in the process.
ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING
(2023)
Article
Construction & Building Technology
Zhaoyang Luo, Cheng Sun, Qi Dong, Xuanning Qi
Summary: This paper comprehensively analyzes the importance of control-related variables for machine learning-assisted automated louvers in open-plan offices and provides an efficient but compact set of control variables.
BUILDING AND ENVIRONMENT
(2022)
Article
Computer Science, Artificial Intelligence
Felix Mohr, Jan N. van Rijn
Summary: Traditional cross-validation methods have drawbacks in terms of speed and providing limited information on the learning process. This article introduces a new validation approach called learning curve cross-validation (LCCV) which iteratively increases the training instances. Experiments on 75 datasets show that LCCV achieves comparable performance to 5/10-fold CV while significantly reducing runtime (median runtime reductions of over 50%) with a maximum difference of 2.5% in performance compared to CV.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Urology & Nephrology
Craig Peter Coorey, Ankit Sharma, Samuel Muller, Jean Yee Hwa Yang
Summary: This mini-review examines the potential applications of machine learning methods in key stages of a kidney transplant recipient's journey, including donor selection, personalization of immunosuppression, and prediction of post-transplantation events. Both unsupervised and supervised machine learning methods are discussed, as well as the challenges associated with these approaches.
KIDNEY INTERNATIONAL
(2021)
Article
Chemistry, Multidisciplinary
Yonghua Wei, Jin Wu, Yixuan Wu, Hongjiang Liu, Fanqiang Meng, Qiqi Liu, Adam C. Midgley, Xiangyun Zhang, Tianyi Qi, Helong Kang, Rui Chen, Deling Kong, Jie Zhuang, Xiyun Yan, Xinglu Huang
Summary: This study demonstrates a novel method using machine learning algorithms to understand the enzyme-like activity of nanozymes, enabling classification and quantitative prediction. It offers a promising strategy to predict and design desirable nanozymes.
ADVANCED MATERIALS
(2022)