Machine learning
Researchers from Instituto de Investigacao e Inovacao em Saude (i3S) have developed a machine learning algorithm to improve mutation modelling in forensic genetics. The study analyzed father-son segregation data for 28 Y-chromosomal short tandem repeats (Y-STRs) and found that logistic regression analyses could estimate marker-specific mutation rates depending on paternal
Machine learning
Researchers at Sanming University have developed a novel machine learning method to improve the reliability of fault diagnosis in nuclear power plant gear lubrication systems. This integrated data-driven approach combines sensor measurements with advanced machine learning algorithms to detect faults more accurately. Key Takeaways: * The proposed method, called Integrated Data-Driven
Machine learning
Research by King Abdullah University of Science and Technology has revealed that Deep Neural Networks (DNNs) exhibit superior performance in solving complex problems such as computer vision and natural language processing compared to classic machine learning techniques. The study, which focuses on Convolutional Neural Networks (CNNs) as an example of
Machine learning
Researchers from the Harbin Institute of Technology have proposed a new framework, RefineCatDiff, for high-quality medical image segmentation. According to the study, existing convolutional neural network-based or vision transformer-based segmentation models often struggle to produce accurate masks when dealing with complex images. However, diffusion models are particularly effective at capturing
Machine learning
Researchers from the Beijing University of Chemical Technology have made significant strides in applying artificial intelligence (AI) to advance polymer science and engineering. The study, published in the Journal of Chemical Information and Modeling, explores the influence of polymer composition and sequence structure on determining the physical properties of polymers.
Machine learning
A recent investigation by the Computer Science Department has introduced a novel hybrid multiobjective wrapper method for feature selection, which combines the technique for Order Preference by Similarity to Ideal Solution (TOPSIS) with a multiobjective Crow Search Algorithm (CSA) to improve classification accuracy. The study aims to address the challenges
Machine learning
A new research study from Leipzig University Hospital has identified the potential of using Graph Neural Networks (GNNs) in clinical data analysis, particularly in diagnosing sepsis from blood count data. The study, published in PLOS ONE, finds that GNNs can achieve superior results compared to traditional machine learning algorithms when
Machine learning
Research from the University of Gondar in Ethiopia has analyzed data from 147,466 men to identify predictors of tobacco use in Sub-Saharan Africa. The study utilized machine learning algorithms to determine the key factors influencing tobacco use among men in the region between 2018 and 2023. The findings highlight
Machine learning
Researchers at the Vellore Institute of Technology in Tamil Nadu, India, have conducted a comprehensive review of the Deep Q-Network (DQN) algorithm, a type of machine learning that enables sequential decision-making. The team focused on the pivotal aspects of DQN, including the deep neural network and experience replay, and explored
Machine learning
Researchers have developed new dynamic programming algorithms within the NUPACK software suite, enabling the analysis of equilibrium base-pairing properties for complex systems containing mixed-material nucleic acid strands. Currently, calculations are limited to single-material systems, but the new algorithms can analyze mixed-material systems, which are critical for modern applications in vitro,
Machine learning
A new patent granted to John Keith Bentley and his team from Texas Instruments Incorporated has unveiled a novel method for detecting end effector bumps in robots using contact shock transients sensing units. This technology is expected to significantly impact the manufacturing industry, particularly in the semiconductor sector, where precise
Machine learning
A team of researchers from the United Arab Emirates University has made significant strides in forecasting energy commodity prices using Artificial Intelligence-based models. The study, which covered a comprehensive daily dataset from 2006 to 2023, found that machine learning models incorporating global energy transition factors perform better than traditional ANN
