Machine learning
Researchers at Guizhou Normal University have made significant strides in addressing the challenges of data privacy and security in cloud computing. By proposing Einocchio, a verifiable computation scheme that combines Pinocchio system with homomorphic encryption, they have achieved IND-CPA security and demonstrated its effectiveness in cloud computing environments. This breakthrough
Machine learning
A new study from the Civil Aviation University of China has introduced a data-driven psychophysiological model that uses machine learning algorithms to evaluate pilots' mental workload. The research, funded by the National Natural Science Foundation of China and Fundamental Research Funds for the Central Universities, involved a flight-simulation experiment
Machine learning
Researchers from North China University of Technology have proposed a novel data-driven methodology for calculating insurance premiums of Beijing-based insurance companies. This study utilized machine learning algorithms to predict healthcare insurance costs, incorporating 11 features such as age and weight as input variables. The research found that age was the
Machine learning
Research conducted at Daejeon University in South Korea has utilized machine learning algorithms and oversampling techniques to classify seafloor sediments, achieving high accuracy rates. The study, funded by the National Research Foundation of Korea and the Ministry of Science & ICT, utilized field data including porosity, S- and P-wave velocities,
Machine learning
A new study from the Department of General Surgery at Mianyang Hospital of Traditional Chinese Medicine, People's Republic of China, has developed and validated a machine learning-based nomogram for preoperative prediction of laparoscopic surgical difficulty in gallstone patients. The research aimed to improve surgical outcomes by incorporating key
Machine learning
A team of researchers from Leshan Normal University in China has developed a gesture recognition system that uses advanced deep learning architectures and machine learning algorithms to recognize complex gestures with high precision. The system, which has been tested on a diverse dataset, achieves an accuracy of 95.4% and
Machine learning
Current research in machine learning has led to the development of effective ecological security patterns in regional ecosystems. A study published in the Journal of Cleaner Production utilized machine learning algorithms, circuit theory, and the ecological protection red line method to construct an ecological security pattern in Kunming City, China.
Machine learning
Accurately predicting crop traits like flowering time is crucial for modern rice breeding, especially under changing environmental conditions. Researchers at Nanjing Agricultural University have employed a novel integrative modeling approach that combines process-based rice growth models, genome-wide association studies, single nucleotide polymorphisms, and machine learning algorithms to predict rice flowering
Machine learning
Researchers in Moscow, Russia have identified the challenges in creating training and testing datasets for machine learning algorithms to measure linear dimensions of organs, including the spleen. The study found that relying solely on mathematical statistics or machine learning methods is inadequate to detect outliers and anomalies in such datasets.
Machine learning
A recent research study published in PLOS Computational Biology has shed new light on the challenges of identifying novel and functional RNA structures. The study, conducted by researchers at New York University (NYU), introduces a computational topology-based approach with unsupervised machine-learning algorithms to estimate the database size and content of
Machine learning
Researchers at Daffodil International University have established a comprehensive database of knitted fabrics, incorporating various parameters related to yarn types, machinery, and fabric characteristics. This database was created by processing raw data from a knitting factory with pre-processing techniques using domain knowledge and the Python programming language. The dataset comprises
Machine learning
Researchers at the University of Tehran have developed a deep learning-based solution for designing optimal contracts in federated learning, a method that enables collaborative model training without sharing raw data. The study proposes a novel application of Deep Neural Networks (DNNs) to achieve near-optimal contracts, which can incentivize vehicle participation
