Article
Materials Science, Multidisciplinary
Yongfeng Ding, P. P. Camanho, Arlindo Silva
Summary: This article proposes a new quantitative method, the D-index, to evaluate the departure of a given fibre arrangement from the completely spatial randomness pattern. An explicit model is created to investigate the correlation between the D-index and the degree of randomness. The derivation of the D-index with respect to the degree of randomness is analyzed under different fibre volume fractions.
JOURNAL OF MATERIALS RESEARCH AND TECHNOLOGY-JMR&T
(2023)
Article
Multidisciplinary Sciences
Bryan A. Dawkins, Trang T. Le, Brett A. McKinney
Summary: The performance of nearest-neighbor feature selection and prediction methods is influenced by neighborhood computation metrics and data distribution properties. Recent work has focused on improving algorithms through new estimation methods and metrics, but little attention has been paid to the distributional properties of pairwise distances. Analytical formulas for mean and variance of pairwise distances for different data types and metrics have been derived, providing insights into the distance properties commonly used in nearest-neighbor methods.
Article
Telecommunications
Kaushlendra Pandey, Abhishek K. Gupta
Summary: This letter discusses the derivation of cumulative distribution functions for the kth contact distance and nearest neighbor distance of the n-dimensional Matern cluster process, as well as a new approach based on the relationship between probability mass and generating functions. The analysis is validated through numerical simulations and provides insights into the impact of clustering on performance, with applications in cellular networks and D2D networks highlighted for further study.
IEEE COMMUNICATIONS LETTERS
(2021)
Article
Materials Science, Multidisciplinary
R. K. Everett, M. Zupan
Summary: This study investigates the two-dimensional mean nearest neighbor distance and other distance distribution metrics in uniform and unimodal normal distributions of random sequential addition hard disc computer-generated patterns. An accurate method for estimating the mean nearest neighbor distance is proposed and the changes in distance are discussed. The findings have implications for higher dimension problems and multi-modal distributions.
MATERIALS TODAY COMMUNICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Ke Yuan, Daoming Yu, Jingkai Feng, Longwei Yang, Chunfu Jia, Yiwang Huang
Summary: This article proposes a new ensemble learning-based model for analyzing and identifying encryption algorithms in cryptographic systems, achieving higher classification accuracy.
PEERJ COMPUTER SCIENCE
(2022)
Article
Computer Science, Information Systems
Min Li, Ting Wu, Weitao Li, Chun Wang, Wen Dai, Xu Su, Yuanyuan Zhao
Summary: This paper proposes a new terrain skeleton that includes three types of terrain skeleton points and two types of terrain skeleton lines, which can reflect the structure and characteristics of the terrain. By selecting three analysis indicators, the development of landforms can be described, and differences in different areas of the same landform can be discovered.
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION
(2022)
Article
Computer Science, Interdisciplinary Applications
Sherif Ahmed Abu El-Magd, Sk Ajim Ali, Quoc Bao Pham
Summary: This study implemented different machine learning algorithms to predict landslide events in the Jabal Farasan area of northwest Jeddah, Saudi Arabia, identifying factors such as mining activities and high slope angles as contributors to landslide susceptibility. The model accuracy for predicting landslide occurrence locations using remote sensing data ranged between 86% and 89%. The generated landslide susceptibility map aims to assist in hazard management and control for natural disasters in the region.
EARTH SCIENCE INFORMATICS
(2021)
Review
Computer Science, Interdisciplinary Applications
Satyanarayana Murthy Nimmagadda, Sowmya Sree Agasthi, Abbas Shai, Dimple Kavitha Raj Khandavalli, Janaki Ram Vatti
Summary: Kidneys are essential for maintaining body balance, and machine learning techniques are valuable for accurate detection and analysis of chronic kidney disease. The performance of different algorithms varies in detecting CKD.
ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING
(2023)
Article
Environmental Sciences
Sumayh S. Aljameel, Dina A. Alabbad, Norah A. Alzahrani, Shouq M. Alqarni, Fatimah A. Alamoudi, Lana M. Babili, Somiah K. Aljaafary, Fatima M. Alshamrani
Summary: This study collected Arabic COVID-19 related tweets and used machine learning models to predict the awareness of precautionary procedures in five main regions in Saudi Arabia. The results showed that the south region had the highest awareness towards COVID-19 containment measures, while the middle region had the lowest.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2021)
Article
Social Sciences, Mathematical Methods
Mehdi Dagdoug, Camelia Goga, David Haziza
Summary: Nonparametric and machine learning methods are flexible and accurate for prediction, especially for data sets with a large number of predictors and complex structures. Machine learning procedures can be a useful alternative to traditional imputation procedures in the presence of item nonresponse, generating imputed values for the estimation of study parameters.
JOURNAL OF SURVEY STATISTICS AND METHODOLOGY
(2023)