Article
Engineering, Marine
Marvin F. Li, Patricia M. Glibert, Vyacheslav Lyubchich
Summary: The frequency, magnitude, and impact of harmful algal blooms (HABs) have increased globally. Machine learning algorithms, such as RVM and NB, show better abilities in predicting blooms. The importance of upwelling-favorable winds, onshore winds, and river flows in regulating blooms has been quantified.
JOURNAL OF MARINE SCIENCE AND ENGINEERING
(2021)
Article
Environmental Sciences
Chris G. Tzanis, Anastasios Alimissis, Ioannis Koutsogiannis
Summary: An essential aspect of environmental science is studying air quality using statistical methods and large datasets of climatic parameters. Monitoring networks in urban areas provide crucial pollutant data, which can be utilized through environmental statistics to develop continuous pollutant concentration surfaces. Mapping air quality can guide policymakers and researchers in minimizing adverse effects, with spatial interpolation methods used to estimate air quality distributions.
Article
Environmental Sciences
Ahmed Mohsen, Timea Kiss, Ferenc Kovacs
Summary: Despite the lack of attention to riverine litter, this study aims to detect riverine litter using middle-scale multispectral satellite images and machine learning. The study focuses on the Tisza River in Hungary and utilizes very high-resolution images obtained from the Google Earth database. Five supervised machine-learning algorithms were trained and validated using the litter spots identified in the images, but their performance on larger unseen data and different litter sizes was only medium to poor. The study provides preliminary insights into the automatic detection of riverine litter and highlights the need for further research with a larger dataset and finer spatial resolution images.
ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH
(2023)
Article
Agronomy
Yingpeng Fu, Hongjian Liao, Longlong Lv
Summary: This study employed various methods to impute missing soil property data in UNSODA, with random forest and multiple imputation showing the best performance in explaining data variability. The fluctuation of geological features was within acceptable ranges after imputation, suggesting the reliability of the RF and MI methods for imputing missing data in UNSODA.
Article
Computer Science, Artificial Intelligence
Feras Al-Obeidat, Alvaro Rocha, Maryam Akram, Saad Razzaq, Fahad Maqbool
Summary: Cancer, a genetic disease with high mortality, can be detected early to reduce death rates. Oncogenomics utilizes RNA sequencing and gene expression profiling to identify cancer-related genes. The CDRGI approach, based on a Discrete Filtering technique, shows promise in identifying relevant genes and detecting cancer.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Engineering, Mechanical
Timur Karimov, Olga Druzhina, Artur Karimov, Aleksandra Tutueva, Valerii Ostrovskii, Vyacheslav Rybin, Denis Butusov
Summary: A metal detector sensor based on chaotic oscillators with continuous response to target distance and material was developed and compared with a sensor based on harmonic oscillators, showing at least a 20% larger operating range for the chaotic sensor. Data processing techniques were modified for accurate measurement results from chaotic data series.
NONLINEAR DYNAMICS
(2022)
Article
Computer Science, Information Systems
Monalisa Jena, Satchidananda Dehuri
Summary: This work presents an integrated framework combining rule based decision tree and Support Vector Machine for imputation of missing values and prediction of class label, outperforming other methods in terms of performance. Additionally, a new variant of kNN, approximated kNN, is proposed to reduce computational time without compromising classification accuracy.
Article
Green & Sustainable Science & Technology
Divya Biligere Shivanna, Thompson Stephan, Fadi Al-Turjman, Manjur Kolhar, Sinem Alturjman
Summary: The paper introduces an image-based computer-aided automatic autoimmune disease diagnosis method, which effectively improves the accuracy of autoimmune disease diagnosis. Through the Multistage Classification Scheme, combined with various basic modules, the recognition and classification of autoantibodies in serum are performed.
Article
Geochemistry & Geophysics
Xintao Chai, Genyang Tang, Shangxu Wang, Kai Lin, Ronghua Peng
Summary: The incompleteness of seismic data poses a longstanding challenge in exploration seismology, which researchers have addressed by extending deep learning techniques from 2D to 3D data reconstruction. The developed 3D CNN reconstruction method has shown superior performance compared to 2D CNN, particularly in handling seismic data with high missing percentages or large gaps. This approach offers significant improvements in data fidelity and efficiency over existing rank-reduction-based methods.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Computer Science, Artificial Intelligence
Esther-Lydia Silva-Ramirez, Juan-Francisco Cabrera-Sanchez
Summary: Data imputation is a common problem in today's applications, and various techniques have been proposed to address it, from statistical methods to machine learning models. The CANFIS-ART model proposed in this paper outperforms other state-of-the-art techniques and demonstrates a higher level of generalization capability, enhancing the data quality in databases with missing values.
NEURAL COMPUTING & APPLICATIONS
(2021)
Article
Nanoscience & Nanotechnology
H. R. Jafarian, Yan Zeng, S. H. Mousavi Anijdan, A. R. Eivani
Summary: The ARB process can control the microstructural evolution and precipitation characteristics of the Al-Ag-Sc alloy, with different cycles leading to varying grain structure evolution. Artificial aging at different temperatures affects the hardness and microstructural evolution of the alloy.
MATERIALS SCIENCE AND ENGINEERING A-STRUCTURAL MATERIALS PROPERTIES MICROSTRUCTURE AND PROCESSING
(2021)
Article
Environmental Sciences
Kadir Gezici, Selim Senguel
Summary: Considering the importance of limited natural resources, accurate recording and evaluation of temperature data were conducted using artificial neural network (ANN), support vector regression (SVR), and regression tree (RT) methods. Various statistical evaluation criteria and the Taylor diagram were used to compare the output values generated by different machine learning methods. ANN6, ANN12, medium gaussian SVR, and linear SVR were found to be the most suitable methods, particularly for estimating temperatures at high and low extremes. All the models and network architectures employed achieved successful results (NSE-R-2 >0.90). Some deviations in estimation results were observed in mountainous areas with heavy snowfall, particularly in the range of -1 to 5 degrees Celsius, where fresh snowfall affects heat emission from the ground. The increase in the number of layers positively affected the accuracy of estimation in models with high neuron counts.
ENVIRONMENTAL MONITORING AND ASSESSMENT
(2023)
Article
Water Resources
Muhammad Adnan Khan, Juergen Stamm
Summary: This study evaluated the performance and hydrologic utility of four different satellite precipitation datasets (SPDs) for predicting daily streamflow and SL in the mountainous Upper Jhelum River Basin, Pakistan. The results showed that the GPM dataset had the highest performance, and the SWAT-RF model performed better than other models in simulating streamflow. The SWAT-ANN model showed the best performance in simulating SL. Hydrological coupled SCMs based on SPDs can be effective in simulating hydrological parameters in complex terrain with limited gauge network density.
JOURNAL OF WATER AND CLIMATE CHANGE
(2023)
Article
Computer Science, Artificial Intelligence
Dalwinder Singh, Birmohan Singh
Summary: Feature weighting is a well-known approach for improving machine learning algorithm performance, but the sensitivity of algorithms to weighting has not been explored in depth. This study empirically assesses the sensitivity of four popular algorithms to changes in the feature space, comparing performance with weighted and unweighted features to identify the best algorithms.
PATTERN ANALYSIS AND APPLICATIONS
(2022)
Article
Computer Science, Interdisciplinary Applications
Ali Daher, Hassan Al Sakka, Khaled Chaaban
Summary: An accurate and low complexity model for precipitation estimation is essential for monitoring hydrological and water resource applications. In this study, a supervised learning algorithm using a single-layer neural network (the perceptron) was proposed as an alternative method for rainfall estimation. Experimental results demonstrated that this machine learning approach outperformed the R-k-based method. This promising alternative method could significantly improve the efficiency of various applications, such as real-time urban flood risk management.
JOURNAL OF HYDROINFORMATICS
(2023)
Article
Engineering, Civil
Ramesh S. Teegavarapu, Aneesh Goly
WATER RESOURCES MANAGEMENT
(2018)
Article
Environmental Sciences
Sina Borzooei, Ramesh Teegavarapu, Soroush Abolfathi, Youri Amerlinck, Ingmar Nopens, Maria Chiara Zanetti
WATER AIR AND SOIL POLLUTION
(2019)
Article
Meteorology & Atmospheric Sciences
Subash Yeggina, Ramesh S. Teegavarapu, Sekhar Muddu
INTERNATIONAL JOURNAL OF CLIMATOLOGY
(2019)
Article
Environmental Sciences
Sina Borzooei, Gisele H. B. Miranda, Ramesh Teegavarapu, Gerardo Scibilia, Lorenza Meucci, Maria Chiara Zanetti
JOURNAL OF ENVIRONMENTAL MANAGEMENT
(2019)
Article
Meteorology & Atmospheric Sciences
Ari D. Preston, Henry E. Fuelberg, Mary C. Barth
JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES
(2019)
Article
Water Resources
Fahad Khan Khadim, Ramesh S. V. Teegavarapu
HYDROLOGICAL SCIENCES JOURNAL-JOURNAL DES SCIENCES HYDROLOGIQUES
(2020)
Article
Engineering, Civil
Ramesh S. V. Teegavarapu
JOURNAL OF HYDROLOGY
(2020)
Article
Environmental Sciences
Hao Chen, Ramesh S. Teegavarapu
Article
Meteorology & Atmospheric Sciences
Subash Yeggina, Ramesh S. V. Teegavarapu, Sekhar Muddu
THEORETICAL AND APPLIED CLIMATOLOGY
(2020)
Article
Environmental Sciences
Marco Arrieta-Castro, Adriana Donado-Rodriguez, Guillermo J. Acuna, Fausto A. Canales, Ramesh S. V. Teegavarapu, Bartosz Kazmierczak
Article
Engineering, Civil
Aneesh Goly, Ramesh S. V. Teegavarapu
JOURNAL OF HYDROLOGIC ENGINEERING
(2020)
Article
Geosciences, Multidisciplinary
Caiyun Zhang, Hongbo Su, Tiantian Li, Weibo Liu, Diana Mitsova, Sudhagar Nagarajan, Ramesh Teegavarapu, Zhixiao Xie, Fred Bloetscher, Yan Yong
Summary: The study evaluated the feasibility of multiple linear regression and support vector machine techniques for predicting and mapping high water table in coastal landscapes, with SVM showing better predictive performance. Fine spatial resolution lidar-derived DEMs can effectively assist in predicting and mapping high water tables.
Proceedings Paper
Engineering, Environmental
Sina Borzooei, Ramesh Teegavarapu, Soroush Abolfathi, Youri Amerlinck, Ingmar Nopens, Maria Chiara Zanetti
NEW TRENDS IN URBAN DRAINAGE MODELLING, UDM 2018
(2019)
Proceedings Paper
Engineering, Civil
Hao Chen, Ramesh S. V. Teegavarapu
WORLD ENVIRONMENTAL AND WATER RESOURCES CONGRESS 2019: GROUNDWATER, SUSTAINABILITY, HYDRO-CLIMATE/CLIMATE CHANGE, AND ENVIRONMENTAL ENGINEERING
(2019)