Article
Food Science & Technology
Marco Camardo Leggieri, Marco Mazzoni, Sihem Fodil, Maurizio Moschini, Terenzio Bertuzzi, Aldo Prandini, Paola Battilani
Summary: The study evaluated the potential use of an electronic nose for rapid identification of mycotoxin contamination in maize samples and found that it had high accuracy in distinguishing contamination levels above or below legal limits. Artificial neural network (ANN) was the best method with 78% and 77% accuracy for AFB1 and FBs, respectively, indicating that the e-nose supported by ANN could be a rapid and reliable tool for mycotoxin detection.
Article
Computer Science, Information Systems
Chao-Yu Guo, Yi-Jyun Lin
Summary: A proper statistical approach is crucial in medical and health sciences to avoid erroneous conclusions. Different genders and drug interactions can have significant impacts on therapeutic effects and efficacy. This study proposes a new method called random interaction forest (RIF) based on random forest, which outperforms other models in considering interactions and making predictions under various scenarios.
Article
Chemistry, Analytical
Ahmad Elleathy, Faris Alhumaidan, Mohammed Alqahtani, Ahmed S. Almaiman, Amr M. Ragheb, Ahmed B. Ibrahim, Jameel Ali, Maged A. Esmail, Saleh A. Alshebeili
Summary: This paper presents an intruder detection system that uses a strain-based optical fiber Bragg grating (FBG), machine learning (ML), and adaptive thresholding to classify intruders in low signal-to-noise ratio conditions. The proposed method achieves an average accuracy of 99.17% when the optical signal-to-noise ratio (OSNR) is <0.5 dB.
Article
Green & Sustainable Science & Technology
Clariandys Rivera-Kempis, Leobardo Valera, Miguel A. Sastre-Castillo
Summary: This study uses predictive analysis from the machine learning approach to determine if an individual is an entrepreneur based on measures of 20 entrepreneurial competence attributes. The dynamic and complicated nature of entrepreneurship is taken into consideration, and a series of algorithms are utilized for analysis.
Article
Computer Science, Artificial Intelligence
Germanno Teles, Joel J. P. C. Rodrigues, Sergei A. Kozlov, Ricardo A. L. Rabelo, Victor Hugo C. Albuquerque
Summary: The study explores the expected recovery of credit using predictive models, finding that a simple logistic regression model can be extended to a multiple logistic regression model by integrating more prediction variables. However, obtaining multiple observations becomes increasingly difficult as the number of independent variables increases. By comparing logistic regression with linear regression, the study identifies the best model for predicting whether recovery is due in a credit operation.
Article
Geosciences, Multidisciplinary
Tom Horrocks, Eun-Jung Holden, Daniel Wedge, Chris Wijns
Summary: 3-D geochemical subsurface models are constructed by spatial interpolation of drill-core assays, but their accuracy is limited by the spatial sparsity of underlying data. Integrating collocated 3-D models could improve accuracy, but standard machine learning algorithms struggle to incorporate spatial autocorrelation.
GEOSCIENCE FRONTIERS
(2021)
Article
Biochemistry & Molecular Biology
Fatemeh Eshari, Fahime Momeni, Amirreza Faraj Nezhadi, Soudabeh Shemehsavar, Mehran Habibi-Rezaei
Summary: A novel machine-learning approach based on logistic regression (LR) is used to predict protein aggregation propensity (PAP) using a dataset of hexapeptides and eight physiochemical features. The LR model, combined with sequence and feature information, achieves high accuracy and outperforms other existing methods in PAP prediction.
INTERNATIONAL JOURNAL OF BIOLOGICAL MACROMOLECULES
(2023)
Article
Computer Science, Information Systems
Eman H. Alkhammash, Myriam Hadjouni, Ahmed M. Elshewey
Summary: Gender recognition by voice is an important research topic in speech processing, and recent research has focused on using ensemble learning to improve the accuracy of classifiers. This paper presents a stacked ensemble model for gender voice recognition, which combines multiple classifiers and achieves good results.
Review
Computer Science, Information Systems
Xuan Song, Xinyan Liu, Fei Liu, Chunting Wang
Summary: The study found that machine learning models perform similarly to logistic regression models in predicting acute kidney injury (AKI), but some machine learning models show exceptional performance, with gradient boosting models performing the best.
INTERNATIONAL JOURNAL OF MEDICAL INFORMATICS
(2021)
Article
Engineering, Multidisciplinary
Mohammad Ayoub Khan, Rijwan Khan, Fahad Algarni, Indrajeet Kumar, Akshika Choudhary, Aditi Srivastava
Summary: Research is crucial in addressing the COVID-19 pandemic, with technology, particularly machine learning models, playing a significant role in predicting cases and planning for future outbreaks. This study recommends the use of machine learning to demonstrate the outbreak and proposes regression models for future research. The research highlights the importance of precise validation and data analysis in developing strategies for early healing and prevention.
AIN SHAMS ENGINEERING JOURNAL
(2022)
Article
Computer Science, Information Systems
Kai Yang, Jie Lu, Wanggen Wan, Guangquan Zhang, Li Hou
Summary: This paper proposes a transfer learning method based on sparse Gaussian process to tackle regression problems, and maintains transfer performance through adaptive neural kernel network and transfer inducing point algorithm.
INFORMATION SCIENCES
(2022)
Article
Engineering, Multidisciplinary
Abdullahi Abubakar Mas'ud
Summary: This study investigates the application of different machine learning models to predict the PV power output at Jubail Industrial City in Saudi Arabia. The results show that the kNN model performs the best in terms of prediction accuracy. This research highlights the potential of using machine learning techniques to accurately predict PV output power across different regions.
AIN SHAMS ENGINEERING JOURNAL
(2022)
Article
Engineering, Multidisciplinary
Ahmad Hammad Khaliq, Muhammad Basharat, Malik Talha Riaz, Muhammad Tayyib Riaz, Saad Wani, Nadhir Al-Ansari, Long Ba Le, Nguyen Thi Thuy Linh
Summary: This study utilized machine learning techniques to analyze and assess the landslide susceptibility in the Hattian Bala district of NW Himalayas, Pakistan. Historical satellite imageries were used to generate spatiotemporal landslide inventories, and a spatial database was created for various factors. The results showed that the Random Forest model outperformed the Logistic Regression model. The study aims to minimize losses and effectively manage landslide hazards in the region.
AIN SHAMS ENGINEERING JOURNAL
(2023)
Article
Green & Sustainable Science & Technology
Huaiping Jin, Lixian Shi, Xiangguang Chen, Bin Qian, Biao Yang, Huaikang Jin
Summary: A novel probabilistic wind power forecasting method based on selective ensemble of finite mixture Gaussian process regression models (SEFMGPR) is proposed in this study, which enhances prediction accuracy by constructing diverse GPR models, integrating FMGPR models, and selecting highly influential models using genetic algorithm. Additionally, an incremental adaptation mechanism is employed to alleviate performance degradation. Application results demonstrate that SEFMGPR outperforms traditional global and ensemble wind power prediction methods in handling time-varying changes and maintaining high prediction accuracy.
Article
Automation & Control Systems
Ismail Adewale Olumegbon, Ibrahim Olanrewaju Alade, Mojeed Opeyemi Oyedeji, Talal F. Qahtan, Aliyu Bagudu
Summary: This study presents a machine learning (ML) approach for estimating the diffusion coefficient of molecular gas systems. The ML models, including support vector regression, Gaussian process regression, and artificial neural networks, were built using simple descriptors such as temperature, pressure, molar masses, and mole fractions. The artificial neural network model showed the best generalization capability for evaluating the binary diffusion coefficient of gases.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)