Article
Computer Science, Artificial Intelligence
Ganesan Kalaiarasi, Sureshbabu Maheswari
Summary: In this study, an effective classification of hyperspectral images was modeled and simulated with the proximal support vector machine (PSVM) by integrating them with the deep learning approach. The new deep PSVM classifiers, designed to handle the complexity, discrepancies, and irregularities in traditional hyperspectral image classifiers, showed better classification accuracy compared to other techniques.
NEURAL COMPUTING & APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Ceren Atik, Recep Alp Kut, Reyat Yilmaz, Derya Birant
Summary: This paper proposes a novel method called support vector machine chains (SVMC) that involves chaining together multiple SVM classifiers in a special structure, decrementing one feature at each stage. The paper also introduces a new voting mechanism called tournament voting, where classifiers' outputs compete in groups and the winning class label of the final round is assigned as the prediction. Experimental results show that SVMC outperforms SVM in terms of accuracy and achieves a 6.88% improvement over state-of-the-art methods.
Article
Economics
Ahmad Hammami, Mohammad Hendijani Zadeh
Summary: In this paper, six novel ensemble classifiers are used to predict earnings management (EM) in both forms, accrual-based earnings management (AEM) and real earnings management (REM). The paper compares the EM prediction accuracy of wrapper feature selection (FS) and filtering FS techniques. The results show that the ABC-SVM ensemble classifier outperforms others in predicting both AEM and REM. It is also found that wrapper FS ensemble classifiers generally outperform filtering FS ensemble classifiers in predicting AEM and REM, and that predicting REM is more difficult than predicting AEM. This paper contributes to the literature on EM prediction by introducing six new ensemble classifiers, and it is the first work (to the best of our knowledge) to consider both REM and AEM in one context and to compare the performance of wrapper and filtering FS techniques in the EM prediction setting.
JOURNAL OF FORECASTING
(2022)
Article
Chemistry, Analytical
Donghui Chen, Bingyang Wang, Xiao Yang, Xiaohui Weng, Zhiyong Chang
Summary: Accurate and rapid prediction of pesticides in groundwater is crucial for protecting human health. This study introduced the TrAdaBoost transfer learning method to recognize pesticides in groundwater using an electronic nose. Two steps were conducted: qualitative identification of pesticide type and semi-quantitative prediction of pesticide concentration. The integration of support vector machine with TrAdaBoost showed significantly improved recognition rates compared to methods without transfer learning. These results demonstrate the potential of TrAdaBoost-based support vector machine approaches in recognizing pesticides in groundwater with limited samples.
Article
Food Science & Technology
Dongbing Yu, Yu Gu
Summary: The study introduces a new classification framework using CNN-SVM for the sub-categories of Maofeng green tea and Maojian green tea, leveraging electronic nose data for deep feature extraction and utilizing an SVM classifier to enhance classification performance. The CNN-SVM framework shows great potential for fine-grained classification of highly similar teas with high accuracy and robustness.
Article
Chemistry, Analytical
Xiuguo Zou, Chenyang Wang, Manman Luo, Qiaomu Ren, Yingying Liu, Shikai Zhang, Yungang Bai, Jiawei Meng, Wentian Zhang, Steven W. Su
Summary: In this study, an electronic nose detection system based on KNN-SVM was designed for apple quality grading. By optimizing the nasal cavity structure and reducing data dimension using PCA and LDA, the classifier achieved improved accuracy.
Article
Chemistry, Analytical
Abeer Alshejari, Vassilis S. Kodogiannis, Stavros Leonidis
Summary: This paper presents a prediction system based on multispectral imaging technique for predicting the total viable counts of microorganisms in beef fillet samples. The paper also explores a feature fusion approach and compares the performance of neurofuzzy models with other regression algorithms. The results confirm the validity of using feature selection methods and neurofuzzy models for assessing the microbiological quality of meat products.
Article
Chemistry, Analytical
Shengming Li, Lin Feng, Yunfei Ge, Li Zhu, Liang Zhao
Summary: The novel electronic nose with active perception and ensemble learning method can differentiate the smell of different objects effectively. Experimental results show that the accuracy of active odor perception can reach over 90%, even with only 30% training data.
Article
Environmental Sciences
Rashmi Saini, Sanjay Kumar Ghosh
Summary: This study successfully identified major crops in the Roorkee, India using Sentinel-2A data, with Xgboost showing the best performance for crop mapping and support vector machine (SVM) performing the worst. Major crops, such as wheat and sugarcane, were accurately classified with high accuracy rates.
GEOCARTO INTERNATIONAL
(2021)
Article
Computer Science, Artificial Intelligence
Rosita Guido, Maria Carmela Groccia, Domenico Conforti
Summary: Hyperparameter tuning is essential for improving model performance in machine learning. This research focuses on classifying imbalanced data using cost-sensitive support vector machines and proposes a multi-objective approach to optimize the model's hyperparameters. The algorithm is presented in a basic version and an improved version utilizing genetic algorithms and decision trees. Experimental results demonstrate the importance of using appropriate evaluation measures for assessing the classification performance of imbalanced data classification models.
Article
Computer Science, Artificial Intelligence
Gherardo Varando, Salvador Catsis, Emiliano Diaz, Gustau Camps-Valls
Summary: Bivariate causal discovery is the task of inferring the causal relationship between two random variables from observational data. This paper proposes an ensemble algorithm that combines classical and data-driven methods, achieving superior performance on various synthetic and real-world problems.
APPLIED SOFT COMPUTING
(2024)
Article
Computer Science, Artificial Intelligence
Pawel Ksieniewicz, Pawel Zyblewski, Robert Burduk
Summary: Ensembles of classifiers are known for their stability and accuracy, often outperforming single classifiers. This study proposes a fusion method in geometric space using decision boundaries of base classifiers, introducing a new function for measuring central tendency and removing the limit on the number of base classifiers. Experiments on multiple binary datasets show the effectiveness of this approach.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Chemistry, Multidisciplinary
Mengli Cao, Xingwei Ling
Summary: This study focuses on a problem in perfume identification, which is how to improve accuracy using ensemble learning methods. The authors tested four ensemble learning methods and found that XGBoost performed the best in terms of accuracy mean and variance on their self-designed electronic nose.
APPLIED SCIENCES-BASEL
(2022)
Article
Optics
Renato Luiz Faraco Filho, Felipe Oliveira Barino, Joao Calderano, Italo Fernando Valle Alvarenga, Deivid Campos, Alexandre Bessa Dos Santos
Summary: This paper presents the application of an in-fiber Mach-Zehnder interferometer (MZI) for monitoring coffee bean fermentation. Two MZIs, based on a combination of a fiber taper cascaded by a micro-tapered long-period fiber grating, were installed in a fermentation barrel to monitor the liquids and gases released during the fermentation process. The research demonstrates that in-fiber MZIs are promising optical noses for this scenario.
Article
Computer Science, Information Systems
Zichen Zhang, Shifei Ding, Yuting Sun
Summary: This paper introduces a new method called multiple birth support vector regression (MBSVR), which constructs the regressor from multiple hyperplanes obtained by solving small quadratic programming problems, aiming for faster computation and better fitting precision.
INFORMATION SCIENCES
(2021)
Article
Agriculture, Dairy & Animal Science
Sofia Tsaloumi, Zafiro Aspridou, Evgenia Spyrelli, George-John E. Nychas, Konstantinos Koutsoumanis
Summary: This study investigated and modeled the growth of naturally contaminated pseudomonads on fresh poultry fillets as a function of temperature. The results showed that the muscle type does not significantly affect the temperature dependence of pseudomonads growth kinetics. A unified model was developed and validated, and it was used to predict the shelf life of fresh poultry products for risk assessment.
Article
Chemistry, Analytical
George Pampoukis, Anastasia E. Lytou, Anthoula A. Argyri, Efstathios Z. Panagou, George-John E. Nychas
Summary: Unsafe food is estimated to cause 600 million cases of foodborne disease annually. This review summarizes recent progress in rapid microbial assessment methods, including spectroscopic techniques, spectral imaging techniques, biosensors, and sensors designed to mimic human senses.
Article
Chemistry, Analytical
Dimitrios Kolosov, Lemonia-Christina Fengou, Jens Michael Carstensen, Nette Schultz, George-John Nychas, Iosif Mporas
Summary: This article proposes an architecture for estimating microbial populations in meat samples using multispectral imaging and deep convolutional neural networks. The deep learning models operate on embedded platforms, and the experimental results show the advantages of different platforms in terms of latency, throughput, efficiency, and value.
Editorial Material
Microbiology
Olga S. Papadopoulou, Agapi Doulgeraki, Efstathios Panagou, Anthoula A. Argyri
FRONTIERS IN MICROBIOLOGY
(2023)
Review
Pediatrics
Victoria Ramos-Garcia, Isabel Ten-Domenech, Alba Moreno-Gimenez, Laura Campos-Berga, Anna Parra-Llorca, Amparo Ramon-Beltran, Maria J. Vaya, Fady Mohareb, Corentin Molitor, Paulo Refinetti, Andrei Silva, Luis A. Rodrigues, Serge Rezzi, Andrew C. C. Hodgson, Stephane Canarelli, Eirini Bathrellou, Eirini Mamalaki, Melina Karipidou, Dimitrios Poulimeneas, Mary Yannakoulia, Christopher K. Akhgar, Andreas Schwaighofer, Bernhard Lendl, Jennifer Karrer, Davide Migliorelli, Silvia Generelli, Maria Gormaz, Miltiadis Vasileiadis, Julia Kuligowski, Maximo Vento
Summary: This study aims to compare the effect of mother's own milk (OMM) and pasteurized donor human milk (DHM) on % weight gain/month in preterm and term infants. It also evaluates the impact of diet, lifestyle habits, psychological stress, and pasteurization on milk composition and their influence on infant growth, health, and development. Biological samples and various characteristics are collected at different time points for analysis. The study also involves the use of sensor prototypes, measurement of maternal psychosocial status, examination of postpartum bonding and parental stress, and application of neurodevelopment scales.
FRONTIERS IN PEDIATRICS
(2023)
Article
Microbiology
Agapi I. Doulgeraki, Christina S. Kamarinou, George-John E. Nychas, Anthoula A. Argyri, Chrysoula C. Tassou, Georgios Moulas, Nikos Chorianopoulos
Summary: Microbial interactions have a significant impact on biofilm formation and the efficacy of disinfectants. This study assessed the effect of microbial interactions on biofilm formation and the disinfection activity of a photocatalytic surfactant based on TiO2 nanoparticles. The presence of different species or dual-species biofilms influenced the microbial load and population, while the disinfectant enhanced the antimicrobial activity of UV light. The presence of multiple species also affected the resistance of biofilm cells to UV light and disinfectants. These findings highlight the importance of microbial interactions in biofilm formation and decontamination.
Article
Plant Sciences
Frank Gyan Okyere, Daniel Cudjoe, Pouria Sadeghi-Tehran, Nicolas Virlet, Andrew B. B. Riche, March Castle, Latifa Greche, Fady Mohareb, Daniel Simms, Manal Mhada, Malcolm John Hawkesford
Summary: This study aimed to develop a fast and robust neural-network-based segmentation tool for high-throughput phenotyping of plants in both field and glasshouse environments. The proposed method outperformed other methods in producing quality segmented images with over 98%-pixel classification accuracy, making it an essential tool for the development of a data analysis pipeline for high-throughput plant phenotyping.
Article
Plant Sciences
Mateus Henrique Vicente, Kyle MacLeod, Feng Zhu, Diego D. Rafael, Antonio Figueira, Alisdair R. Fernie, Fady Mohareb, Zoltan Kevei, Andrew J. Thompson, Agustin Zsogon, Lazaro Eustaquio Pereira Peres
Summary: The study identified a genetic locus on chromosome 7 in tomato that controls both vegetative and reproductive organ size, with alleles from wild species leading to lower cell number and reduced size of leaves, flowers, and fruits in introgression lines. The findings suggest that selection for large fruit during domestication also influences leaf size by altering cell division, potentially allowing for fine-tuning of parameters important for crop adaptation.
Review
Food Science & Technology
Maria Kazou, Eleni Nikolopoulou, Efstathios Z. Panagou
Summary: The consumption of fermented foods, especially traditional or ethnic ones like table olives, has become important for improving human health. Greece has a long tradition in producing table olives, but there has been a recent effort to modernize the industry and adopt scientifically based processing methods. This review provides an up-to-date overview of the transformation of phenolic compounds in Greek table olives during processing and discusses their functional and antioxidant potential as well as their nutritional implications.
EUROPEAN FOOD RESEARCH AND TECHNOLOGY
(2023)
Article
Food Science & Technology
Zoe Gounari, Stamatoula Bonatsou, Ilario Ferrocino, Luca Cocolin, Olga S. Papadopoulou, Efstathios Z. Panagou
Summary: This study investigated the physicochemical characteristics and microbiological composition of naturally black dry-salted olives obtained from different retail outlets in the Greek market. The results showed significant variability in the physicochemical attributes among the samples, with pH values ranging from 4.0 to 5.0 and water activity values ranging from 0.58 to 0.91. The dominant yeast species identified were Pichia membranifaciens, Candida sorbosivorans, Citeromyces nyon-sensis, Candida etchelsii, Wickerhamomyces subpelliculosus, Candida apicola, Wickerhamomyces anomalus, Torulaspora delbrueckii, and Candida versatilis. The majority of the samples met the microbiological and hygienic quality requirements, but there was a lack of standardization in the processing of this commercial style.
INTERNATIONAL JOURNAL OF FOOD MICROBIOLOGY
(2023)
Review
Plant Sciences
Daniel K. Cudjoe, Nicolas Virlet, March Castle, Andrew B. Riche, Manal Mhada, Toby W. Waine, Fady Mohareb, Malcolm J. Hawkesford
Summary: Improving crop productivity is essential to meet the dietary demands of the fast-growing African population. Developing high-yielding crop cultivars resilient to biotic and abiotic stresses is crucial. High-throughput plant phenotyping approaches enable the African plant science community to measure complex quantitative phenotypes and establish the genetic basis of agriculturally relevant traits. However, Africa faces practical, financial, geographical, and political barriers in the development and implementation of these systems.
FRONTIERS IN PLANT SCIENCE
(2023)
Article
Plant Sciences
Frank Gyan Okyere, Daniel Cudjoe, Pouria Sadeghi-Tehran, Nicolas Virlet, Andrew B. Riche, March Castle, Latifa Greche, Daniel Simms, Manal Mhada, Fady Mohareb, Malcolm John Hawkesford
Summary: Sustainable fertilizer management in precision agriculture is crucial. Hyperspectral imaging is a remote sensing tool used to monitor plant nutrient status. This study proposes a hybrid convolution neural network (CNN) that can extract both spatial and spectral information to identify the nutrient status of quinoa and cowpea plants at different growth stages.
FRONTIERS IN PLANT SCIENCE
(2023)
Article
Biochemistry & Molecular Biology
Nikolaos Zacharodimos, Christina Athanasaki, Stamatia Vitsou-Anastasiou, Olga S. S. Papadopoulou, Natalia Moniaki, Agapi I. I. Doulgeraki, George-John E. Nychas, Chrysoula C. C. Tassou, Emilia Papakonstantinou
Summary: The glycemic index (GI) of a commercial mixed fruit juice fortified with vitamin D3, n-3 polyunsaturated fatty acids (PUFA), probiotics, and their combination was determined in this study. The enriched fruit juices produced different postprandial glycemic and insulinemic responses and affected satiety scores compared to the control. All types of fruit juice attenuated postprandial glycemic responses, regardless of the added biofunctional ingredients, offering potential benefits for glycemic control.
Article
Biotechnology & Applied Microbiology
Patra Sourri, Anthoula A. Argyri, George-John E. Nychas, Chrysoula C. Tassou, Efstathios Z. Panagou
Summary: The population dynamics of A. acidoterrestris spores and vegetative cells in orange juice, treated with temperature-assisted HHP and stored in different isothermal conditions, were investigated in this study. The results showed that HHP treatment effectively reduced the spore population, and there were minimal changes in the quantity of spores and vegetative cells during storage, especially at low temperatures.
FERMENTATION-BASEL
(2022)
Article
Chemistry, Analytical
Gang Qu, Yuxin Zhao, Qiaoli Zhang, Jina Wu, Xiaosen Li, Yang Yang, Shilei Liu
Summary: In this study, magnetic mesoporous materials combined with real-time in situ mass spectrometry were used for the high-throughput detection of hydrolyzed products of organophosphorus nerve agents. The method showed good linearity, low limits of detection and quantification, and high extraction recoveries. The magnetic preparation method used was quick, cost-effective, rugged, and safe. The results demonstrated the potential of this method for rapid and efficient determination of the target analytes in environmental samples.
ANALYTICAL METHODS
(2024)
Article
Chemistry, Analytical
Anna Hildebrand, Mariam Merchant, Danny O'Hare
Summary: Substandard and falsified artemisinin derivatives in antimalarials have caused significant deaths and economic losses. This study evaluates the feasibility of voltammetric methods for identifying and quantifying artemether. The findings suggest that electrochemical analysis shows promise as a method for artemether identification and quantification.
ANALYTICAL METHODS
(2024)
Article
Chemistry, Analytical
Marx Osorio Araujo Pereira, Alvaro Ferreira Junior, Edson Silvio Batista Rodrigues, Helena Mulser, Giovanna Nascimento de Mello e Silva, Wallans Torres Pio dos Santos, Eric de Souza Gil
Summary: Brazilian spotted fever (BSF) is a serious and rapidly evolving disease. A new impedimetric immunosensor was developed for rapid diagnosis by measuring specific antibodies in plasma. The sensor demonstrated selectivity and accuracy, and has potential for important applications in diagnostic testing.
ANALYTICAL METHODS
(2024)
Article
Chemistry, Analytical
Kathrin Schilling, Ronald A. Glabonjat, Olgica Balac, Marta Galvez-Fernandez, Arce Domingo-Relloso, Vesna Slavkovich, Jeff Goldsmith, Miranda R. Jones, Tiffany R. Sanchez, Ana Navas-Acien
Summary: Analysis of trace elements in urine is an important tool for assessing exposures, diagnosing nutritional status, and guiding public health and healthcare intervention. This study provides a sensitive method for analyzing 18 elements in urine samples, using only 100 μL. The results show good accuracy and sensitivity of the method.
ANALYTICAL METHODS
(2024)
Article
Chemistry, Analytical
Mengya Li, Shijie Liu, Shiliang Guo, Dong Liang, Miaoyun Li, Yaodi Zhu, Lijun Zhao, Jong-Hoon Lee, Gaiming Zhao, Yangyang Ma, Yanxia Liu
Summary: In this study, a magnetic flow device was developed to purify spores in a culture medium system. The device used magnetic nanoparticles to absorb vegetative cells, separating them from the spores. The achieved purity of the collected spores was over 95%. The study also demonstrated a rapid quantitative detection method using Raman spectroscopy.
ANALYTICAL METHODS
(2024)
Article
Chemistry, Analytical
Wanqiong Liu, Zixuan Wu, Jianwei Peng, Zebin Xu, Yong Liang
Summary: Metal-organic frameworks (MOFs) are effective carriers for molecular imprinting, but their poor dispersibility in aqueous solution is a significant drawback. In this study, we have applied amphiphilic block copolymers and molecularly imprinted technology on MOFs to improve the hydrophilicity of molecularly imprinted fluorescent materials.
ANALYTICAL METHODS
(2024)