Multilayer Perceptron Artificial Neural Network Research Articles

Vis/NIR Spectroscopy and Vis/NIR Hyperspectral Imaging for Non-Destructive Monitoring of Apricot Fruit Internal Quality with Machine Learning.

The fruit supply chain requires simple, non-destructive, and fast tools for quality evaluation both in the field and during the post-harvest phase. In this study, a portable visible and near-infrared (Vis/NIR) spectrophotometer and a portable Vis/NIR hyperspectral imaging (HSI) device were tested to highlight genetic differences among apricot cultivars, and to develop multi-cultivar and multi-year models for the most important marketable attributes (total soluble solids, TSS; titratable acidity, TA; dry matter, DM). To do this, the fruits of seventeen cultivars from a single experimental orchard harvested at the commercial maturity stage were considered. Spectral data emphasized genetic similarities and differences among the cultivars, capturing changes in the pigment content and macro components of the apricot samples. In recent years, machine learning techniques, such as artificial neural networks (ANNs), have been successfully applied to more efficiently extract valuable information from spectral data and to accurately predict quality traits. In this study, prediction models were developed based on a multilayer perceptron artificial neural network (ANN-MLP) combined with the Levenberg-Marquardt learning algorithm. Regarding the Vis/NIR spectrophotometer dataset, good predictive performances were achieved for TSS (R2 = 0.855) and DM (R2 = 0.857), while the performance for TA was unsatisfactory (R2 = 0.681). In contrast, the optimal predictive ability was found for models of the HSI dataset (TSS: R2 = 0.904; DM: R2 = 0.918, TA: R2 = 0.811), as confirmed by external validation. Moreover, the ANN allowed us to identify the most predictive input spectral regions for each model. The results showed the potential of Vis/NIR spectroscopy as an alternative to traditional destructive methods to monitor the qualitative traits of apricot fruits, reducing the time and costs of analyses.

Machine learning-based estimation of crude oil-nitrogen interfacial tension

Accurate estimation of interfacial tension (IFT) between nitrogen and crude oil during nitrogen-based gas injection into oil reservoirs is imperative. The previous research works dealing with prediction of IFT of oil and nitrogen systems consider synthetic oil samples such n-alkanes. In this work, we aim to utilize eight machine learning methods of Decision Tree (DT), AdaBoost (AB), Random Forest (RF), K-nearest Neighbors (KNN), Ensemble Learning (EL), Support Vector Machine (SVM), Convolutional Neural Network (CNN) and Multilayer Perceptron Artificial Neural Network (MLP-ANN) to construct data-driven intelligent models to predict crude oil – nitrogen IFT based upon experimental data of real crude oils samples encountered in underground oil reservoirs. Several statistical indices and graphical approaches are used as accuracy performance indicators. The results show that virtually all the gathered datapoints are suitable for the purpose of model development. The sensitivity analysis indicated that pressure, temperature and crude oil API all negatively affect the IFT, with pressure being the most effective factor. The evaluation study proved that Random Forest is the most accurate developed intelligent model as it was characterized with acceptable R-squared (0.959), mean square error (1.65), average absolute relative error (6.85%) of unseen test datapoints as well as with correct trend prediction of IFT with regard to all input parameters of pressure, temperature and crude oil API. The developed model can be considered an accurate an easy-to-use tool for the prediction of crude oil/N2 IFT values for enhance oil recovery study optimization and upstream reservoir investigations.

Comparison of Different Machine Learning Methodologies for Predicting the Non-Specific Treatment Response in Placebo Controlled Major Depressive Disorder Clinical Trials.

Placebo effect represents a serious confounder for the assessment of treatment effect to the extent that it has become increasingly difficult to develop antidepressant medications appropriate for outperforming placebo. Treatment effect in randomized, placebo-controlled trials, is usually estimated by the mean baseline adjusted difference of treatment response in active and placebo arms and is function of treatment-specific and non-specific effects. The non-specific treatment effect varies subject by subject conditional to the individual propensity to respond to placebo. This effect is not estimable at an individual level using the conventional parallel-group study design, since each subject enrolled in the trial is assigned to receive either active treatment or placebo, but not both. The objective of this study was to conduct a comparative analysis of the machine learning methodologies to estimate the individual probability of a non-specific treatment effect. The estimated probability is expected to support novel methodological approaches for better controlling effect of excessively high placebo response. At this purpose, six machine learning methodologies (gradient boosting machine, lasso regression, logistic regression, support vector machines, k-nearest neighbors, and random forests) were compared to the multilayer perceptrons artificial neural network (ANN) methodology for predicting the probability of individual non-specific treatment response. ANN achieved the highest overall accuracy among all methods tested. A fivefold cross-validation was used to assess performances and risks of overfitting of the ANN model. The analysis conducted without subjects with non-specific effect indicated a significant increase of signal detection with significant increase in effect size.

Analysis of reservoir rock permeability changes due to solid precipitation during waterflooding using artificial neural network

Waterflooding can lead to formation damage, which is affected by various parameters. This study experimentally examined changes in the permeability of rock samples due to calcium sulfate precipitation over a wide range of temperatures, sulfate ion concentrations, injection rates, and injected pore volumes. The results showed that with increasing temperature, ion concentration, injected pore volume, and decreasing injection rate, salt precipitation and formation damage are increased. These effects were associated with a reduction in solubility, collisions of the particles, and large numbers of ions in the porous media. Experimental rock permeability data were then used to develop a model using a multi-layer perceptron (MLP) artificial neural network (ANN). In order to create a high-performing MLP-ANN model, various transfer functions and training algorithms were evaluated. Tansig was the most effective transfer function for the hidden layers. The ANN model using the Levenberg-Marquardt (LM) algorithm showed superior performance with the lowest mean squared error (MSE) for training, testing, and all data sets. Through the implementation of this algorithm in the network, it was determined that the optimal configuration consisted of three layers - comprising of two hidden layers with eight and four neurons in the first and second layer, respectively, along with one output layer. A comparison of the predicted values of rock permeability under scaling conditions with experimental data showed that the proposed ANN model makes it possible to predict formation damage due to scaling with high accuracy. At the same time, no deviations were identified between the predicted and laboratory data. Finally, using a scale inhibitor prevented formation damage at any temperature, injection rate, and ion concentration, confirming the high effectiveness of the scale inhibitor injection for the field application.

Wavelet as a feature vector reduction method used in the recognition of urban buildings

Building recognition is a very difficult task to achieve, since their images have different angles and lighting conditions and partial obstructions from trees, moving vehicles or other buildings. Thus, due to all these difficulties, this work proposes a recognition system that uses the Gabor filter representation for feature extraction and the Wavelet transform for dimensionality reduction, and the Multilayer Perceptron artificial neural network for recognition. To verify the efficiency of this system, a database of images of new and old buildings was used. The performance of the method stands out significantly in relation to other methods found in other published works.

Multi-layer perceptron artificial neural network for environmental risks prediction of SW-induced pollution in Dar es Salaam, Tanzania

<p>Many metropolitan areas face significant environmental challenges posed by improper disposal and management of solid waste. As a result, environmental risks have emerged as a pressing concern, prompting dedicated research efforts. This study on environmental risk prediction of Dar es Salaam SW coincides with a mounting governmental effort over rising pollution levels from inadequate SW management. Using the multi-layer perceptron artificial neural network (MLP-ANN) model, it effectively examines the prevailing conditions and forecasts waste generation rates (WGRs) and environmental risk index (ERI) associated with SW pollution. As confirmed with 94.5% prediction accuracy and 86.5% success rate of the MLP-ANN model, WGRs in Dar es Salaam have doubled in less than two decades. Besides, over 40% of the overall generated SW is left unattended. Consequently, the ERI exhibits a consistent upward trajectory throughout the assessment period, with intermittent fluctuations between Level II and III but a persistent overall increase. Projections indicate an escalation of ERI to Level IV by 2025/26 and to a critical threshold (Level V) by 2038. The key indices such as pressure, state, and impact are anticipated to reach critical thresholds ahead of the comprehensive ERI. This underscores the imperative for timely interventions and the urgency of addressing SW management issues to curb the escalating environmental risks in Dar es Salaam and other metropolises with similar challenges.<strong></strong></p>

Detection of partial shading and short-circuit faults in a pv system using the artificial neural network multilayer perceptron technique

Photovoltaic (PV) systems play a crucial role in renewable energy production but face challenges due to environmental factors such as partial shading and short circuits, which reduce efficiency and reliability. Continuous monitoring and early fault detection are essential to mitigate these issues. In this work, we propose a fault detection method that focuses on identifying partial shading and short-circuit faults using an artificial neural network (ANN), specifically a Multi-Layer Perceptron (MLP). The MLP model is trained to classify these faults under various environmental scenarios, including situations where both faults occur simultaneously. The methodology involves simulating different fault scenarios using MATLAB/Simulink and training the MLP on input parameters such as voltage, current, and power output from the PV system. Results indicate that the proposed approach achieves high fault detection accuracy, minimizing classification errors and ensuring reliable identification of system failures. This research highlights the advantages of integrating ANN-based fault detection methods into PV system monitoring frameworks, enabling real-time diagnostics and proactive maintenance. By improving fault detection accuracy, the proposed method contributes to the operational efficiency, safety, and longevity of PV installations. This approach can be extended to large-scale PV systems, addressing the growing demand for reliable renewable energy solutions globally. The findings demonstrate the potential of ANN-based methods to enhance fault diagnostics, providing a scalable solution for modern PV systems and supporting the global transition to sustainable energy.

Artificial intelligence for herbicide recommendation: case study for the use of clomazone in Brazilian soils

The use of herbicides has been carried out without considering the physical and chemical properties of the soil and its variation in the different producing regions. Its interaction is dependent on the properties of the soil and the herbicide itself and, therefore, is considered complex. An alternative to optimize the use of herbicides is the application of mathematical models to estimate the sorption in the soil, contributing to a more efficient control of weeds and bringing more environmental safety. In this study, the efficiency of Artificial Neural Networks (ANNs) was evaluated as a tool to predict clomazone sorption in different Brazilian soils. The herbicide sorption coefficients were determined in the laboratory with 45 soils from different soil environments. Multilayer perceptron (MLP) ANN models were used to predict clomazone sorption. The variables were selected using the feature selection tool using the physical and chemical properties of the soils. Multivariate statistics were performed to determine the correlations between the physical and chemical properties of the soil with the sorption coefficient. Potassium (K), phosphorus (P), magnesium (Mg), organic matter (OM), silt, clay, and cation exchange capacity (CEC) were the ANN inputs and the sorption coefficient (Kfs) the output. The general performance of the model was evaluated by its precision and error values, namely: coefficient of determination (R2), mean absolute relative error (RMSE), mean absolute error (MAE), mean estimation error (MBE), and the coefficient Pearson correlation test (r). The ANN models were able to predict the Kfs of clomazone in the studied soils. The input variables that determined the best performing network for Kfs prediction were: CTC, Silt, Mg, and K. The variables that showed greater relative importance in the sensitivity analysis in the construction of the model for the Kfs of clomazone were: K (37%) and Silt (29%). The best RNA model was able to recommend reducing the dose of clomazone compared to the commercial package insert method.

Meteorological and satellite-based data for drought prediction using data-driven model

This work presents a data-driven model, the Artificial Neural Network-Multilayer Perceptron Neural Network (ANN-MLP), for use in meteorological drought deciles index (DDI) predictions over various climatic sub-zone. Two types of rainfall data from meteorological weather stations (WSs) and satellite-based estimates of PERSIANN (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Network) were adopted. This work considered the calculated DDI (DDI original) from WSs to train and develop the proposed algorithm at three sub-zones (ANN-MLP-DDI models). The newly developed model was tested for DDI prediction using PERSIANN, and compared with the calculated DDI original from WSs. The results positively revealed that the ANN-MLP-DDI models showed high performance (Correlation coefficient r= 0.981) for DDI prediction against the DDI original from WSs. It can be concluded that data-driven models are feasible for drought prediction, and this work could help water managers in mitigating drought impacts and in providing information for policy makers

Groundwater Nitrate Modeling in Tehran Metropolis Using Artificial Neural Network and Kriging Methods

This study examined the relationship between groundwater quality and land use in Tehran. For this purpose, the possible relationship between the types of land uses and the concentration of nitrate in groundwater parameters was modelled using a Multi-Layer Perceptron (MLP) artificial neural network in geographic information system (GIS). The optimal network model was selected based on the mean root mean square error (RMSE) and correlation coefficient. Interpolation through Kriging was also performed to compare its results with those of the predicted model derived from an artificial neural network. The results showed that the neural network has a high capability for predicting and modelling groundwater nitrate concentration compared to the Kriging method. The high accuracy (RMSE: 0.003) of the neural network makes it a useful tool in relevant management issues. Our results of network sensitivity analysis were similar to scientific findings regarding the factors influencing the formation of nitrate in groundwater. Model outputs in the form of maps, tables, and graphs allowed the study of the role of each variable and the extent of its impact on groundwater quality. Performing various simulations and modelling of groundwater pollution provides an effective benchmark towards optimizing the management, control, planning, and decision-making in urban areas and can lead to economic and environmental savings.

Groundwater Nitrate Modeling in Tehran Metropolis Using Artificial Neural Network and Kriging Methods

This study examined the relationship between groundwater quality and land use in Tehran. For this purpose, the possible relationship between the types of land uses and the concentration of nitrate in groundwater parameters was modelled using a Multi-Layer Perceptron (MLP) artificial neural network in geographic information system (GIS). The optimal network model was selected based on the mean root mean square error (RMSE) and correlation coefficient. Interpolation through Kriging was also performed to compare its results with those of the predicted model derived from an artificial neural network. The results showed that the neural network has a high capability for predicting and modelling groundwater nitrate concentration compared to the Kriging method. The high accuracy (RMSE: 0.003) of the neural network makes it a useful tool in relevant management issues. Our results of network sensitivity analysis were similar to scientific findings regarding the factors influencing the formation of nitrate in groundwater. Model outputs in the form of maps, tables, and graphs allowed the study of the role of each variable and the extent of its impact on groundwater quality. Performing various simulations and modelling of groundwater pollution provides an effective benchmark towards optimizing the management, control, planning, and decision-making in urban areas and can lead to economic and environmental savings.

Accurate modeling of crude oil and brine interfacial tension via robust machine learning approaches

Interfacial tension (IFT) between water and crude oil is a crucial variable that enhanced oil recovery (EOR) techniques can adjust to increase oil extraction from depleted fields. Most of the developed intelligent models in the literature are based on synthetic oil samples rather than real crude oil samples or brine total salinity rather than salinity of each salt type. Hence, this study applies various machine learning approaches, such as Convolutional Neural Networks (CNN), Adaptive Boosting (AdaBoost), Decision Trees (DT), Random Forest (RF), K-Nearest Neighbors (KNN), Ensemble Learning, Support Vector Machines (SVM), and Multi-Layer Perceptron Artificial Neural Networks (MLP-ANN) to develop advanced models for predicting the IFT between brine and crude oil considering real crude oil samples and taking the account of each salt type prevalent within the brine phase, which represent the realistic circumstances encountered in the oil reservoirs. These predictions are based on factors like the type and concentration of salt, the API of the crude oil, and the properties of the system (pressure and temperature) using previously published experimental data. A sensitivity analysis, incorporating a relevancy factor, is performed to highlight the influence of various input parameters on the IFT. Among these models, the Decision Tree is highlighted for its high accuracy and low training cost compared to ANN-based models, as evidenced by its emerged evaluation metrics (R-squared of 0.9796 and mean square error of 5e-4). It is noted that the AdaBoost model is the least accurate with an R2 of 0.6696. Furthermore, the sensitivity analysis indicates that the molecular weight of the salt has the smallest impact on the IFT, whereas temperature has the most significant effect. The developed smart model may be used to accurately estimate crude oil/brine IFT without needing tedious, time-consuming and expensive experimental workflows.

Experimental and machine learning insights on heat transfer and friction factor analysis of novel hybrid nanofluids subjected to constant heat flux at various mixture ratios

This study explores the combined effects of aluminum oxide (Al₂O₃)/graphene oxide (GO) hybrid nanofluids in 50:50 and 80:20 ratios, offering a notable improvement over conventional Al₂O₃ or GO nanofluids. It delivers a thorough comparison of thermophysical properties such as thermal conductivity and viscosity and heat transfer performance across water, Al₂O₃ nanofluids, and the Al₂O₃/GO hybrids. Nanofluids at 0.1–0.5 % volume concentrations were tested in a horizontal circular pipe under constant heat flux and turbulent flow with an inlet temperature of 60 °C. The maximum Nu enhancements of 64, 56 and 41 % were noted for Al₂O₃/GO (50:50), Al₂O₃/GO (80:20), and Al₂O₃ nanofluids, respectively at 0.5 vol%, compared to water. The maximum pressure drop of Al2O3/GO (50:50) nanofluid is 5.64 and 8.3 % greater than that of Al2O3/GO (80:20) and Al2O3 nanofluid, respectively at 0.5 vol%. The peak thermal performance index of 1.56, 1.48, and 1.33 is observed for Al₂O₃/GO (50:50), Al₂O₃/GO (80:20), and Al₂O₃ nanofluids. The integration of a multi-layer perceptron artificial neural network further enhances accuracy in predicting thermal performance, surpassing the precision of conventional empirical models. The adopted model showed excellent predictive accuracy, with correlation coefficients of 0.98493 in training, 0.9837 in validation, and 0.98698 in testing.

Advancing Geotechnical Evaluation of Wellbores: A Robust and Precise Model for Predicting Uniaxial Compressive Strength (UCS) of Rocks in Oil and Gas Wells

This study examines the efficacy of various machine learning models for predicting the uniaxial compressive strength (UCS) of rocks in oil and gas wells, which are essential for ensuring wellbore stability and optimizing drilling operations. The investigation encompasses Linear Regression, ensemble methods (including Random Forest, Gradient Boosting, XGBoost, and LightGBM), support vector machine-based regression (SVM-SVR), and multilayer perceptron artificial neural network (MLP-ANN) models. The results demonstrate that XGBoost and Gradient Boosting offer superior predictive accuracy for UCS in drillability, as indicated by low Mean Absolute Percentage Error (MAPE) values of 3.87% and 4.18%, respectively, and high R2 scores (0.8542 for XGBoost). These models emerge as optimal choices for UCS prediction focused on drillability, offering increased accuracy and reliability in practical engineering scenarios. Ensemble methods and MLP-ANN emerge as frontrunners, providing valuable tools for improving wellbore stability assessments, optimizing drilling parameter selection, and facilitating informed decision-making processes in oil and gas drilling operations. Moreover, this study lays a foundation for further research in drillability-centred predictive modelling for geotechnical parameters, advancing our understanding of rock behaviour under drilling conditions.

Stock Market Forecasting Using a Neural Network Through Fundamental Indicators, Technical Indicators and Market Sentiment Analysis

AbstractThe objective of this research is to provide evidence that it is possible to obtain a prediction that better aligns with the future performance of a stock if a neural network model is trained with stock market analysis variables and qualitative variables. As a case study, thirty-three companies’ representative of the S&P 500 are selected, and a multilayer perceptron artificial neural network is built and trained with input parameter indicators of fundamental analysis, technical analysis, and market sentiment. By incorporating the latter as an additional variable, the model's accuracy increases by 1.5% for 66% of the companies analyzed. The results confirm the crucial role played by the selection of the neural network model and its variables depending on the type of company to be analyzed. The main contributions of this research are the identification of the best variables combination to train a neural network model depending on the market sector to be analyzed, likewise it is demonstrated that, by using market sentiment, it is possible obtain a high accuracy or increase the accuracy to an existing model.

Optimisation of Flexible Forming Processes Using Multilayer Perceptron Artificial Neural Networks and Genetic Algorithms: A Generalised Approach for Advanced High-Strength Steels.

Flexibility is crucial in forming processes as it allows the production of different product shapes without changing equipment or tooling. Single-point incremental forming (SPIF) provides this flexibility, but often results in excessive sheet metal thinning. To solve this problem, a pre-forming phase can be introduced to ensure a more uniform thickness distribution. This study represents advances in this field by developing a generalised approach that uses a multilayer perceptron artificial neural network (MLP ANN) to predict thinning results from the input parameters and employs a genetic algorithm (GA) to optimise these parameters. This study specifically addresses advanced high-strength steels (AHSSs) and provides insights into their formability and the optimisation of the forming process. The results demonstrate the effectiveness of the proposed method in minimising sheet metal thinning and represent a significant advance in flexible forming technologies applicable to a wide range of materials and industrial applications.

Thermal and rheological behavior of aluminium-doped zinc oxide nanofluids: Experimental study and application of artificial neural network model

In this study, the influence of Al-doping concentration on the thermal and rheological characteristics of pristine zinc oxide (ZnO) nanofluids was investigated across a range of temperatures and nanoparticle concentrations. A multilayer perceptron (MLP) artificial neural network (ANN) with an optimized topology was employed to model the thermal conductivity and viscosity of nanofluids. The addition of 0.13 M of Al-doped zinc oxide nanoparticles at 1 % rate led to a remarkable 64 % increase in the thermal conductivity of the base fluid (50:50 ratio water + ethylene glycol mixture). Furthermore, incorporating 1 % of these Al-doped nanoparticles outperformed pure zinc oxide, resulting in a 44 % boost in thermal conductivity. The enhanced heat transfer capabilities of Al-doped zinc oxide were evident across different doping and nanoparticle concentrations. The viscosity of nanofluids increases with an increase in nanoparticle concentration and decreases with an increase in temperature. ZnO and Al-doped ZnO nanoparticles have been prepared and studied their structural, morphological and elemental properties using XRD, SEM and EDS respectively. The predicted thermal and rheological behaviours are in close agreement with the experimental values.

MODEL OF MAXIMAL WEIGHTS INVERSE CHAINS FOR THE ANALYSIS OF THE INFLUENCE FACTORS OF THE SOFTWARE COMPLEXES SUPPORT

Context. The problem of identification, formation and restoration of the boundaries of influencing factors, lost as a result of the implementation of multi-layer perceptron models into the models of subjective perception of the object of software complexes support, as well as the applied practical problem of primary monitoring of the frequency manifestation of a given influencing factor in the post-real-time mode, is considered. The object of research is the influencing factors of support of software complexes. Objective – the goal of the work is to develop a model of inverse chains of maximum weights for the analysis of influencing factors of the software complexes support. Method. A model of maximal weights inverse chains for the analysis of the influence factors of the software complexes support was developed for the analysis of the influencing factors of the software complexes support. The developed model provides possibility to identify and form feedback chains of maximum weights for the identification and further analysis of influencing factors that are reflected into the results of the object perception (the supported software complex or its support processes), by the relevant subjects of interaction which directly or indirectly interact with it. Results. Results of the resolved applied practical problem of primary monitoring of the frequency manifestation of a given influencing factor in the post-real-time mode have been provided as an example of the applied practical use of the developed model. The output results of the developed models functioning – are the reverse chains of maximum weights. In the future, the results obtained by the developed model are used to solve the applied-scientific problem of identification, formation and restoration of the boundaries of influencing factors, lost as a result of the implementation of the appropriate models of multilayer perceptron inside the models of subjective perception of the software complexes support. So the developed model of maximal weights inverse chains for the analysis of the influence factors of the software complexes support resolves this applied-scientific problem, initially caused by the implementation of the corresponding multilayer perceptron models inside the model of the subjective perception of the object of software complexessupport. The developed model provides the possibility of carrying out a qualitative analysis of the transformation of the input characteristics of the object of support into the output resulting characteristics of its subjective perception. Conclusions. Developed model allows to resolve the described problems. At the same time, the developed model improves the classical understanding of multilayer perceptron artificial neural networks, as it introduces an additional value to the neurons of hidden layers, which (starting from now) are able to perform a completely new role of influencing factors markers, while in the classical understanding of multilayer perceptron artificial neural networks they did not perform any functions other than arithmetic to ensure the possibility of correct learning and functioning of a multilayer perceptron artificial neural networks.

Predicting biohydrogen production from dark fermentation of organic waste biomass using multilayer perceptron artificial neural network (MLP–ANN)

The focus on sustainable energy has increased interest in biohydrogen production through dark fermentation of organic waste biomass, offering dual benefits of energy production and waste management. Optimizing this process is challenging due to complex interactions among substrate composition, microbial consortia, and fermentation parameters. A multilayer perceptron artificial neural network model was developed to predict biohydrogen yield from organic waste. The model, trained on 180 data points from 35 studies, uses inputs, such as substrate type, inoculum type, concentration, pH, and temperature, with hydrogen yield as the output. The multilayer perceptron artificial neural network model achieved high accuracy, with a root mean square error of 0.3838, a mean absolute percentage error of 0.1938, and a coefficient of determination of 0.8381. These results demonstrate the model's effectiveness in predicting biohydrogen production, providing a valuable tool for optimizing the fermentation process and advancing sustainable energy solutions.

Multivariate Modelling Based on Isotopic, Elemental, and Fatty Acid Profiles to Distinguish the Backyard and Barn Eggs.

The ability to trace the origin of eggs from backyard-raised hens is important due to their higher market value compared to barn-raised eggs. This study aimed to differentiate eggs from these two rearing systems using isotopic, elemental, and fatty acid profiles of egg yolks. A total of 90 egg yolk samples were analyzed, analytical results being followed by statistical tests (Student's t-test) showing significant differences in δ18O, several elements (Mg, K, Sc, Mn, Fe, Ni, Cu, Zn, As, Cd, Ba, Pb), and fatty acids compositions (C23:0, C17:0, C18:0, C16:1n7, C18:1n9, C18:2n6, C20:1n7, C20:4n6, C20:5n3, C22:6n3), as well as in the ratios of SFA, PUFA, and UFA. The results indicated a nutritional advantage in backyard eggs due to their lower n-6 polyunsaturated fatty acid content and a more favorable n-6 to n-3 ratio, linked to differences in the hens' diet and rearing systems. To classify the production system (backyard vs. barn), three pattern recognition methods were applied: linear discriminant analysis (LDA), k-nearest neighbor (k-NN), and multilayer perceptron artificial neural networks (MLP-ANN). LDA provided perfect initial separation, achieving 98.9% accuracy in cross-validation. k-NN yielded classification rates of 98.4% for the training set and 85.7% for the test set, while MLP-ANN achieved 100% accuracy in training and 92.3% in testing, with minor misclassification. These results demonstrate the effectiveness of fusion among isotopic, elemental, and fatty acid profiles in distinguishing backyard eggs from barn eggs and highlight the nutritional benefits of the backyard-rearing system.