Artificial Neural Network Analysis Research Articles

Detection and diagnosis of bacterial pathogens in blood using laser-induced breakdown spectroscopy

The ability to rapidly and accurately detect and identify pathogenic bacteria in clinically-obtained blood specimens with laser-induced breakdown spectroscopy (LIBS) is evaluated. Samples of blood obtained from five patients in a local hospital were confirmed to be negative for the presence of bacteria by the pathology department and were then tested with LIBS. Specimens of blood were tested as obtained from the hospital with no other alteration as control samples and were also intentionally spiked with known aliquots of Escherichia coli, Staphylococcus aureus, Enterobacter cloacae, and Pseudomonas aeruginosa to simulate blood infections. LIBS spectra were acquired from blood deposited on nitrocellulose filters. The intensities of 15 emission lines measured in the spectra and 92 ratios of those line intensities were used as 107 independent variables in a partial least squares discriminant analysis (PLS-DA) to discriminate between sterile control samples and those spiked with bacteria. In addition, the entire LIBS spectrum from 200 nm – 590 nm was input into an artificial neural network analysis with principal component analysis pre-processing (PCA-ANN) to diagnose the bacterial species once detected.The PLS-DA test possessed a 96.3% sensitivity and a 98.6% specificity for the detection of pathogenic bacteria in blood when 776 spectra from 26 filters were tested by removing one entire filter at a time from the model and testing each spectrum individually. When all the spectra obtained from a filter were averaged to enhance the signal to noise of the spectrum, 19 of 19 filters of infected blood tested positive and 7 of 7 filters with sterile blood tested negative, yielding 100% sensitivity and 100% specificity. The PCA-ANN test performed on the entire LIBS spectrum possessed a 100% sensitivity and 100% specificity when using 80% of the data to build a model and withholding 20% for cross-validation testing. The same PCA-ANN performed on each of the 19 filters individually, using the other 18 filters to build the model, possessed an average sensitivity of 85.5%, an average specificity of 95.0%, and a classification accuracy of 92.5%. These results indicate the potential usefulness of LIBS for detecting and diagnosing blood infections in a clinical setting.

Investigating Barriers to the Adoption of Energy Management Practices for Sustainable Construction Projects: SEM and ANN Approaches

This research addresses the critical challenges hindering the integration of Energy Management Practices (EMPs) within the construction industry, impeding its progress toward sustainability. Recognizing the pivotal role of EMPs in fostering sustainable practices, this study aims to fill a notable research gap by conducting a meticulous survey involving 100 industry professionals. Through the application of Partial Least Squares Structural Equation Modeling (PLS-SEM) and Artificial Neural Network (ANN) analyses, this study provides a comprehensive exploration of the intricate barriers and their interrelated dynamics within the construction sector. The findings reveal significant financial obstacles, including higher initial costs and limited financing options, underscoring the need for interventions to alleviate financial constraints. Additionally, policy and regulatory challenges, such as limited government incentives and shifting energy management rules, are identified, highlighting the necessity for stable and supportive regulatory environments to foster EMP adoptions. This research provides unique insights into the barriers hindering EMP adoption within the construction sector. The implications of this study extend beyond EMP adoption, offering a foundation for advancing sustainable practices in the construction industry. The insights gained can inform both academic research and practical decision-making, contributing to the ongoing discourse on sustainability in construction. Doi: 10.28991/CEJ-2024-010-04-015 Full Text: PDF

Agarwood oil quality identification using artificial neural network modelling for five grades

Agarwood (Aquilaria Malaccensis) oil stands out as one of the most valuable and highly sought-after oils with a hefty price tag due to its widespread use of fragrances, incense, perfumes, ceremonial practices, medicinal applications and as a symbol of luxury. However, nowadays the conventional method that rely on color alone has its limitations as it yields varying results depending on individual panelists' experiences. Hence, the quality identification system of Agarwood oil using its chemical compounds had been proposed in this study to enhance the precision of the Agarwood oil grades thus addressing the shortcomings of traditional methods. This study indicates that the primary chemical compounds of Agarwood oil encompass ɤ-Eudesmol, ar-curcumene, β-dihydroagarofuran, ϒ-cadinene, α-agarofuran, allo-aromadendrene epoxide, valerianol, α-guaiene, 10-epi-ɤ-eudesmol, β-agarofuran and dihydrocollumellarin. This study employed artificial neural network analysis with the implementation of Levenberg-Marquardt algorithm to identify the Agarwood oil grades. The study's findings revealed that this modeling system of five grades got 100% accuracies with mean square error of 0.14338×10-08. Notably, this lowest mean square error (MSE) value falls within the best hidden neuron 3. These study outcomes play a pivotal role in highlighting the Levenberg Marquardt- artificial neural network (LM-ANN) modeling that contribute to the successful of Agarwood oil quality identification using its chemical compounds.

Comparative Analysis of Ensemble Learning Techniques and ANN for Prediction of Compressive Strength of Concrete Mix

Abstract: In this study, non-linear machine learning regressors and an artificial neural network (ANN) were modeled to predict the compressive strength of concrete mixes. A dataset of 1030 mix design data points with a wide range of compressive strength varying from M10 to M80 is taken into account. The data set comprised of 8 input parameters namely- cement, water, fine & coarse aggregate, flyash, slag, superplasticizer, and age. The paper also analyzes effects on the prediction of compressive strength due to waiving of the parameters with lesser importance factors. To improve the performances of models by reducing the variance, biases, and errors, K-fold cross-validation was examined. To examine the performances of the models, metrics such as coefficient of determination (R2) and mean square error (MSE) were employed. Most of the models can predict the compressive strength with good accuracy. Out of the 7-machine learning regressors and ANN models employed, Gradient Boosting Regressor yielded the best prediction accuracy of 0.9561 with the lowest MSE value of 0.0513.

Floating-Point Quantization Analysis of Multi-Layer Perceptron Artificial Neural Networks

The impact of quantization in Multi-Layer Perceptron (MLP) Artificial Neural Networks (ANNs) is presented in this paper. In this architecture, the constant increase in size and the demand to decrease bit precision are two factors that contribute to the significant enlargement of quantization errors. We introduce an analytical tool that models the propagation of Quantization Noise Power (QNP) in floating-point MLP ANNs. Contrary to the state-of-the-art approach, which compares the exact and quantized data experimentally, the proposed algorithm can predict the QNP theoretically when the effect of operation quantization and Coefficient Quantization Error (CQE) are considered. This supports decisions in determining the required precision during the hardware design. The algorithm is flexible in handling MLP ANNs of user-defined parameters, such as size and type of activation function. Additionally, a simulation environment is built that can perform each operation on an adjustable bit precision. The accuracy of the QNP calculation is verified with two publicly available benchmarked datasets, using the default precision simulation environment as a reference.

Artificial neural network (ANN) analysis on thermophysical properties of magnetohydrodynamics flow with radiation in an arc-shaped enclosure with a rotating cylinder

The objective of this research is to examine the thermophysical features of magnetic parameter (Ha) and time step (τ) in a lid-driven cavity using a water-based Al2O3 nanofluid and the efficacy of ANN models in accurately predicting the average heat transfer rate. The Galerkin weighted residual approach is used to solve a set of dimensionless nonlinear governing equations. The Levenberg-Marquardt back propagation technique is used for training ANN using sparse simulated data. The findings of the investigation about the flow and thermal fields are shown. Furthermore, a comparative study and prediction have been conducted on the impact of manipulating factors on the average Nusselt number derived from the numerical heat transfer analysis. The findings of the research indicate that, in the absence of magnetohydrodynamics, a rise in the Hartmann number resulted in a drop in both the fluid velocity profile and magnitude. Conversely, it was observed that the temperature and Nusselt number exhibited an increase under these conditions. The mean temperature of the fluid rises as the Hartmann number drops, reaching a peak value of 0.114 when Ha = 0. The scenario where Ha = 0, representing the lack of magnetohydrodynamics, shows the highest average Nusselt number, whereas the instance with Ha = 45 presents the lowest Nusselt number. The ANN model has a high level of accuracy, as seen by an MSE value of 0.00069 and a MAE value of 0.0175, resulting in a 99% accuracy rate.

Investing in our planet: Examining retail investors' preference for green bond investment

AbstractThe green bond has emerged as an important financial instrument to advance environmentally friendly projects. While institutional investors have shown ample interest in green bonds, retail investors have lagged in their adoption. This study intends to examine the determinants of retail investors' attitude and intention toward green bond investment, utilizing the theory of planned behavior as the fundamental model and incorporating several context‐specific factors. Data from 506 Indian retail investors, representing the majority of Indian states and union territories (UTs), were collected through multi‐stage stratified random sampling. The PLS‐SEM method, coupled with artificial neural network (ANN) analysis and fuzzy set qualitative comparative analysis (fsQCA), was employed to test the hypothesized relationships, ensure the robustness of outcomes, and derive important practical insights. The findings suggest that intrinsic factors (perceived behavioral control and attitude) are superior predictors of investors' behavioral intentions relative to external factors (government policy support and social influence). Further, for green bonds to receive favorable evaluations from investors, they should display adequate financial cum environmental performance. Investors' attitude is significantly influenced by issuers' ratings and their willingness to pay premiums. Environmental concerns and perceived risk also influence investors' attitude toward green bonds, albeit with relatively lower strength. This study holds significance as it offers crucial implications for researchers, market participants, policymakers, and regulators involved in the development of the green bond market.

Multi-Objective Optimization and Prediction of Responses Using GRA and ANN in Micro-EDM of Ti6Al4V Using Different Tool Materials

An experimental investigation was conducted to evaluate the machinability of a titanium alloy (Ti6Al4V) using copper (Cu), tungsten carbide (WC), and graphite (C) tools. Voltage (V), capacitance (pF), pulse-on time (Ton), and pulse-off time (Toff) were considered as the input machining parameters, whereas the material removal rate (MRR) and tool wear rate (TWR) were considered as the output machining parameters. A Taguchi L[Formula: see text] orthogonal array and gray relational analysis (GRA) were utilized to design and optimize the machining parameters for both responses. Artificial neural network (ANN) analysis was performed to predict the experimental outcomes. Scanning electron microscopy (SEM) and energy-dispersive spectroscopy (EDS) were used to assess the surface morphology and determine the elemental composition of the machined surface. The results indicated that the optimum machining conditions for the copper tool were 150 V, 1000 pF, 15[Formula: see text]s (Ton), and 15[Formula: see text]s (Toff). However, the optimal machining conditions for the WC were 200 V, 100 pF, 25 [Formula: see text]s (Ton), and 10[Formula: see text]s (Toff), and the optimal conditions for the C were 200 V, 1000 pF, 20[Formula: see text]s (Ton), and 25[Formula: see text]s (Toff), respectively. The highest MRR achieved using the WC tool was 9.4510 mg/s, whereas the TWR of the Cu, WC, and C tools were 1.1039 mg/s, 1.0307 mg/s, and 1.2796 mg/s, respectively. The results showed that machining with the graphite tool had a higher TWR than machining with the Cu and WC tools.

Predictive modeling of copper (II) adsorption from aqueous solutions by sawdust: a comparative analysis of adaptive neuro-fuzzy interference system (ANFIS) and artificial neural network (ANN) approaches

Heavy metal ions are considered to be the most prevalent and toxic water contaminants. The objective of thois work was to investigate the effectiveness of employing the adsorption technique in a laboratory-size reactor to remove copper (II) ions from an aqueous medium. An adaptive neuro-fuzzy inference system (ANFIS) and a feed-forward artificial neural network (ANN) were used in this study. Four operational factors were chosen to examine their influence on the adsorption study: pH, contact duration, initial Cu (II) ions concentration, and adsorbent dosage. Using sawdust from wood, prediction models of copper (II) ions adsorption were optimized, created, and developed using the ANN and ANFIS models for tests. The result indicates that the determination coefficient for copper (II) metal ions in the training dataset was 0.987. Additionally, the ANFIS model’s R 2 value for both pollutants was 0.992. The findings demonstrate that the models presented a promising predictive approach that can be applied to successfully and accurately anticipate the simultaneous elimination of copper (II) and dye from the aqueous solution.

Comparative study of time-series forecasting models for wind power generation in Gujarat, India

The rapid rate of transformation in the power sector of India has placed a significant emphasis on robust grids and distributed generation units. The observable shift in the energy sector, especially in wind and solar energy, also requires smooth integration of Distributed Generation units with the existing power grid. Precise wind power generation forecast, therefore, becomes an important and complex task for the strategic planning and management of the systems. We, thus, aim towards a system that can actually provide precise wind power forecasts by applying machine learning techniques. This work proposes a comparative and comprehensive analysis of Artificial Neural Network (ANN), Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and Autoregressive Integrated Moving Average (ARIMA) model. The experimentations and modelling are performed considering meteorological and historical power generation data. The study is concentrated in Kutch, Gujarat and is validated on the data collected from the Central Electricity Authority (CEA), India for power generation data and weather data collected from regional weather centres. The findings show that ARIMA outperforms the other models for non-linear data in multivariate analysis, with a MAPE score of less than 5.5 on the prediction dataset.

Perceptions of how occupants adopt water conservation behaviors under psychosocial processes: A complementary dual-stage SEM-ANN perspective

This study delved into socio-psychological determinants driving residential water curtailment actions, an area relatively overlooked in the existing literature. The theory of planned behavior was innovatively expanded with moral norms, perceived expectations, self-identity, and risk-related components. The proposed framework was empirically examined by conducting an online self-administered survey of 343 citizens residing in Isfahan, Iran. What sets this research apart is introducing a new multi-analytical hybrid approach integrating linear and non-linear techniques to leverage their respective strengths. Artificial neural network (ANN), coupled with Sobol sensitivity analysis, was employed to robustly validate structural equation modeling (SEM) results regarding the framework's explanatory power and predictors' ranking. Through providing reliable performance, the ANN analysis confirmed SEM findings, revealing that the significant influences of moral norms and self-identity on intention are mediated by attitude and perceived behavioral control. Moreover, self-identity and attitude were the strongest direct predictors of behavior and intention. Several policy recommendations were proposed, with the following highlighted as crucial: (1) fostering favorable water conservation attitudes, which entails targeting individuals' moral considerations and inducing self-reward feelings within the community; (2) cultivating a sense of water conservation identity among citizens; and (3) boosting individuals’ self-confidence and facilitating their engagement in water conservation behaviors.

Embodimetrics: A Principal Component Analysis Study of the Combined Assessment of Cardiac, Cognitive and Mobility Parameters.

There is a growing body of literature investigating the relationship between the frequency domain analysis of heart rate variability (HRV) and cognitive Stroop task performance. We proposed a combined assessment integrating trunk mobility in 72 healthy women to investigate the relationship between cognitive, cardiac, and motor variables using principal component analysis (PCA). Additionally, we assessed changes in the relationships among these variables after a two-month intervention aimed at improving the perception-action link. At baseline, PCA correctly identified three components: one related to cardiac variables, one to trunk motion, and one to Stroop task performance. After the intervention, only two components were found, with trunk symmetry and range of motion, accuracy, time to complete the Stroop task, and low-frequency heart rate variability aggregated into a single component using PCA. Artificial neural network analysis confirmed the effects of both HRV and motor behavior on cognitive Stroop task performance. This analysis suggested that this protocol was effective in investigating embodied cognition, and we defined this approach as "embodimetrics".

Comparative Analysis of Linear Models and Artificial Neural Networks for Sugar Price Prediction

Sugar is an important commodity that is used beyond the food industry. It can be produced from sugarcane and sugar beet, depending on the region. Prices worldwide differ due to high volatility, making it difficult to estimate their forecast. Thus, the present work aims to predict the prices of kilograms of sugar from four databases: the European Union, the United States, Brazil, and the world. To achieve this, linear methods from the Box and Jenkins family were employed, together with classic and new approaches of artificial neural networks: the feedforward Multilayer Perceptron and extreme learning machines, and the recurrent proposals Elman Network, Jordan Network, and Echo State Networks considering two reservoir designs. As performance metrics, the MAE and MSE were addressed. The results indicated that the neural models were more accurate than linear ones. In addition, the MLP and the Elman networks stood out as the winners.

Thermosonication Processing of Purple Onion Juice (Allium cepa L.): Anticancer, Antibacterial, Antihypertensive, and Antidiabetic Effects

Onion (Allium cepa L.) juice is an important product used in gastronomy and food formulations. The first objective of this study was to optimize the content of bioactive compounds in purple onion juice (POJ) after the thermosonication process using response surface methodology (RSM) and artificial neural network (ANN) application models. Second, the anticancer, antibacterial, antihypertensive, and antidiabetic effects of POJ obtained after thermal pasteurization (P-POJ) or thermosonication (TS-POJ) were investigated after obtaining the ANN and RSM analysis reports. The optimization process for TS-POJ was carried out at 44 °C, for 13 min, with a 68% amplitude. The findings demonstrated that the angiotensin-converting enzyme (ACE) inhibition level was greater in TS-POJ samples than in the untreated control (C-POJ) sample (p > 0.05). C-POJ, TS-POJ, and P-POJ exhibited the inhibition of cell proliferation in vitro in a dose-dependent manner in lung (A549), cervical (HeLa), and colon cancer cells following 24 h incubation. Thermosonication or thermal pasteurization did not markedly affect the cell proliferation of the examined cancer cells compared to the untreated control group. While no antibacterial effect was observed with low concentrations of samples, they showed an antibacterial effect at pure concentrations (100%). The thermosonication treatment for processing purple onion juice was successful in this study’s results.

Experimental and ANN Analysis of Shearing Rate Effects on Coarse Sand Crushing

The present study analyzes laboratory experiments on how shearing rate affects the shear strength and crushability of natural coarse sand, employing artificial neural network (ANN) analysis. This study tested three different coarse sands obtained from the crushing of natural rocks: Black Virgin Tuff, weathered Zeolitic Tuff, and calcareous limestone. The behavior of crushed sand specimens with consistent grading, which passed through sieve #4 and were retained on sieve #8, was analyzed using a direct shear box. The specimens were subjected to varied normal loads and shearing speeds to examine their behavior at different relative densities. The test results were analyzed using ANN to investigate the significance of shearing rates on shearing strength parameters, specifically internal mobilized peak friction, the constant volume (residual) internal friction angle, and the consequence of shearing rate on the particle's breakage index. The selected normal (Gaussian) rate significantly affected both the shear strength parameters and breakage. The loading rate increased both shear strength parameters and particle breakage. Therefore, it's highly recommended to maintain secure sets of shear strength values and comprehensive test data for assessing parameters at typical strain rates, prioritizing using slower rates whenever possible. Doi: 10.28991/CEJ-2024-010-03-011 Full Text: PDF

Investigation of optimal machining Monel 400 superalloy considering carbon emissions using FEM, regression and ANN methods

In this study, the optimal machining parameters for the machining of Monel 400 superalloy considering carbon emission were investigated. Monel 400 stands out among nickel alloys with its resistance to high corrosive environments. First, the Johnson-Cook material model for Monel 400 was developed for FEM simulations. Then, FEM simulations were performed using seven different machining parameters, including cutting and geometric, and validated with experimental results in the literature. The obtained FEM results were trained and predicted by regression and Artificial Neural Network (ANN) analysis. With this study, the Johnson-Cook material model of Monel 400 superalloy and the optimal machining parameters considering carbon emission with different cutting and geometric machining parameters are provided to the literature. As a result, it was determined that the most effective machining parameters in terms of cutting force were 139.04% increase in depth of cut, 74.23% increase in feed and 9.57% decrease in side rake angle. In terms of carbon emission, the most effective machining parameters were found to be −338.9g reduction in depth of cut, −294.3g reduction in feed rate and −171.9g reduction in cutting speed. In terms of training and prediction of FEM results, it was observed that the regression model was 89.19% for cutting force data and 95.39% for carbon emission, while the ANN model was 98.081% for cutting force data and 99.978% for carbon emission. Optimal machining parameters were determined by ANN method. The optimal machining parameters (cutting speed, feed, cutting depth, side rake angle, back rake angle, lead angle and nose radius) of Monel 400 superalloy with the lowest cutting force and carbon emission were determined as 100 m/min, 0.15mm/rev, 0.375 mm, 0°, 5°, 0°, 0°, 0°, 0.4 mm and 100 m/min, 0.2mm/rev, 0.25 mm, −5°, 5°, 0°, 0.8 mm, respectively.

Analysis of Artificial Neural Networks on Predicting Dementia

Dementia, a chronic neurological disorder characterized by memory loss and impaired cognitive ability, has affected tens of millions of people globally and projections estimate this number to rise in the coming decade. Despite its widespread impact, dementia remains one of many diseases with no true cure or definitive method of diagnosis. This study aims to explore the question of how artificial neural networks can be used to predict dementia. Analyzing current research on neural networks for disease diagnosis highlights a notable gap, known as the "black box conundrum," indicating the challenge of tracing steps leading to a given output. Another gap in the literature is the absence of comparative data on factors influencing successful neural network algorithms. Conducting a comprehensive cross-continental comparative analysis across North America, Asia, and Europe, and utilizing an experimental design with the UCI ML Repository’s Parkinson’s Database, this study specifically aims to fill the identified gaps. The findings reveal key elements for successful neural networks: a data sample exceeding 100, clinically intelligible data values, and a backpropagation model. In the experimental phase, a comparison between a backpropagation model and logistic regression, paired with Shapley value analysis, elucidated a 94.87% accuracy for the former, surpassing the latter's 92.3%. These findings extend beyond dementia, providing valuable insights for neural network applications in all healthcare settings. Future research endeavors should focus on refining and optimizing neural networks with the ultimate goal of achieving 100% accuracy and maximized efficiency.

Artificial Neural Network and Multiple Regression Analysis Applied to 2D-QSAR Studies: the Case of Imidazolidine-2,4-dione as PTP1B Inhibitors

In this investigation, we undertook a comprehensive quantitative structure-activity relationship (QSAR) analysis using both modern artificial neural network (ANN) methods and classical multiple linear regression methods (MLR). Our primary focus was on revealing the intricate connection between antidiabetic activity and the molecular structure of thirty-nine imidazolidine- 2,4-dione derivatives. The B3LYP hybrid functional and 6-31G (d) basis set computed electronic properties at the quantum level. Rigorous benchmarking against experimental data validated the reliability of our quantum theory approach. Our statistical model effectively predicted activities closely aligned with experimental antidiabetic activities, quantified by the IC50 values. They also revealed a significant superiority of the ANN architecture (6-4-1) over the MLR method. This study stands as a meaningful contribution to the field, providing valuable insights for designing new antidiabetic drugs, particularly as potential inhibitors of tyrosine phosphate 1B (PTP1B). The explicit articulation of our primary aim and emphasis on the study’s significance underscore its potential impact on advancing drug development within the realm of antidiabetic therapeutics.

An experimental study and ANN analysis of utilizing ammonia as a hydrogen carrier by real-time emulsion fuel injection to promote cleaner combustion

This work utilized Real-Time Emulsion Fuel Injection (RTEFI) methods to incorporate emulsion fuel, which included citronella oil (20%), ammonia (10%), and water (5%), into diesel fuel without the addition of surfactants. The primary aim of this study was to investigate the influence of hydrogen presence in liquid ammonia on diesel engines. This was accomplished by using a water-emulsified diesel-biodiesel blend and forecasting performance using an artificial neural network (ANN). Citronella oil (20%) was blended with standard fuel (SD), followed by the addition of 10% ammonia and 5% water, labeled as SD20CO, SD20CO10A, and SD20CO10A5W, respectively. The results indicated that the brake thermal efficiency of SD20CO10A5W is comparable to that of diesel. The fuel combination significantly decreased hydrocarbon emissions by 70–75%, smoke by 10–15%, and oxides of nitrogen by 2–5%, while increasing carbon dioxide by 3–8% under all load situations compared to other fuels. The trained ANN models had a coefficient of determination of 97%, with values ranging from 0.9076 to 0.9965. Additionally, the models exhibited low mean absolute percentage error values, ranging from 0.98% to 4.26%. The analytical results provide strong evidence supporting the efficacy of SD20CO10A5W, ultimately indicating that it is a feasible substitute fuel for diesel engines.

Machine Learning Offers Insights into the Impact of In Vitro Drought Stress on Strawberry Cultivars

This study aimed to assess the susceptibility of three strawberry cultivars (“Festival”, “Fortuna”, and “Rubygem”) to drought stress induced by varying polyethylene glycol (PEG) concentrations in the culture medium. Plantlets were cultivated on a solid medium supplemented with 1 mg/L BAP, and PEG concentrations (0, 2, 4, and 6 mg/L) were introduced to simulate drought stress. Morphological changes were observed, and morphometric analysis was conducted. Additionally, artificial neural network (ANN) analysis and machine learning approaches were integrated into this study. The results showed significant effects of PEG concentrations on plant height and multiplication coefficients, highlighting genotype-specific responses. This study employed various machine learning models, with random forest consistently demonstrating superior performance. Our findings revealed the random forest model outperformed others with a remarkable global diagnostic accuracy of 91.164%, indicating its superior capability in detecting and predicting water stress effects in strawberries. Specifically, the RF model excelled in predicting root length and the number of roots for “Festival” and “Fortuna” cultivars, demonstrating its reliability across different genetic backgrounds. Meanwhile, for the “Rubygem” cultivar, the multi-layer perceptron (MLP) and Gaussian process (GP) models showed particular strengths in predicting proliferation and plant height, respectively. These findings highlight the potential of ML models, particularly RF, to enhance agricultural breeding and cultivation strategies through accurate phenotypic predictions, suggesting a promising direction for future research to improve these predictions further. This research contributes to understanding strawberry responses to drought stress and emphasizes the potential of machine learning in predicting plant characteristics.