Performance Of Machine Learning Models Research Articles

Comparing Human-Level and Machine Learning ModelPerformance in White Blood Cell MorphologyAssessment.

There is an increasing research focus on the role of machine learning in the haematology laboratory, particularly in blood cell morphologic assessment. Human-level performance is an important baseline and goal for machine learning. This study aims to assess the interobserver variability and human-level performance in blood cell morphologic assessment. A dataset of 1000 single white blood cell images were independently labelled by 10 doctors and morphology scientists. Interobserver variability was calculated using Fleiss' kappa. Observers' labels were then separated into consensus labels used to determine ground truth, and performance labels used to assess observer performance. A machine learning model was trained and assessed using the same cell images. Explainability images (XRAI and IG) were generated for each of the test images. The Fleiss kappa for all 10 observers was 0.608, indicating substantial agreement between observers. The accuracy of human observers was 95%, with sensitivity 72% and specificity 97%. The accuracy of the machine learning model was 95%, with sensitivity 71% and specificity 97%. The model shared similar performance across labels when compared to humans. Explainability metrics demonstrated that the machine learning model was able to differentiate between the cytoplasm and nucleus of the cells, and used these features to perform predictions. The substantial, though not perfect, agreement between human observers highlights the inherent subjectivity in white blood cell morphologic assessment. A machine learning model performed similarly to human observers in single white blood cell identification. Further research is needed to compare human-level and machine learning performance in ways that more closely reflect the typical process of morphologic assessment.

Performance of Machine Learning Models to Predict Customer Satisfaction Scores for Lao National Convention Center

This research paper investigates the effectiveness of various machine learning (ML) models in predicting customer satisfaction scores for the Lao National Convention Center (LNCC). By utilizing a dataset containing customer feedback, demographic data, facilities, and service-related information, the study aims to identify the best performing model for predicting customer satisfaction. In this paper, the performance of ML to predict customer satisfaction scores from the questionnaire survey dataset is evaluated. The customer satisfaction score is categorized into five classes: strongly agree, agree, neutral, disagree and strongly disagree. The analysis will compare the performance of different models, including linear regression (LR), k-nearest neighbor (KNN), support vector machines (SVM), decision trees (DT) and random forest (RF). The results show that SVM model achieved the highest accuracy rate of 95.91%, followed by KNN and LR with an accuracy rate of 95.55% and 95.49%, respectively. The findings of this study have important implications for the use of ML in improving the service-related information for LNCC and providing valuable insights for decision-makers and developers.

An intelligent learning system based on electronic health records for unbiased stroke prediction

Stroke has a negative impact on people’s lives and is one of the leading causes of death and disability worldwide. Early detection of symptoms can significantly help predict stroke and promote a healthy lifestyle. Researchers have developed several methods to predict strokes using machine learning (ML) techniques. However, the proposed systems have suffered from the following two main problems. The first problem is that the machine learning models are biased due to the uneven distribution of classes in the dataset. Recent research has not adequately addressed this problem, and no preventive measures have been taken. Synthetic Minority Oversampling (SMOTE) has been used to remove bias and balance the training of the proposed ML model. The second problem is to solve the problem of lower classification accuracy of machine learning models. We proposed a learning system that combines an autoencoder with a linear discriminant analysis (LDA) model to increase the accuracy of the proposed ML model for stroke prediction. Relevant features are extracted from the feature space using the autoencoder, and the extracted subset is then fed into the LDA model for stroke classification. The hyperparameters of the LDA model are found using a grid search strategy. However, the conventional accuracy metric does not truly reflect the performance of ML models. Therefore, we employed several evaluation metrics to validate the efficiency of the proposed model. Consequently, we evaluated the proposed model’s accuracy, sensitivity, specificity, area under the curve (AUC), and receiver operator characteristic (ROC). The experimental results show that the proposed model achieves a sensitivity and specificity of 98.51% and 97.56%, respectively, with an accuracy of 99.24% and a balanced accuracy of 98.00%.

Pre-operative prediction of BCR-free survival with mRNA variables in prostate cancer.

Technological innovation yielded opportunities to obtain mRNA expression data for prostate cancer (PCa) patients even prior to biopsy, which can be used in a precision medicine approach to treatment decision-making. This can apply in particular to predict the risk of, and time to biochemical recurrence (BCR). Most mRNA-based models currently proposed to this end are designed for risk classification and post-operative prediction. Effective pre-operative prediction would facilitate early treatment decision-making, in particular by indicating more appropriate therapeutic pathways for patient profiles who would likely not benefit from a systematic prostatectomy regime. The aim of this study is to investigate the possibility to leverage mRNA information pre-operatively for BCR-free survival prediction. To do this, we considered time-to-event machine learning (ML) methodologies, rather than classification models at a specific survival horizon. We retrospectively analysed a cohort of 135 patients with clinical follow-up data and mRNA information comprising over 26,000 features (data accessible at NCBI GEO database, accession GSE21032). The performance of ML models including random survival forest, boosted and regularised Cox models were assessed, in terms of model discrimination, calibration, and predictive accuracy for overall, 3-year and 5-year survival, aligning with common clinical endpoints. Results showed that the inclusion of mRNA information could yield a gain in performance for pre-operative BCR prediction. ML-based time-to-event models significantly outperformed reference nomograms that used only routine clinical information with respect to all metrics considered. We believe this is the first study proposing pre-operative transcriptomics models for BCR prediction in PCa. External validation of these findings, including confirmation of the mRNA variables identified as potential key predictors in this study, could pave the way for pre-operative precision nomograms to facilitate timely personalised clinical decision-making.

Resampling approach for imbalanced data classification based on class instance density per feature value intervals

In practical applications, imbalanced datasets significantly degrade the classification performance of machine learning models. However, most conventional resampling approaches fall short in adequately addressing the varying contributions of individual features to the classification model. In response to this defect, this study introduces three novel resampling approaches. The first approach, Oversampling based on class instance density per feature value intervals (OCF), focuses on augmenting the dataset. The second approach, Undersampling based on class instance density per feature value intervals (UCF), seeks to reduce dataset size. The third approach, Hybrid sampling based on class instance density per feature value intervals (HSCF), which can perform oversampling and undersampling simultaneously. These approaches categorize feature value into different intervals based on their varying information content, calculate class instance densities within these intervals, and generate feature values in intervals with high discriminative information. Subsequently, these generated features are combined to synthesize minority class data, effectively achieving oversampling. Additionally, the study combines class instance density and feature importance to identify majority class data at the classification boundary with minimal contribution and subsequently executes undersampling. The flexibility to adjust sampling ratios and the integration of OCF and UCF enable the implementation of hybrid sampling. Finally, experiments on the benchmark dataset demonstrate the superiority and effectiveness of the proposed method. Furthermore, it is observed that the method proposed in this study enhances the feature dividing capability of decision tree classifiers. Hence, the best results are achieved when working in synergy with decision tree classifiers, leading to the most significant improvements in classification performance. All codes have been published at https://github.com/Wangfeiopen/HSCF.

ILKD: An Incremental Learning and Knowledge Distillation Framework for Coronavirus Risk Prediction

Abstract The coronavirus pandemic has seriously affected public health and social order. Prediction methods based on machine learning can identify the infectivity phenotype and pandemic risk of coronavirus. Currently, six types of coronaviruses that infect humans have been discovered, with significant differences in viral genome sequences. Continuous genetic variation of the virus will lead to reduced performance of machine learning models and potential learning forgetting. To solve this challenge, we propose an incremental learning and knowledge distillation framework (ILKD). First, we employ Dna2Vec to extract virus features and encode the virus sequence into virus feature vector. Second, we use hierarchical clustering to continuously identify new coronavirus groups. Third, ILKD employ a combined strategy of incremental learning and knowledge distillation to transform the Back Propagation (BP) neural network to continuously learn and predict the phenotypes of human-to-human coronavirus infection. Experimental results show that ILKD can effectively alleviate the learning forgetting phenomenon. Further analysis reveals ILKD has better performance than other incremental learning models, and has important public health application value.

Optimization of drug solubility inside the supercritical CO2 system via numerical simulation based on artificial intelligence approach

In this research paper, we explored the predictive capabilities of three different models of Polynomial Regression (PR), Extreme Gradient Boosting (XGB), and LASSO to estimate the density of supercritical carbon dioxide (SC-CO2) and the solubility of niflumic acid as functions of the input variables of temperature and pressure. The optimization of hyperparameters for these models is achieved using the innovative Barnacles Mating Optimizer (BMO) algorithm. For SC-CO2 density estimation, PR exhibits remarkable accuracy, showing an R-squared value of 0.99207 for data fitting. XGB performs admirably with an R2 of 0.92673, while LASSO model demonstrates good predictive ability, showing an R2 of 0.81917. Furthermore, we assess the models’ performance in predicting the solubility of niflumic acid. PR exhibits excellent predictive capabilities with an R2 of 0.96949. XGB also delivers strong performance, yielding an R-squared score of 0.92961. LASSO performs well, achieving an R-squared score of 0.82094. The results indicated promising performance of machine learning models and optimizer in estimating drug solubility in supercritical CO2 as the solvent applicable for pharmaceutical industry.

A generalized machine learning framework to estimate fatigue life across materials with minimal data

In this research, a generalized machine learning (ML) framework is proposed to estimate the fatigue life of epoxy polymers and additively manufactured AlSi10Mg alloy materials, leveraging their failure surface void characteristics. An extreme gradient boosting algorithm-based ML framework encompassing Synthetic Minority Over-sampling TEchnique (SMOTE), categorical data encoding, and external loop cross-validation is developed to evaluate the fatigue life across materials. The influence of different training strategies based on materials, input features, encoding method, and data standardization on the model performance is explored. Additionally, the importance of anti-data-leakage and anti-overfitting measures over the ML model performance is addressed. The result shows that the data-leakage-free, external loop cross-validated model can estimate the fatigue life of selective epoxy polymers and metal alloys with an average R2 of 0.71 ± 0.06 using a mere 12 to 27 experimental data points per material category. Whereas the model trained with data-leakage and overfitting results in high R2 of 0.9.

Spatial Prediction of Landslide Susceptibility using Various Machine Learning Based Binary Classification Methods

ABSTRACT This study compares the performance of various machine learning models for predicting landslide susceptibility using a geospatial dataset from the Lai Chau province, Vietnam. The dataset consisted of 850 landslide locations and ten influencing factors. Eight models, including Forest by Penalizing Attributes (FPA), Bagging-based FPA (BFPA), Artificial Neural Network (ANN), Logistic Regression (LR), Support Vector Machine (SVM), Multilayer Perceptron (MLP), Bayesian Network (BN), and Naïve Bayes (NB), were evaluated based on different evaluation metrics. The results revealed distinct variations in the performance of the models across the evaluation metrics. Based on the overall rankings, the ensemble BFPA model with sensitivity=90%, specificity= 95.98%, accuracy=92.86%, Kappa=0.857, and area under the curve=0.98 demonstrated the highest capability in predicting landslide susceptibility. It was followed by BN, FPA, MLP, ANN, SVM, LR, and NB. These findings suggest that the BFPA model outperformed other models in terms of its ability to accurately identify potential landslide-prone areas in the study region. This study provides valuable insights into the comparative analysis of machine learning models for landslide susceptibility prediction. Furthermore, it supports the effectiveness of ensemble models for landslide susceptibility prediction, which can inform decision-makers, land-use planners, and disaster management agencies in making informed decisions regarding potential landslide hazards and implementing effective risk mitigation strategies in Vietnam. Continued research in this area will enhance our understanding of machine learning techniques and their application in mitigating the impact of landslides on society and the environment.

An ensemble deep learning model for fast classification of Twitter spam

Twitter spam needs to be detected and arrested quickly. The paper examines methods for classification of spam in terms of determination of important features, comparative performance of classification models, and improvement in time performance for classification. It presents a conceptualization of several novel rich, deep, and naïve features. The extraction processes for rich and deep features increase the time complexity of spam classification. To address this, the proposed model selectively segregates and combines features to enable near real-time processing. This supersedes the time performance of standard machine learning and deep learning models, with no compromise on the quality of classification.

Feasibility study on discrimination of Polygonatum kingianum origins by NIR and MIR spectra data.

Most existing studies have focused on identifying the origin of species with protected geographical indications while neglecting to determine the proximate geographical origin of different species. In this study, we investigated the feasibility of using near- and mid-infrared spectroscopy to identify the origin of 156 Polygonatum kingianum samples from six regions in Yunnan, China. In this work, spectral images of different modes reveal more information about the P. kingianum. Comparing the performance of traditional machine learning models according to single spectrum and data fusion, the middle-level data fusion-principal component model has the best performance, and its sensitivity, specificity, and accuracy are all 1, and the model has the least number of variables. The residual convolutional neural network (ResNet) model constructed in the 1050-850cm-1 band confirms that fewer variables are beneficial in improving the accuracy of the model. In conclusion, this study verifies the feasibility of the proposed strategy and establishes a practical model to determine the source of P. kingianum.

DEVELOPING MACHINE LEARNING APPLICATION FOR EARLY CARDIOVASCULAR DISEASE (CVD) RISK DETECTION IN FIJI: A DESIGN SCIENCE APPROACH

CVD (cardiovascular disease) has become a significant contributor to premature deaths for many years in Fiji. CVD's late detection also significantly impacts annual deaths and casualties. Currently, Fiji lacks diagnosis tools that would enable people to know their risk levels. In this paper, a machine learning mobile application was developed that can be easily accessible to the local population for early prediction of CVD risk. The design science approach was used to guide the development of the application. The design process involved identifying the problem and motivation, setting objectives, creating a machine-learning mobile application for medical record analysis, demonstrating the application to selected participants, evaluating its usability and the machine-learning model's performance, and communicating the findings. The results revealed that the application proposed in this paper is an effective tool for CVD prediction in Fiji.

Classification of Vase Life Day Based on Petal Colorimetric Data in Cut Lisianthus Using AutoML

This study investigated the potential of petal colorimetric data to classify vase life stages in cut lisianthus flowers (Eustoma grandiflorum). We analyzed the changes in the petal color space over time, focusing on the b* value as an indicator of senescence due to increasing yellowing caused by copigmentation. A comparative analysis was conducted between two cultivation methods: soil (S) and hydroponic (H) cultivation. The objective was to evaluate the performance of machine learning models trained to classify vase life stages based on petal color data. Automated machine learning models exhibited better performance in H-cultivated cut flowers, effectively distinguishing days within the vase life stages from Days 1 to 14 for H cultivation. Cut flowers cultivated under S conditions showed less variation in the color space from Days 1 to 9, maintaining a relatively uniform color range. This made it more difficult to distinguish the vase life stages compared to H cultivation. These findings demonstrate that petal color metrics can serve as reliable indicators of cut flower senescence and potentially facilitate nondestructive methods for classifying vase life stages. This technology holds promise for wider applications in the floriculture industry, improving quality control, and extending the vase life of various cut-flower crops.

Polynomial-SHAP analysis of liver disease markers for capturing of complex feature interactions in machine learning models

Liver disease diagnosis is pivotal for effective patient management, and machine learning techniques have shown promise in this domain. In this study, we investigate the impact of Polynomial-SHapley Additive exPlanations analysis on enhancing the performance and interpretability of machine learning models for liver disease classification. Our results demonstrate significant improvements in accuracy, precision, recall, F1_score, and Matthews correlation coefficient across various algorithms when polynomial- SHapley Additive exPlanations analysis is applied. Specifically, the Light Gradient Boosting Machine model achieves exceptional performance with 100 % accuracy in both scenarios. Furthermore, by comparing the results obtained with and without the approach, we observe substantial differences in the performance, highlighting the importance of incorporating Polynomial-SHapley Additive exPlanations analysis for improved model performance. The Polynomial features and SHapley Additive exPlanations values also enhance the interpretability of machine learning models by capturing complex feature interactions, enabling users to gain deeper insights into the underlying mechanisms driving the diagnosis. Moreover, data rebalancing using Synthetic Minority Over-sampling Technique and parameter tuning were employed to optimize the performance of the models. These findings underscore the significance of employing this analytical approach in machine-learning-based diagnostic systems for liver diseases, offering superior performance and enhanced interpretability for informed decision-making in clinical practice.

Evaluating machine learning performance in predicting sodium adsorption ratio for sustainable soil-water management in the eastern Mediterranean

Soil salinization is a critical global issue for sustainable agriculture, impacting crop yields and posing a threat to achieving the Sustainable Development Goal (SDG) of ensuring food security. It is necessary to monitor it in detail and uncover its underlying factors at a regional scale. In this context, the present study aimed to evaluate soil health in the eastern Mediterranean region by using the Sodium Adsorption Ratio (SAR) as an indicator of soil salinity in three distinct soil horizons. The main objective of the research was to evaluate the performance of four machine learning (ML) models, including Random Forest (RF), Nu Support Vector Regression (NuSVR), Artificial Neural Network-Multi Layer Perceptron (ANN-MLP), and Gradient Boosting Regression (GBR), for accurate prediction of SAR following the Recursive Feature Elimination (RFE) as a feature selection method. Moreover, SHapely Additive exPlanations (SHAP) was applied as sensitivity analysis to identify the most influential covariates. Main findings of the research revealed that the average clay content in the surface horizon (H10-25cm) was 50.5% ± 10.4, which significantly increased to 57.5% ± 8.7 (p < 0.05). No significant mean differences were detected between the studied horizons for SAR and Na+. ML output revealed that NuSVR outperformed other algorithms in accurately predicting outcomes during both the training and testing stages. Moreover, Scenario 2 (SC2) with seven selected features from the RFE method facilitated highly accurate SAR predictions. Overall, the performance of ML models is ranked as NuSVR > GBR > ANN-MLP > RF. Lastly, SHAP sensitivity analysis identified CEC, Ca+2, Mg+2, and Na+ as the most influential variables for SAR prediction in both the training and testing stages. Hence, the research yielded valuable insights for efficient agricultural soil management at a regional level using state-of-the-art technology.

Predicting diabetes in adults: identifying important features in unbalanced data over a 5-year cohort study using machine learning algorithm

BackgroundImbalanced datasets pose significant challenges in predictive modeling, leading to biased outcomes and reduced model reliability. This study addresses data imbalance in diabetes prediction using machine learning techniques. Utilizing data from the Fasa Adult Cohort Study (FACS) with a 5-year follow-up of 10,000 participants, we developed predictive models for Type 2 diabetes.MethodsWe employed various data-level and algorithm-level interventions, including SMOTE, ADASYN, SMOTEENN, Random Over Sampling and KMeansSMOTE, paired with Random Forest, Gradient Boosting, Decision Tree and Multi-Layer Perceptron (MLP) classifier. We evaluated model performance using F1 score, AUC, and G-means—metrics chosen to provide a comprehensive assessment of model accuracy, discrimination ability, and overall balance in performance, particularly in the context of imbalanced datasets.Resultsour study uncovered key factors influencing diabetes risk and evaluated the performance of various machine learning models. Feature importance analysis revealed that the most influential predictors of diabetes differ between males and females. For females, the most important factors are triglyceride (TG), basal metabolic rate (BMR), and total cholesterol (CHOL), whereas for males, the key predictors are body Mass Index (BMI), serum glutamate Oxaloacetate Transaminase (SGOT), and Gamma-Glutamyl (GGT). Across the entire dataset, BMI remains the most important variable, followed by SGOT, BMR, and energy intake. These insights suggest that gender-specific risk profiles should be considered in diabetes prevention and management strategies. In terms of model performance, our results show that ADASYN with MLP classifier achieved an F1 score of 82.17 ± 3.38, AUC of 89.61 ± 2.09, and G-means of 89.15 ± 2.31. SMOTE with MLP followed closely with an F1 score of 79.85 ± 3.91, AUC of 89.7 ± 2.54, and G-means of 89.31 ± 2.78. The SMOTEENN with Random Forest combination achieved an F1 score of 78.27 ± 1.54, AUC of 87.18 ± 1.12, and G-means of 86.47 ± 1.28.ConclusionThese combinations effectively address class imbalance, improving the accuracy and reliability of diabetes predictions. The findings highlight the importance of using appropriate data-balancing techniques in medical data analysis.

Machine learning-based individualized survival prediction model for prognosis in osteosarcoma: Data from the SEER database.

Patient outcomes of osteosarcoma vary because of tumor heterogeneity and treatment strategies. This study aimed to compare the performance of multiple machine learning (ML) models with the traditional Cox proportional hazards (CoxPH) model in predicting prognosis and explored the potential of ML models in clinical decision-making. From 2000 to 2018, 1243 patients with osteosarcoma were collected from the Surveillance, Epidemiology, and End Results (SEER) database. Three ML methods were chosen for model development (DeepSurv, neural multi-task logistic regression [NMTLR]) and random survival forest [RSF]) and compared them with the traditional CoxPH model and TNM staging systems. 871 samples were used for model training, and the rest were used for model validation. The models' overall performance and predictive accuracy for 3- and 5-year survival were assessed by several metrics, including the concordance index (C-index), the Integrated Brier Score (IBS), receiver operating characteristic curves (ROC), area under the ROC curves (AUC), calibration curves, and decision curve analysis. The efficacy of personalized recommendations by ML models was evaluated by the survival curves. The performance was highest in the DeepSurv model (C-index, 0.77; IBS, 0.14; 3-year AUC, 0.80; 5-year AUC, 0.78) compared with other methods (C-index, 0.73-0.74; IBS, 0.16-0.17; 3-year AUC, 0.73-0.78; 5-year AUC, 0.72-0.78). There are also significant differences in survival outcomes between patients who align with the treatment option recommended by the DeepSurv model and those who do not (hazard ratio, 1.88; P < .05). The DeepSurv model is available in an approachable web app format at https://survivalofosteosarcoma.streamlit.app/. We developed ML models capable of accurately predicting the survival of osteosarcoma, which can provide useful information for decision-making regarding the appropriate treatment.

An automated approach to predict diabetic patients using KNN imputation and effective data mining techniques

Diabetes is thought to be the most common illness in underdeveloped nations. Early detection and competent medical care are crucial steps in reducing the effects of diabetes. Examining the signs associated with diabetes is one of the most effective ways to identify the condition. The problem of missing data is not very well investigated in existing works. In addition, existing studies on diabetes detection lack accuracy and robustness. The available datasets frequently contain missing information for the automated detection of diabetes, which might negatively impact machine learning model performance. This work suggests an automated diabetes prediction method that achieves high accuracy and effectively manages missing variables in order to address this problem. The proposed strategy employs a stacked ensemble voting classifier model with three machine learning models. and a KNN Imputer to handle missing values. Using the KNN imputer, the suggested model performs exceptionally well, with accuracy, precision, recall, F1 score, and MCC of 98.59%, 99.26%, 99.75%, 99.45%, and 99.24%, respectively. In two scenarios one with missing values eliminated and the other with KNN imputer, the study thoroughly compared the suggested model with seven other machine learning techniques. The outcomes demonstrate the superiority of the suggested model over current state-of-the-art methods and confirm its efficacy. This work demonstrates the capability of KNN imputer and looks at the problem of missing values for diabetes detection. Medical professionals can utilize the results to improve care for diabetes patients and discover problems early.

Beach nourishment for coastal aquifersimpacted by climate change and population growth using machine learning approaches

Groundwater in coastal regions is threatened by saltwater intrusion (SWI). Beach nourishment is used in this study to manage SWI in the Biscayne aquifer, Florida, USA, using a 3D SEAWAT model nourishment considering the future sea level rise and freshwater over-pumping. The present study focused on the development and comparative evaluation of seven machine learning (ML) models, i.e., additive regression (AR), support vector machine (SVM), reduced error pruning tree (REPTree), Bagging, random subspace (RSS), random forest (RF), artificial neural network (ANN) to predict the SWI using beach nourishment. The performance of ML models was assessed using statistical indicators such as coefficient of determination (R2), Nash–Sutcliffe efficiency (NSE), means absolute error (MAE), root mean square error (RMSE), and root relative squared error (RRSE) along with the graphical inspection (i.e., Radar and Taylor diagram). The findings indicate that applying SVM, Bagging, RSS, and RF models has great potential in predicting the SWI values with limited data in the study area. The RF model emerged as the best fit and closely matched observed values; it obtained R2 (0.999), NSE (0.999), MAE (0.324), RRSE (0.209), and RMSE (0.416) during the testing process. The present study concludes that the RF model could be a valuable tool for accurate predictions of SWI and effective water management in coastal areas.

Multiple remotely sensed datasets and machine learning models to predict chlorophyll-a concentration in the Nakdong River, South Korea.

The Nakdong River is a crucial water resource in South Korea, supplying water for various purposes such as potable water, irrigation, and recreation. However, the river is vulnerable to algal blooms due to the inflow of pollutants from multiple points and non-point sources. Monitoring chlorophyll-a (Chl-a) concentrations, a proxy for algal biomass is essential for assessing the trophic status of the river and managing its ecological health. This study aimed to improve the accuracy and reliability of Chl-a estimation in the Nakdong River using machine learning models (MLMs) and simultaneous use of multiple remotely sensed datasets. This study compared the performances of four MLMs: multi-layer perceptron (MLP), support vector machine (SVM), random forest (RF), and eXetreme Gradient Boosting (XGB) using three different input datasets: (1) two remotely sensed datasets (Sentinel-2 and Landsat-8), (2) standalone Sentinel-2, and (3) standalone Landsat-8. The results showed that the MLP model with multiple remotely sensed datasets outperformed other MLMs with 0.43 - 0.86 greater in R2 and 0.36 - 5.88 lower in RMSE. The MLP model demonstrated the highest performance across the range of Chl-a concentrations and predicted peaks above 20 mg/m3 relatively well compared to other models. This was likely due to the capacity of MLP to handle imbalanced datasets. The predictive map of the spatial distribution of Chl-a generated by MLP well captured the areas with high and low Chl-a concentrations. This study pointed out the impacts of imbalanced Chl-a concentration observations (dominated by low Chl-a concentrations) on the performance of MLMs. The data imbalance likely led to MLMs poorly trained for high Chl-a values, producing low prediction accuracy. In conclusion, this study demonstrated the value of multiple remotely sensed datasets in enhancing the accuracy and reliability of Chl-a estimation, mainly when using the MLP model. These findings would provide valuable insights into utilizing MLMs effectively for Chl-a monitoring.