Random Forest Regression Research Articles

A machine learning and DFT assisted analysis of benzodithiophene based organic dyes for possible photovoltaic applications

We present a synergistic approach to combine Machine Learning (ML), Density Functional Theory (DFT), and molecular descriptor analysis for designing high-performance benzodithiophene (BDT) based chromophores. A dataset of 366 BDT incorporated moieties is compiled from literature while their molecular descriptors are designed by using Python programming language. Linear and Random Forest Regression models produces best results to predict their exciton binding energy (Eb) with their R-Squared (R2) value 0.87 and 0.94 respectively. Their DFT calculations provides additional features, including molecular charges. Their ML models also reveals that their Eb values are a crucial predictor for their photovoltaic (PV) performance as its lower value could facilitate efficient charge carrier separation. For this, their hydrogen bond acceptors (HBA) and topological polar surface area (TPSA) emerges as key descriptors during their regression analysis. Their DFT validation shows negligible differences in their molecular charges to suggest their electron donor/acceptor moieties can significantly impact their chromophore nature. The current research work is helpful for efficiently screening the suitability of organic chromophores for their PV applications through advanced computational tools.

Distinctive impacts of ESG pillars on corporate financial performance: A random forest analysis of Korean listed firms

This study investigates the distinct effects of the environmental, social, and governance (ESG) pillars on corporate financial performance (CFP) measured by ROE and ROA. Using a random forest regression on a sample of Korean-listed firms, we address the positive correlations among these pillars and explore the nonlinear relationships between CFP and individual pillars without pre-specified functional forms. The findings indicate that both environmental and governance performances positively affect CFP, with environmental performance having a more substantial impact but diminishing returns at higher levels. Social performance negatively affects CFP. The environmental pillar emerges as the most reliable predictor of CFP.

The effectiveness and factors influencing frequency of endoscopic bougie dilatation in treating postoperative anastomotic stenosis of congenital esophageal atresia.

This study is to evaluate the effectiveness and frequency factors of endoscopic bougie dilatation in treating postoperative anastomotic stenosis of congenital esophageal atresia (CEA). The clinical data of patients of anastomotic stenosis with endoscopic bougie were retrospectively analyzed. According to the number of dilation times (ND), patients were divided into two groups (Group 0: ND < 3; Grooup1: ND ≥ 3), and the differences in multiple clinical data were compared. Lasso regression and Ridge regression were used to screen important variables. Classification models were built utilizing various machine learning algorithms and their performance were evaluated. Finally, Kaplan-Meier model was used to estimate the probable-time distribution of children achieving normal feeding. Seventy-five patients underwent a total of 210 times of dilation, with a median of 3 times of dilation. The overall effectiveness was 98.67% (74/75), with perforation in 2 case (0.95%), and obvious bleeding in 3 cases (1.43%). Initial diameter of bougie, final diameter of bougie, treatment pattern (Regular: dilation each 4weeks; Wait-and-see: dilation until symptoms present), age at final dilation, esophageal obstruction by food were the factors related to ND. Random Forest (RF) and Logistic regression (LR) model were excellent models for predicting ND. The median age for achieving normal eating in Group0 was 120 days (95% CI: 90-160), while it was 270 days (95% CI: 240-460) in Group1 with a statistically significant difference (P < 0.0001). Endoscopic bougie dilatation is a safe and effective treatment for anastomotic stenosis. Selecting the appropriate bougie, using symptoms as the criterion for dilation, and minimizing the dilations under 3 times constitute a rational strategy.

Circular rubber aggregate CFST stub columns under axial compression: prediction and reliability analysis

Extensive studies support using steel tubes to enhance the structural integrity of rubber aggregate concrete (RBAC), namely RBAC-filled steel tubes (RCFST). However, current design codes for assessing the axial compressive behaviour of circular stub RCFST (CS-RCFST) columns are limited. Furthermore, there is a scarcity of studies focused on ensuring the structural safety of these columns. Based on an extensive experimental database comprising 145 columns, this study explores machine learning (ML) capabilities for predicting the axial strength of CS-RCFST columns, using six typical machine-learning models, i.e., symbolic regression (SR), XGBoost, CatBoost, random forest, LightGBM, and Gaussian process regression models. The hyperparameter tuning of the introduced ML models is performed using the Bayesian Optimization technique. The comparison results show that the CatBoost model is the most reliable and accurate ML model (R2 = 0.999 and 0.993 for the training and testing sets, respectively). In addition, a simple and practical design expression for CS-RCFST columns has been developed with acceptable accuracy based on the SR model (an average test-to-prediction ratio of 0.99 and CoV of 0.132). Meanwhile, the axial strength predicted by ML models was compared with two prominent practice codes (i.e., AISC360 and EC4). The comparison results indicated that the ML models could introduce a highly reliable and accurate approach over current design standards for strength prediction. Furthermore, a reliability analysis is conducted on two different ML models to evaluate the reliability of utilising ML models in practical design applications. This assessment involves identifying the statistical properties associated with the compressive strength of RBAC, as well as introducing the required resistance design factors aligned with the target reliability recommended by code standards.

Intraoperative short-term blood pressure variability and postoperative acute kidney injury: a single-center retrospective cohort study using sample entropy analysis

BackgroundTo investigate if intraoperative very short-term variability in blood pressure measured by sample entropy improves discrimination of postoperative acute kidney injury after noncardiac surgery.MethodsAdult surgical patients undergoing general, thoracic, urological, or gynecological surgery between August 2016 to June 2017 at Seoul National University Hospital were included. The primary outcome was acute kidney injury stage 1, defined by the Kidney Disease: Improving Global Outcomes guidelines. Exploratory and explanatory variables included sample entropy of the mean arterial pressure and standard demographic, surgical, anesthesia and hypotension over time indices known to be associated with acute kidney injury respectively. Random forest classification and L1 logistic regression were used to assess four models for discriminating acute kidney injury: (1) Standard risk factors which included demographic, anesthetic, and surgical variables (2) Standard risk factors and cumulative hypotension over time (3) Standard risk factors and sample entropy (4) Standard risk factors, cumulative hypotension over time and sample entropy.ResultsTwo hundred and thirteen (7.4%) cases developed postoperative acute kidney injury. The median and interquartile range for sample entropy of mean arterial pressure was 0.34 and [0.26, 0.42] respectively. C-statistics were identical between the random forest and L1 logistic regression models. Results demonstrated no improvement in discrimination of postoperative acute kidney injury with the addition of the sample entropy of mean arterial pressure: Standard risk factors: 0.81 [0.76, 0.85], Standard risk factors and hypotension over time indices: 0.80 [0.75, 0.85], Standard risk factors and sample entropy of mean arterial pressure: 0.81 [0.76, 0.85] and Standard risk factors, sample entropy of mean arterial pressure and hypotension over time indices: 0.81 [0.76, 0.86].ConclusionAssessment of very short-term blood pressure variability does not improve the discrimination of postoperative acute kidney injury in patients undergoing non-cardiac surgery in this sample.

Students' perceptions of skills needed for labor in digital economy

This study examines the skills required for individuals to thrive in the digital economy and emphasizes the significance of education and training in developing these skills. The study focuses on four key skills: information technology (IT) skills, creative thinking skills, teamwork skills, and problem-solving skills. This study explores the definition and impact of the digital economy on workers, economic development, job creation, and sustainability. The research employed multiple correspondence analysis and random forest regression to analyze the data and evaluate the research model. The results demonstrated the reliability and meaningfulness of the scales used in the study, with evaluation metrics indicating a high level of accuracy in the research model. The article concludes by emphasizing the importance of equipping individuals with the necessary skills to succeed in the digital economy, and suggests future research directions.

Transferring spectroscopic stellar labels to 217 million Gaia DR3 XP stars with SHBoost

With Gaia Data Release 3 (DR3), new and improved astrometric, photometric, and spectroscopic measurements for 1.8 billion stars have become available. Alongside this wealth of new data, however, there are challenges in finding efficient and accurate computational methods for their analysis. In this paper, we explore the feasibility of using machine learning regression as a method of extracting basic stellar parameters and line-of-sight extinctions from spectro-photometric data. To this end, we built a stable gradient-boosted random-forest regressor (xgboost), trained on spectroscopic data, capable of producing output parameters with reliable uncertainties from Gaia DR3 data (most notably the low-resolution XP spectra), without ground-based spectroscopic observations. Using Shapley additive explanations, we interpret how the predictions for each star are influenced by each data feature. For the training and testing of the network, we used high-quality parameters obtained from the StarHorse code for a sample of around eight million stars observed by major spectroscopic stellar surveys, complemented by curated samples of hot stars, very metal-poor stars, white dwarfs, and hot sub-dwarfs. The training data cover the whole sky, all Galactic components, and almost the full magnitude range of the Gaia DR3 XP sample of more than 217 million objects that also have reported parallaxes. We have achieved median uncertainties of 0.20 mag in V-band extinction, 0.01 dex in logarithmic effective temperature, 0.20 dex in surface gravity, 0.18 dex in metallicity, and 12% in mass (over the full Gaia DR3 XP sample, with considerable variations in precision as a function of magnitude and stellar type). We succeeded in predicting competitive results based on Gaia DR3 XP spectra compared to classical isochrone or spectral-energy distribution fitting methods we employed in earlier works, especially for parameters AV and Teff, along with the metallicity values. Finally, we showcase some potential applications of this new catalogue, including extinction maps, metallicity trends in the Milky Way, and extended maps of young massive stars, metal-poor stars, and metal-rich stars.

Nondestructive freshness prediction of large yellow croaker (Pseudosciaena crocea) using computer vision and machine learning techniques based on pupil color.

Conventional methods for evaluating of fish freshness based on physiological and biochemical methods are often destructive, complicated, and costly. This study aimed to predict the freshness of large yellow croaker which was sampled every second day in 9 consecutive days at 4°C, using computer vision technology combined with pupil color parameters and different machine learning algorithms (back propagation neural network, BPNN; radial basis function neural network; support vector regression; and random forest regression, RFR). In the process of model building, the RFR model provided the most accurate prediction for the value of total volatile basic nitrogen (TVB-N), with the R-square of the test set ( ) of 0.993. The BPNN model exhibited the best fit for predicting the value of thiobarbituric acid (TBA), with of 0.959. Additionally, the RFR model was the most effective in forecasting total viable count (TVC), with of 0.935. After validation, the root mean square error values of the RFR model for predicting TVB-N value, TBA value, and TVC value were the lowest, which were 0.764, 0.067, and 0.219, respectively. It demonstrated the applicability and good predictive performance of the RFR model for predicting biochemical and microbiological indicators. These findings also demonstrated that monitoring the changes in pupil color could successfully predict the freshness of chilled fish. PRACTICAL APPLICATION: Application Scenario: Quality inspectors detect changes in thefreshness of large yellow croaker in real time from the beginning of distribution to the selling site.

Effective Machine Learning Techniques for Dealing with Poor Credit Data

Credit risk is a crucial component of daily financial services operations; it measures the likelihood that a borrower will default on a loan, incurring an economic loss. By analysing historical data for assessment of the creditworthiness of a borrower, lenders can reduce credit risk. Data are vital at the core of the credit decision-making processes. Decision-making depends heavily on accurate, complete data, and failure to harness high-quality data would impact credit lenders when assessing the loan applicants’ risk profiles. In this paper, an empirical comparison of the robustness of seven machine learning algorithms to credit risk, namely support vector machines (SVMs), naïve base, decision trees (DT), random forest (RF), gradient boosting (GB), K-nearest neighbour (K-NN), and logistic regression (LR), is carried out using the Lending Club credit data from Kaggle. This task uses seven performance measures, including the F1 Score (recall, accuracy, and precision), ROC-AUC, and HL and MCC metrics. Then, the harnessing of generative adversarial networks (GANs) simulation to enhance the robustness of the single machine learning classifiers for predicting credit risk is proposed. The results show that when GANs imputation is incorporated, the decision tree is the best-performing classifier with an accuracy rate of 93.01%, followed by random forest (92.92%), gradient boosting (92.33%), support vector machine (90.83%), logistic regression (90.76%), and naïve Bayes (89.29%), respectively. The classifier is the worst-performing method with a k-NN (88.68%) accuracy rate. Subsequently, when GANs are optimised, the accuracy rate of the naïve Bayes classifier improves significantly to (90%) accuracy rate. Additionally, the average error rate for these classifiers is over 9%, which implies that the estimates are not far from the actual values. In summary, most individual classifiers are more robust to missing data when GANs are used as an imputation technique. The differences in performance of all seven machine learning algorithms are significant at the 95% level.

Comparative Study of XGBoost, Random Forest, and Logistic Regression Models for Predicting Customer Interest in Vehicle Insurance

In today’s competitive insurance market, accurately predicting customer interest in additional products, such as vehicle insurance, is crucial for optimizing marketing strategies and maximizing sales. This study presents a comparative analysis of three machine learning models such as XGBoost, RandomForest, and Logistic Regression to predict customer interest in vehicle insurance based on a dataset that includes demographic, vehicle, and policy-related features. The dataset was analyzed using five-fold cross-validation, and the performance of the models was evaluated using AUC-ROC, precision, recall, and F1-score. XGBoost demonstrated the highest recall (0.9525) and AUC-ROC (0.7854), making it the most effective model for identifying customers interested in vehicle insurance, though at the expense of lower precision (0.2585). RandomForest showed a more balanced trade-off between precision (0.3064) and recall (0.5341) but performed lower overall. Logistic Regression, while the most interpretable model, exhibited high variability in performance across different folds, with a lower average precision (0.2372). The findings of this research highlight that XGBoost is ideal for maximizing recall in high-volume campaigns, while RandomForest may be better suited for applications requiring fewer false positives. These results offer valuable insights into model selection based on business objectives and resource allocation.

Analysis of the power during machining of Inconel 718 for different geometrical profiles and the development of power prediction models using multisensor data

This article presents analyses and prediction of machining power during end milling of Inconel 718 for different geometric features namely angular slot, half slot, square pocket, rectangular pocket, and blind hole. The experimental study revealed that among the machining features examined, angular slot required the highest specific cutting energy, while blind holes consumed the least, across different machining conditions. A 40% increase in cutting speed raised specific cutting energy by 4–9%, while a similar increase in feed rate lowered specific cutting energy by 40–44% across all profiles. Doubling depth of cut reduced specific cutting energy by 50% for angular slots, half slots, square pockets, and rectangular pockets, but increased it for blind holes. Power prediction models were developed using multisensor data, employing two machine learning and two deep learning architectures. The performance of the four power prediction models was compared for all geometric profiles using both raw data and statistical feature data extracted from machining signals from force dynamometer, accelerometer, and acoustic emission (AE) sensors. Long–Short-Term Memory model outperformed other models in case of raw data for all geometric profiles. The Random Forest Regression method consistently fared better for power predictions models developed based on feature-extracted data. Predictive models trained without AE sensor data performed better than those trained with AE.

Integrating multi-source remote sensing and machine learning for root-zone soil moisture and yield prediction of winter oilseed rape (Brassica napus L.): A new perspective from the temperature-vegetation index feature space

Accurately assessing root-zone soil moisture is crucial for precision irrigation, as it directly influences crop yield. The Temperature-Vegetation Index (Ts-VI) Feature Space, which combines land surface temperature (Ts) and vegetation index (VI), is widely used to evaluate root-zone soil moisture in vegetated areas. However, its effectiveness in estimating crop yield remains unclear. Therefore, the objectives of this study are: (1) to collect multispectral and thermal infrared remote sensing data from a two-year (2021–2023) field experiment on winter oilseed rape (Brassica napus L.), and to optimize and evaluate the fitting methods of the dry and wet edges of the Ts-VI feature space based on the selected vegetation indices; (2) to analyze the spatiotemporal patterns of the Temperature Vegetation Dryness Index (TVDI) derived from the optimized Ts-VI feature space and estimate root-zone soil moisture (SM) and crop yield; and (3) to precisely invert the SM and yield of winter oilseed rape in the 0–60 cm root-zone using three machine learning algorithms—Support Vector Regression (SVR), Extreme Gradient Boosting Regression (XGBR), and Random Forest Regression (RFR)—based on the optimized TVDI. Results indicate that, among the various fitting methods, the polynomial fitting method shows the best performance. The performance of the root-zone soil moisture prediction models across different growth stages follows the order of budding stage > seedling stage > flowering stage, and with the increase of soil depth, the performance of the model gradually deteriorates.In the yield inversion of winter oilseed rape, TVDI effectively predicts yield, with the coefficient of determination (R2) ranging from 0.430 to 0.480 and RMSE ranging from 213.399 to 267.212 kg ha−1 during the seedling stage, R2 ranging from 0.640 to 0.747 and RMSE ranging from 110.712 to 178.133 kg ha−1 during the budding stage, and R2 ranging from 0.680 to 0.773 and RMSE ranging from 83.815 to 147.301 kg ha−1 during the flowering stage. The flowering stage effectively reflects crop yield trends and allows for accurate yield prediction of winter oilseed rape up to two months in advance. A comparison of the modeling results from XGBR, SVR, and RFR shows that XGBR provides the best fit for both root-zone soil moisture and yield predictions. Compared to linear regression models, the three machine learning models significantly improve accuracy and fit, providing more precise evaluations of root-zone soil moisture and yield. In addition, through the comparison and verification of this method in other regions, it shows that the results also have certain reference value. The combination of the Ts-VI feature space and machine learning algorithms not only enables precise monitoring of root-zone soil moisture conditions but also predicts future crop yield trends, offering valuable insights for water resource management and irrigation decision-making in precision agriculture.

Optimisation of artefact detection in photoplethysmography heart rate data: influence of different classifiers in machine learning models

Abstract Background Heart rate (HR) tracking by wrist-worn devices using photoplethysmography (PPG) could assist in continuously following up physical activity. However, the accuracy can be impacted by (motion) artefacts. Machine learning models could help to recognise artefacts in PPG-based HR data. The choice of classifier in these machine learning models is a determing factor for task performance of the model. Purpose This study evaluates and determines the optimal classifier for a new machine learning-based approach to enhance the reliability of artefact detection in PPG-based HR data. Methods A total of 62 participants (27 cardiac rehabilitation patients, 35 healthy athletes) wore both a test device and a reference device measuring HR continuously for 24 hours. A training dataset was prepared, assigning two independent labels (i.e. anomaly and activity) to each HR episode based on the reference device data. Fitbit data were processed using our in-house designed artefact removal procedure, which involves the application of two classification models: one for anomaly detection and another for activity detection. Four distinct classifiers were employed for both models: Balanced Bagging, Balanced Bagging with Random Forest, Balanced Random Forest, and Logistic Regression. Each classifier was evaluated using area under the receiver operating characteristic curve (ROC-AUC), accuracy, sensitivity and specificity. Results Of the 1,647,328 HR data points collected, 103,095 (6.26%) were identified as artefacts. Figure 1 and Figure 2 summarise the performance of the distinct classifiers for the anomaly model and the activity model, respectively. Balanced Bagging and Balanced Bagging with Random Forest consistently demonstrate the highest AUC values and accuracies across both anomaly and activity detection models (anomaly detection: AUC = 0.95, accuracy = 89-85%; activity detection: AUC = 0.98, accuracy = 95%). Comparing these two, Balanced Bagging with Random Forest emerges as the preferred option, given the highest sensitivity in both anomaly detection (93%&amp;gt;86%) and activity detection models (99%&amp;gt;96%). In contrast, Balanced Random Forest and Logistic Regression exhibit inferior performance. In the anomaly detection model, Balanced Random Forest exhibits a lower sensitivity of 75%, while Logistic Regression performs even worse with a sensitivity of 25%. Similarly, in the activity detection model, both Balanced Random Forest and Logistic Regression demonstrate diminished performance. Conclusions Balanced Bagging with Random Forest emerges as the optimal classifier to detect anomalies and activities in continuous PPG-based HR data, thus contributing to the optimisation of our in-house designed procedure for removing artefacts. This processing aims to provide a reliable and automatic way for continuous HR monitoring, which can help monitor and guide physical activities.

Artificial intelligence for automated left ventricular systolic dysfunction detection using a wearable cardiac patch incorporated with synchronized phonocardiogram and electrocardiogram

Abstract Background Left ventricular systolic dysfunction (LVSD) is characterized by a reduced left ventricular ejection fraction (LVEF) and is associated with three times the risk of developing Heart failure (HF). We aimed to test an artificial intelligence (AI) algorithm applied to synchronized phonocardiography (PCG) and electrocardiography (ECG) signals, recorded with a wearable device, to validate its potential as a screening tool for LVSD. Methods Using ECG and PCG data collected from 1960 admitted patients in a hospital, we trained an AI algorithm to detect the LVSD. There are two ways of defining LVSD, one group is defined by LVEF &amp;lt;50% and the other is defined by LVEF 40% as measured by echocardiography. The AI algorithm followed a two-step detection structure. In the first phase, ECG signals were processed by wavelet denoising and baseline wander correction, and PCG signals were converted to the frequency domain via wavelet transformation. We developed a 1D CNN-based U-Net variant model to pinpoint electromechanical activation time (EMAT). Addressing the challenge of processing multimodal signals, we devised a contrastive learning approach that mandates the encoders for different modalities to generate embeddings within the same feature space. This trained encoder, along with the decoder in U-Net, is subsequently fine-tuned for EMAT detection. The AI-EMAT% was obtained by adjusting detected EMAT based on RR interval. For the second phase, random forest regression is used to estimate LVEF, incorporating features such as AI-EMAT% and clinical data, including age and gender. For LVSD detection, we evaluated the performance of methods by bifurcating AI-estimated LVEF% (AI-predict LVEF%) based on thresholds of 40% and 50%, and by directly estimating AI-EMAT% cutoff. The evaluation of LVSD detection algorithm, conducted via 2-fold cross-validation, involves dividing the dataset into two subsets, with each alternately serving as training and testing sets to ensure a fair assessment. Results We recruited 1960 patients (1430 [72.96%] male). 357 (18.21%) had an ejection fraction of 50% or lower, and 157 (8.01%) had an ejection fraction of 40% or lower. In detecting LVSD of LVEF&amp;lt;50%, AI-EMAT% yielded an AUROC of 0.86, sensitivity of 73.7%, and specificity of 80.7%. While AI-predict LVEF% approach resulted in an AUROC of 0.89, sensitivity of 81.8%, and specificity of 81.5%. In detecting LVSD of LVEF&amp;lt;40%, AI-EMAT% yielded an AUROC of 0.89, sensitivity of 85.4%, and specificity of 76.0%. While AI-predict LVEF% outputs resulted in an AUROC of 0.91, sensitivity of 92.4%, and specificity of 76.8%. Conclusions We developed an automated algorithm to detect LVSD using a wearable device that incorporated synchronized PCG and ECG signals, demonstrating robust performance across different subgroups. This approach represents a strategy for automated screening of LV systolic dysfunction, particularly beneficial in resource-limited settings.Baseline and AUROCAI algorithm

Research on Impact Prediction Model for Corn Ears by Integrating Motion Features Using Machine Learning Algorithms

The mechanical damage of corn kernels during harvest leads to mildew in the kernel storage process, seriously affecting food safety and quality. Impact force is the primary source of mechanical damage in the corn threshing process, and its accurate detection is of great significance for corn threshing with low damage. A method for the impact force detection of corn ears was proposed in this manuscript. The momentum theorem determined the main factors influencing impact force (weight, falling height, and space attitude). Corn ear weight, falling height, and space attitude were used as experimental factors. The bench test was carried out with the impact force of corn ear as the output variable. During the experiment, piezoelectric sensors were used to collect the impact force of corn ears under different motion states. Then, the impact force detection model was constructed using four machine learning algorithms: multiple linear regression, ridge regression, random forest, and support vector regression. The results showed that the RF algorithm was more suitable for constructing the prediction model of average and maximum impact force when corn ears fall, SD, RMSE, and r were, respectively: 0.9526, 1.2685, 0.9855; 3.8389, 3.6071, and 0.8510. Secondly, the weight characteristics had the most significant influence on the impact force detection of the ear. Therefore, this method can be used as an accurate, objective, and efficient online detection method for impact force.

Data-driven prediction of early language delay at Youth Health Care Services in the Netherlands

Abstract Introduction Language development in early childhood significantly impacts future academic success, social interaction, and emotional well-being. This study aimed to develop and evaluate a prediction model using data on developmental milestones and parents’ socio-economic background known at age 2, and translate the results into an interactive digital dashboard and pilot it with Youth Health Care (YHC) professionals and parents. Methods A mixed-method study on retrospective cohort of children attending YHC at age 2, followed up until age 4. Data on language development, gender, parental education, and home language environment were obtained from YHC digital dossiers and linked to parental socio-economic status characteristics, and perinatal outcomes and healthcare costs using a secure remote access environment of Statistics Netherlands. Random forest and logistic regression models were computed with language development at age 4 as an outcome, initially using all available data at age 2 as predictors and then with a restricted model using variables readily available in YHC settings. Models performance was assessed using sensitivity, specificity, and AUC value. Results were visualized through an interactive dashboard and pilot-tested in simulated consultations. Results Among 9,148 children, 13.2% had language development delay at age 4. The full model had an AUC of 0.78, while the restricted model (gender, parental education, language environment, and language development), achieved an AUC of 0.77. YHC professionals and parents recognized the value of individualized, data-driven risk assessment for potential language delay to stimulate the discussion of preventive interventions. Conclusions A good quality prediction model for language development at age 4 can be derived from just a few background characteristics of the child and parents at age 2. The digital dashboard presents a practical approach of integrating prediction model results into daily YHC practice. Key messages • Language development at age of 4 can be predicted as early as age of 2 using just a few background characteristics of child and parents, which presents a window of opportunity for prevention. • The digital dashboard presents a practical approach of integrating prediction model results into preventive care practice.

Using Machine Learning to identify the social determinants of the mid-life decline in mental health

Abstract Background The mid-life decline in mental well-being, depicted as a U-shaped pattern, is debated across disciplines. This study seeks to identify the social determinants of this decline and the overall contribution of social determinants to changes in mid-life mental health. Methods We use data from the 1970 British Cohort Study, from the ages 26, 34, and 42 (N = 6992, 51.5% female). Mental distress was measured using the 9-score Malaise Inventory. Mental distress increased from age 34 (M = 1.67) to 42 (M = 1.86; p &amp;lt; 0.001), with 17.8% declining by at least 1 SD on the inventory. We contrasted this group of decliners with those exhibiting stable mental health. We included socioeconomic and family factors, as well as physical health and health behaviours as potential determinants of mental health decline (at ages 26, 34, or at birth). We combine Random Forest and logistic regressions. Results The Random Forest’s Variable Importance (VI; root Mean Squared Error (MSE)), revealed social class at birth (VI = 0.023), physical multimorbidity at age 26 (VI = 0.22), and income quintile at age 34 (VI = 0.022) as most important predictors. Regressions confirmed that low birth social class (OR = 1.51 [1.25-1.83]), high-income at age 34 (OR = 0.68 [0.53-0.88]), and physical multimorbidity at age 26 (OR = 1.25 [1.06-1.48]) predict mental health decline. For those with high income and social class, the predicted probability of mid-life mental health decline is 0.109 (CI = [0.08-0.139]), rising to 0.277 (CI = [0.227-0.324]) for high risk individuals. Social class does not predict changes in mental health between ages 26 - 34. Interpretation We show that various social determinants of health contribute to the observed decline in mental health during mid-life. Social class at birth, income, and physical health are the most relevant predictors; they reduce the risk of decline from up to 27% to 11%. The contribution of social class appears specific to mid-life, and it could help explain the U-shaped pattern. Key messages • Social Factors Matter: Social class, income, and health predict mid-life mental decline. • Mid-life Class Impact: Social class uniquely affects mental health decline in mid-life.

Predictive slope stability early warning model based on CatBoost

A model for predicting slope stability is developed using Categorical Boosting (CatBoost), which incorporates 6 slope features to characterize the state of slope stability. The model is trained using a symmetric tree as the base model, utilizing ordered boosting to replace gradient estimation, which enhances prediction accuracy. Comparative models including Support Vector Machine (SVM), Light Gradient Boosting Machine (LGBM), Random Forest (RF), and Logistic Regression (LR) were introduced. Five performance evaluation metrics are utilized to assess the predictive capabilities of the CatBoost model. Based on CatBoost model, the predicted probability of slope instability is calculated, and the early warning model of slope instability is further established. The results suggest that the CatBoost model demonstrates a 6.25% disparity in accuracy between the training and testing sets, achieving a precision of 100% and an Area Under Curve (AUC) value of 0.95. This indicates a high level of predictive accuracy and robust ordering capabilities, effectively mitigating the problem of overfitting. The slope instability warning model offers reasonable classifications for warning levels, providing valuable insights for both research and practical applications in the prediction of slope stability and instability warning.

Automatic categorization of medical documents in Afaan Oromo using ensemble machine learning techniques

Automatic medical document classification using machine learning techniques can enhance the productivity of healthcare services by reducing processing time and cost. This work proposes an ensemble learning approach to develop a model that classifies electronic medical documents in Afaan Oromo. The main tasks in this work are preparing the corpus, pre-processing, training the models, and the classification process. We used the term frequency-inverse document frequency (TF-IDF) and bag of words (BOW) feature extraction methods. An ensemble technique in this work is that it creates multiple individual classifier predictions from naïve Bayes, random forest, SVM, and logistic regression and then combines them to advance a reliable and more accurate classifier. Evaluation measures were employed using accuracy, F1-score, recall, and precision for performance comparison. The efficiency of the proposed method is compared with the two existing boosting approaches, namely gradient boosting and adaboost. The experimental result shows the efficiency of BOW feature extraction over TF-IDF in this work on our dataset. These results also illustrated the effectiveness of the proposed model by scoring 94.81% accuracy and 94.84% F1-score. This work significantly contributes to the technological enhancement of service delivery, managing documents through classification methods, and advancing the data processing systems in healthcare sectors.

Space-Based Mapping of Pre- and Post-Hurricane Mangrove Canopy Heights Using Machine Learning with Multi-Sensor Observations

Coastal mangrove forests provide numerous ecosystem services, which can be disrupted by natural disturbances, mainly hurricanes. Canopy height (CH) is a key parameter for estimating carbon storage. Airborne Light Detection and Ranging (LiDAR) is widely viewed as the most accurate method for estimating CH but data are often limited in spatial coverage and are not readily available for rapid impact assessment after hurricane events. Hence, we evaluated the use of systematically acquired space-based Synthetic Aperture Radar (SAR) and optical observations with airborne LiDAR to predict CH across expansive mangrove areas in South Florida that were severely impacted by Category 3 Hurricane Irma in 2017. We used pre- and post-Irma LiDAR-derived canopy height models (CHMs) to train Random Forest regression models that used features of Sentinel-1 SAR time series, Landsat-8 optical, and classified mangrove maps. We evaluated (1) spatial transfer learning to predict regional CH for both time periods and (2) temporal transfer learning coupled with species-specific error correction models to predict post-Irma CH using models trained by pre-Irma data. Model performance of SAR and optical data differed with time period and across height classes. For spatial transfer, SAR data models achieved higher accuracy than optical models for post-Irma, while the opposite was the case for the pre-Irma period. For temporal transfer, SAR models were more accurate for tall trees (&gt;10 m) but optical models were more accurate for short trees. By fusing data of both sensors, spatial and temporal transfer learning achieved the root mean square errors (RMSEs) of 1.9 m and 1.7 m, respectively, for absolute CH. Predicted CH losses were comparable with LiDAR-derived reference values across height and species classes. Spatial and temporal transfer learning techniques applied to readily available spaceborne satellite data can enable conservation managers to assess the impacts of disturbances on regional coastal ecosystems efficiently and within a practical timeframe after a disturbance event.