Conventional Random Forest Research Articles

GEOGRAPHICALLY WEIGHTED MACHINE LEARNING MODEL FOR ADDRESSING SPATIAL HETEROGENEITY OF PUBLIC HEALTH DEVELOPMENT INDEX IN JAVA ISLAND

Random Forest (RF) machine learning models have emerged as a prominent algorithm, addressing problems arising from the sole use of decision trees, such as overfitting and instability. However, conventional RF has global coverage that may need to capture spatial variations better. Based on the analysis of the level of public health development, the relationship between the level of health development and risk factors can vary spatially. We use a modified RF algorithm called Geographically Weighted Random Forest (GW-RF) to address this challenge. GW-RF, as a tree-based non-parametric machine learning model, can help explore and visualize relationships between the Public Health Development Index (PHDI) as response variables and factors that are indicators at the district level. GW-RF output is compared with global output, which is RF in 2018 using the percentage of the population with access to clean/decent water (X1), consumption of eggs and milk per capita per week (X2), number of healthcare facilities per 1000 people (X3), number of doctors per 1000 people (X4), pure participation rate ratio female/male (X5), percentage of households that have hand washing facilities with soap and water (X6) as independent variables. Our results show that the non-parametric GW-RF model shows high potential for explaining spatial heterogeneity and predicting PHDI versus a global model when including six major risk factors. However, some of these predictions mean little. Findings of spatial heterogeneity using GW-RF show the need to consider local factors in approaches to increasing PHDI values. Spatial analysis of PHDI provides valuable information for determining geographic targets for areas whose PHDI values need to be improved.

Comparative analysis of novel data‐driven techniques for remaining useful life estimation of wind turbine high‐speed shaft bearings

AbstractAs the global momentum for wind power generation accelerates, the industry faces substantial challenges due to premature failures in wind turbine components. These failures, particularly in critical elements like the high‐speed shaft bearing, lead to significant operational losses, including unplanned downtime and elevated maintenance costs. To mitigate these issues, it's crucial to have precise predictions of the remaining useful life (RUL) of these components, enabling timely interventions and more efficient maintenance schedules. This article proposes advanced, data‐driven approaches for estimating the RUL of wind turbine high‐speed shaft bearings, utilizing cutting‐edge techniques such as long short‐term memory (LSTM), bidirectional LSTM (BiLSTM), gated recurrent units (GRU), and random forest (RF) algorithms. Our analysis leverages vibration data from a 2 MW wind turbine equipped with a 20‐tooth pinion gear, providing a thorough validation and comparison of these methodologies against traditional models. Our results reveal that the LSTM and BiLSTM models excel in both accuracy and computational efficiency for predicting RUL and enhancing system prognosis, surpassing the performance of conventional RF and GRU methods. This research underscores the potential of our innovative data‐driven strategies to develop effective RUL estimation algorithms, significantly advancing wind turbine proactive operation and maintenance operations.

Knowledge-slanted random forest method for high-dimensional data and small sample size with a feature selection application for gene expression data

The use of prior knowledge in the machine learning framework has been considered a potential tool to handle the curse of dimensionality in genetic and genomics data. Although random forest (RF) represents a flexible non-parametric approach with several advantages, it can provide poor accuracy in high-dimensional settings, mainly in scenarios with small sample sizes. We propose a knowledge-slanted RF that integrates biological networks as prior knowledge into the model to improve its performance and explainability, exemplifying its use for selecting and identifying relevant genes. knowledge-slanted RF is a combination of two stages. First, prior knowledge represented by graphs is translated by running a random walk with restart algorithm to determine the relevance of each gene based on its connection and localization on a protein-protein interaction network. Then, each relevance is used to modify the selection probability to draw a gene as a candidate split-feature in the conventional RF. Experiments in simulated datasets with very small sample sizes (n≤30)\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$(n \\le 30)$$\\end{document} comparing knowledge-slanted RF against conventional RF and logistic lasso regression, suggest an improved precision in outcome prediction compared to the other methods. The knowledge-slanted RF was completed with the introduction of a modified version of the Boruta feature selection algorithm. Finally, knowledge-slanted RF identified more relevant biological genes, offering a higher level of explainability for users than conventional RF. These findings were corroborated in one real case to identify relevant genes to calcific aortic valve stenosis.

An interpretable ensemble trees method with joint analysis of static and dynamic features for myocardial infarction detection

Objective. In recent years, artificial intelligence-based electrocardiogram (ECG) methods have been massively applied to myocardial infarction (MI). However, the joint analysis of static and dynamic features to achieve accurate and interpretable MI detection has not been comprehensively addressed. Approach. This paper proposes a simplified ensemble tree method with a joint analysis of static and dynamic features to solve this issue for MI detection. Initially, the dynamic features are extracted by modeling the intrinsic dynamics of ECG via dynamic learning in addition to extracting classical static features. Secondly, a two-stage feature selection strategy is designed to identify a few significant features, which substitute the original variables that are employed in constructing the ensemble tree. This approach enhances the discriminative ability by selecting significant static and dynamic features. Subsequently, this paper presents an interpretable classification method named StackTree by introducing a stacked ensemble scheme to modify the ensemble tree simplification algorithm. The representative rules of the raw ensemble trees are selected as the intermediate training data that is used to retrain a decision tree with performance close to that of the source ensemble model. Using this scheme, the significant precision and interpretability of MI detection are thus comprehensively addressed. Main results. The effectiveness of our method in detecting MI is evaluated using the Physikalisch-Technische Bundesanstalt (PTB) and clinical database. The findings suggest that our algorithm outperforms the traditional methods based on a single type of feature. Additionally, it is comparable to the conventional random forest, achieving 97.1% accuracy under the inter-patient framework on the PTB database. Furthermore, feature subsets trained on PTB are validated using the clinical database, resulting in an accuracy of 84.5%. The chosen important features demonstrate that both static and dynamic information have crucial roles in MI detection. Crucially, the proposed method provides clear internal workings in an easy-to-understand visual manner.

Estimating Forest Aboveground Biomass Using a Combination of Geographical Random Forest and Empirical Bayesian Kriging Models

Accurately estimating forest aboveground biomass (AGB) is imperative for comprehending carbon cycling, calculating carbon budgets, and formulating sustainable forest management plans. Currently, random forest (RF) and other machine learning models are widely used to estimate forest AGB, as they can effectively handle nonlinear relationships. However, by constructing a global model using all the samples collected from a study area, these models fail to account for the spatial heterogeneity in the AGB and cannot correct the prediction biases, thereby constraining the estimation accuracy. To overcome these limitations, we proposed a novel approach termed geographical random forest and empirical Bayesian kriging (GRFEBK). This hybrid model combines the localized modeling capability of geographical random forest (GRF) with the bias correction strength of empirical Bayesian kriging (EBK). GRF adapts RF to account for the spatial heterogeneity of the AGB, while EBK utilizes the spatial autocorrelation of residuals to correct the prediction deviations. This study was conducted in Hainan Island, utilizing spectral bands, vegetation indices, tasseled cap components derived from Landsat-8 imagery, backscattering coefficients from ALOS-2 synthetic aperture radar, topographic features, and the forest canopy height as the explanatory variables. A total of 195 forest aboveground biomass (AGB) samples were collected for modeling and assessing the predictive accuracy. The results demonstrate that, among the tested models, including GRFEBK, RF, support vector machine (SVM), k-nearest neighbor (KNN), geographically weighted regression (GWR), GRF, and EBK, GRFEBK attains the highest R2 (0.78) and the lowest RMSE (36.04 Mg/ha) and RRMSE (22.87%), significantly outperforming the conventional models and using GRF or EBK alone. These results demonstrate that by accounting for local non-stationarity in AGB and correcting prediction biases, GRFEBK achieves significantly higher accuracy than conventional RF and other models. While the results are promising, the computational cost of GRFEBK and its performance under varying geographical conditions warrant further investigation at larger scales to assess its broader applicability. Nevertheless, GRFEBK provides an innovative and more reliable approach for accurate forest AGB estimation with great potential to support global forest resource monitoring.

Balancing Interpretability and Performance: Optimizing Random Forest Algorithm Based on Point-to-Point Federated Learning

Federated learning is extensively applied in collaborative data scenarios involving multiple data owners. While the majority of state-of-the-art federated learning algorithms are currently black-box models, making it challenging for users to comprehend how decisions are made. Random forest models are extensively utilized in medical contexts owing to their exceptional interpretability. However, when faced with multicenter data, the heterogeneity of data from each center often leads to its predictive performance falling short of expectations. To mitigate this challenge, the present study introduces DFLRF (Decentralized Federated Learning Random Forest), a federated learning algorithm based on random forests. Expanding on conventional random forests, DFLRF employs federated learning to disseminate decision tree models. It assesses and consolidates tree models from all client sites, thereby comprehensively addressing data disparities across various centers. The algorithm selects the optimal decision tree model based on the magnitude of model loss to guarantee the predictive performance of the final federated random forest model. The algorithm undergoes testing on a public dataset. Experimental results demonstrate that, compared to baseline algorithms, DFLRF enhances the AUC by 1.5% and the recall rate by 6%, while also ensuring superior interpretability.

Coupling of microbial-explicit model and machine learning improves the prediction and turnover process simulation of soil organic carbon

Modeling soil organic carbon (SOC) is helpful for understanding its distribution and turnover processes, which can guide the implementation of effective measures for carbon (C) sequestration and enhance land productivity. Process-based simulation with high interpretability and extrapolation, and machine learning modeling with high flexibility are two common methods for investigating SOC distribution and turnover. To take advantage of both methods, we developed a hybrid model by coupling of a two-carbon pool microbial model and machine learning for SOC modeling. Here, we assessed the SOC model's predictive, mapping, and interpretability capabilities for the SOC turnover process on Ningbo region. The results indicate that the microbial model with density-dependence (β ​= ​2) and microbial biomass carbon simulation performed better in modeling the parameters of the microbial-based C cycle, such as microbial carbon use efficiency (CUE), microbial mortality rate, and assimilation rate. By integrating this optimal microbial model and random forest (RF) model, the hybrid model improved the prediction accuracy of SOC, with an increased R2 from 0.74 to 0.84, residual prediction deviation increased from 1.97 to 2.50, and reduced the root-mean-square error from 4.65 to 3.67 ​g ​kg−1 compared to the conventional RF model. As a result, the predicted SOC distribution exhibited high spatial variation and provided abundant details. Microbial CUE and potential C input, represented by net primary productivity, emerged as the primary factors driving SOC distribution in Ningbo region. Projections of SOC under the CMIP6 SSP2-4.5 scenario revealed that regional C loss in high SOC areas was mainly caused by decreased microbial CUE and C input, induced by climate change. Our findings highlight the potential of combining the microbial-explicit model and machine learning to improve SOC prediction accuracy and understand SOC feedback in a changing climate.

Development of Novel Methods for QSAR Modeling by Machine Learning Repeatedly: A Case Study on Drug Distribution to Each Tissue.

Artificial intelligence is expected to help identify excellent candidates in drug discovery. However, we face a lack of data, as it is time-consuming and expensive to acquire raw data perfectly for many compounds. Hence, we tried to develop a novel quantitative structure-activity relationship (QSAR) method to predict a parameter more precisely from an incomplete data set via optimizing data handling by making use of predicted explanatory variables. As a case study we focused on the tissue-to-plasma partition coefficient (Kp), which is an important parameter for understanding drug distribution in tissues and building the physiologically based pharmacokinetic model and is a representative of small and sparse data sets. In this study, we predicted the Kp values of 119 compounds in nine tissues (adipose, brain, gut, heart, kidney, liver, lung, muscle, and skin), although some of these were not available. To fill the missing values in Kp for each tissue, first we predicted those Kp values by the nonmissing data set using a random forest (RF) model with in vitro parameters (log P, fu, Drug Class, and fi) like a classical prediction by a QSAR model. Next, to predict the tissue-specific Kp values in a test data set, we constructed a second RF model with not only in vitro parameters but also the Kp values of other tissues (i.e., other than target tissues) predicted by the first RF model as explanatory variables. Furthermore, we tested all possible combinations of explanatory variables and selected the model with the highest predictability from the test data set as the final model. The evaluation of Kp prediction accuracy based on the root-mean-square error and R2 value revealed that the proposed models outperformed other machine learning methods such as the conventional RF and message-passing neural networks. Significant improvements were observed in the Kp values of adipose tissue, brain, kidney, liver, and skin. These improvements indicated that the Kp information on other tissues can be used to predict the same for a specific tissue. Additionally, we found a novel relationship between each tissue by evaluating all combinations of explanatory variables. In conclusion, we developed a novel RF model to predict Kp values. We hope that this method will be applied to various problems in the field of experimental biology which often contains missing values in the near future.

SPATIALLY AWARE LANDSLIDE SUSCEPTIBILITY PREDICTION USING A GEOGRAPHICAL RANDOM FOREST APPROACH

Abstract. Landslide susceptibility prediction practices have been increasingly reliant on non-geographically-oriented (i.e., aspatial) machine learning algorithms. While these approaches have exhibited increasing success, they have often faced criticism for their limited consideration of spatial autocorrelations and local variations across geographical space, thereby neglecting the concept of spatial non-stationarity. To fulfill the research gap, this work applies a geographical random forest (GRF) approach, contrasting it with the conventional random forest (RF) algorithm. To this end, the study area, encompassing the Lake Sapanca Basin and its surroundings, was subdivided into 4,452 slope-based mapping units. The effectiveness of both predictive models was then measured by using overall accuracy (OA) and area under the curve (AUC). The results revealed that the GRF (OA = 80.82% and AUC = 85.22%) outperformed the RF algorithm (OA = 75.34% and AUC = 82.50%) by approximately 5% in OA, and demonstrated a 3% improvement in AUC score. The Wilcoxon signed-rank test confirmed significant differences (95% level) between the predictions of both models. The slope parameter emerged as the globally most influential factor, but local interpretations disclosed notable variations in the importance of causative factors contingent upon location. For instance, the curvature parameter was the most important geospatial covariate in around one-third (34.23%) of the slope units, mostly concentrated in the northernmost zones of the study area. On the other hand, elevation was the most important factor for 14.67% of the slope units primarily located in the southern region.

Toxic Fake News Detection and Classification for Combating COVID-19 Misinformation

The emergence of COVID-19 has led to a surge in fake news on social media, with toxic fake news having adverse effects on individuals, society, and governments. Detecting toxic fake news is crucial, but little prior research has been done in this area. This study aims to address this gap and identify toxic fake news to save time spent on examining nontoxic fake news. To achieve this, multiple datasets were collected from different online social networking platforms such as Facebook and Twitter. The latest samples were obtained by collecting data based on the topmost keywords extracted from the existing datasets. The instances were then labeled as toxic/nontoxic using toxicity analysis, and traditional machine-learning (ML) techniques such as linear support vector machine (SVM), conventional random forest (RF), and transformer-based ML techniques such as bidirectional encoder representations from transformers (BERT) were employed to design a toxic-fake news detection (FND) and classification system. As per the experiments, the linear SVM method outperformed BERT SVM, RF, and BERT RF with an accuracy of 92% and -score, -score, and -score of 95%, 85%, and 87%, respectively. Upon comparison, the proposed approach has either suppressed or achieved results very close to the state-of-the-art techniques in the literature by recording the best values on performance metrics such as accuracy, F1-score, precision, and recall for linear SVM. Overall, the proposed methods have shown promising results and urge further research to restrain toxic fake news. In contrast to prior research, the presented methodology leverages toxicity-oriented attributes and BERT-based sequence representations to discern toxic counterfeit news articles from nontoxic ones across social media platforms.

Predicting Systemic Risk in the Russian Financial Sector with Boosting Techniques

We test the predictive performance of different ensemble methods for forecasting systemic risk in Russia for the period 2008-2024. In contrast to the existing research on machine learning ensemble techniques, we find that conventional random forest works better for the Russian data. Based on this model, we additionally conduct variable importance analysis. We identify that the first three lags of the exchange rate as well as the level of non-performing loans are of utmost importance for systemic risk in the national financial sector.

Quantifying Digital Biomarkers for Well-Being: Stress, Anxiety, Positive and Negative Affect via Wearable Devices and Their Time-Based Predictions.

Wearable devices have become ubiquitous, collecting rich temporal data that offers valuable insights into human activities, health monitoring, and behavior analysis. Leveraging these data, researchers have developed innovative approaches to classify and predict time-based patterns and events in human life. Time-based techniques allow the capture of intricate temporal dependencies, which is the nature of the data coming from wearable devices. This paper focuses on predicting well-being factors, such as stress, anxiety, and positive and negative affect, on the Tesserae dataset collected from office workers. We examine the performance of different methodologies, including deep-learning architectures, LSTM, ensemble techniques, Random Forest (RF), and XGBoost, and compare their performances for time-based and non-time-based versions. In time-based versions, we investigate the effect of previous records of well-being factors on the upcoming ones. The overall results show that time-based LSTM performs the best among conventional (non-time-based) RF, XGBoost, and LSTM. The performance even increases when we consider a more extended previous period, in this case, 3 past-days rather than 1 past-day to predict the next day. Furthermore, we explore the corresponding biomarkers for each well-being factor using feature ranking. The obtained rankings are compatible with the psychological literature. In this work, we validated them based on device measurements rather than subjective survey responses.

Socioeconomic and environmental determinants of asthma prevalence: a cross-sectional study at the U.S. County level using geographically weighted random forests

BackgroundSome studies have established associations between the prevalence of new-onset asthma and asthma exacerbation and socioeconomic and environmental determinants. However, research remains limited concerning the shape of these associations, the importance of the risk factors, and how these factors vary geographically.ObjectiveWe aimed (1) to examine ecological associations between asthma prevalence and multiple socio-physical determinants in the United States; and (2) to assess geographic variations in their relative importance.MethodsOur study design is cross sectional based on county-level data for 2020 across the United States. We obtained self-reported asthma prevalence data of adults aged 18 years or older for each county. We applied conventional and geographically weighted random forest (GWRF) to investigate the associations between asthma prevalence and socioeconomic (e.g., poverty) and environmental determinants (e.g., air pollution and green space). To enhance the interpretability of the GWRF, we (1) assessed the shape of the associations through partial dependence plots, (2) ranked the determinants according to their global importance scores, and (3) mapped the local variable importance spatially.ResultsOf the 3059 counties, the average asthma prevalence was 9.9 (standard deviation ± 0.99). The GWRF outperformed the conventional random forest. We found an indication, for example, that temperature was inversely associated with asthma prevalence, while poverty showed positive associations. The partial dependence plots showed that these associations had a non-linear shape. Ranking the socio-physical environmental factors concerning their global importance showed that smoking prevalence and depression prevalence were most relevant, while green space and limited language were of minor relevance. The local variable importance measures showed striking geographical differences.ConclusionOur findings strengthen the evidence that socio-physical environments play a role in explaining asthma prevalence, but their relevance seems to vary geographically. The results are vital for implementing future asthma prevention programs that should be tailor-made for specific areas.

A Hybrid Model for Predicting Low Oxygen in the Return Air Corner of Shallow Coal Seams Using Random Forests and Genetic Algorithm

In order to better solve the phenomenon of low oxygen in the corner of return airway caused by abnormal gas emission in goaf during shallow coal seam mining, by analyzing the source and reason of low oxygen phenomenon, a prediction model of oxygen concentration in the corner of return airway based on genetic algorithm (GA) and random forest (RF) technology was proposed. The training sample set was established by using the field data obtained from actual monitoring, including the oxygen concentration in the return airway corner, the periodic pressure step distance of the roof, the surface temperature and atmospheric pressure. GA was used to optimize the parameters in the RF model, including trees and leaves in the forest. The results showed that the model prediction error was minimum when the number of trees was 398 and the number of leaves was 1. In addition, GA was used to optimize the number of hidden neurons and the initial weight threshold of the back-propagation neural network (BPNN). In order to verify the superiority of the model, the GA optimized RF and BPNN model are compared with the conventional RF and BPNN model. Analyze the average absolute percentage error (MAPE), root mean square error (RMSE), and average absolute error (MAE) of the prediction data of each model. The results show that the optimized RF prediction model is better than other models in terms of prediction accuracy.

Optimal prediction of viral host from genomic datasets using ensemble classifier

Viruses are common biological agents that are supposed to be the world's greatest repositories of undiscovered genetic diversity. One of the common problems in bioinformatics is gene-disease prediction. Techniques for taxonomic classification, host range, and biological properties of newly discovered viruses are needed for complete functional characterization and annotation. Understanding the behaviors as well as interactions of microbial populations needs research into virus-host infectious associations. The following three main steps of an unique viral host prediction method using genomic datasets are introduced in this research work: “(a) Pre-processing, (b) Feature extraction, and (c) Prediction phase”. In starting stage, raw genomic datasets are exposed to pre-processing, which would include data cleaning activities. The pre-processed data is then used to extract the statistical features, high order statistical features, weighted holoentropy, chi-squared features, relief-based features, and symmetric uncertainty-based features. The ensemble approach, which uses the Support Vector Machine (SVM), Neural Network (NN), Random Forest (RF), and Convolutional Neural Network (CNN), respectively, is then employed for the prediction. Here, SVM, NN, and RF classifiers are fed the retrieved features as input. These classifiers' outputs will be fed into an optimised CNN, which produces the final prediction outcome. Additionally, the Adaptive Flower Pollination Algorithm (AFPA), an upgraded variant of the conventional Flower Pollination Protocol, is used to fine-tune the weights of optimised CNN in order to increase prediction accuracy (FPA). The AFPA+EC's F1-Score is 0.75026, which is correspondingly 61%, 52.4%, 20.2%, and 26.5% better than the conventional methods SVM, KNN, RF, and CNN.

Application of machine learning algorithms in predicting HIV infection among men who have sex with men: Model development and validation.

BackgroundContinuously growing of HIV incidence among men who have sex with men (MSM), as well as the low rate of HIV testing of MSM in China, demonstrates a need for innovative strategies to improve the implementation of HIV prevention. The use of machine learning algorithms is an increasing tendency in disease diagnosis prediction. We aimed to develop and validate machine learning models in predicting HIV infection among MSM that can identify individuals at increased risk of HIV acquisition for transmission-reduction interventions.MethodsWe extracted data from MSM sentinel surveillance in Zhejiang province from 2018 to 2020. Univariate logistic regression was used to select significant variables in 2018–2019 data (P < 0.05). After data processing and feature selection, we divided the model development data into two groups by stratified random sampling: training data (70%) and testing data (30%). The Synthetic Minority Oversampling Technique (SMOTE) was applied to solve the problem of unbalanced data. The evaluation metrics of model performance were comprised of accuracy, precision, recall, F-measure, and the area under the receiver operating characteristic curve (AUC). Then, we explored three commonly-used machine learning algorithms to compare with logistic regression (LR), including decision tree (DT), support vector machines (SVM), and random forest (RF). Finally, the four models were validated prospectively with 2020 data from Zhejiang province.ResultsA total of 6,346 MSM were included in model development data, 372 of whom were diagnosed with HIV. In feature selection, 12 variables were selected as model predicting indicators. Compared with LR, the algorithms of DT, SVM, and RF improved the classification prediction performance in SMOTE-processed data, with the AUC of 0.778, 0.856, 0.887, and 0.942, respectively. RF was the best-performing algorithm (accuracy = 0.871, precision = 0.960, recall = 0.775, F-measure = 0.858, and AUC = 0.942). And the RF model still performed well on prospective validation (AUC = 0.846).ConclusionMachine learning models are substantially better than conventional LR model and RF should be considered in prediction tools of HIV infection in Chinese MSM. Further studies are needed to optimize and promote these algorithms and evaluate their impact on HIV prevention of MSM.

Joint Regression Network and Window Function-Based Piecewise Neural Network for Cuffless Continuous Blood Pressure Estimation Only Using Single Photoplethesmogram

The blood pressure (BP) is generally measured using a cuff based sphygmomanometer. However, it is inconvenient to be used. Recently, an alternative solution only using the photoplethesmograms (PPGs) was proposed. In this case, the continuous BP estimation could be performed. First, the features were extracted from the PPGs. Then, a regression network was employed to estimate the BP values. Nevertheless, the accuracy of this approach was not so high. In order to improve the estimation accuracy, this paper proposes to cascade a two layer piecewise neural network to the output of the existing regression network to correct the estimation error. In particular, the overall system is a three layer network. The first layer of the network is the existing regression network. It generates the initial estimated BP values. The second layer of the network consists of the window functions. It segments the range of the BP values into various regions for the further processing. The final layer of the network performs the estimation correction. The performance of our proposed network is evaluated via two practical datasets and three common regression networks including the three layer artificial neural network (ANN) based regression network, the random forest (RF) based regression network and the support vector regression (SVR) based network. For the first dataset, our proposed method with the RF model and the piecewise neural network achieves the systolic BP (SBP) estimation error and the diastolic BP (DBP) estimation error at <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$3.01{\pm }2.22$ </tex-math></inline-formula> mmHg with the correlation coefficient at 0.926 and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$4.43{\pm }3.37$ </tex-math></inline-formula> mmHg with the correlation coefficient at 0.935, respectively. On the other hand, the conventional RF model without the piecewise neural network achieves the SBP estimation error and the DBP estimation error at <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$5.34{\pm }4.08$ </tex-math></inline-formula> mmHg with the correlation coefficient at 0.740 and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$5.89{\pm }4.98$ </tex-math></inline-formula> mmHg with the correlation coefficient at 0.840, respectively. For the second dataset, our proposed method with the RF model and the piecewise neural network achieves the SBP estimation error and the DBP estimation error at <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$7.91{\pm }8.06$ </tex-math></inline-formula> mmHg with the correlation coefficient at 0.876 and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$3.47{\pm }5.59$ </tex-math></inline-formula> mmHg with the correlation coefficient at 0.859, respectively. On the other hand, the conventional RF model without the piecewise neural network achieves the SBP estimation error and the DBP estimation error at <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$9.77{\pm }9.01$ </tex-math></inline-formula> mmHg with the correlation coefficient at 0.805 and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$7.08{\pm }5.55$ </tex-math></inline-formula> mmHg with the correlation coefficient at 0.612, respectively. It can be seen that our proposed network yields the estimated BP values highly correlated to the reference BP values. Also, our proposed method yields the higher accuracies compared to the existing networks. This demonstrates the effectiveness of our proposed network.

Toward Nonintrusive Camera-Based Heart Rate Variability Estimation in the Car Under Naturalistic Condition

Driver status monitoring systems are a vital component of smart cars in the future, especially in the era when an increasing amount of time is spent in the vehicle. The heart rate (HR) is one of the most important physiological signals of driver status. To infer HR of drivers, the mainstream of the existing research focused on capturing subtle heartbeat-induced vibration of the torso or leveraged photoplethysmography (PPG) that detects cardiac cycle-related blood volume changes in the microvascular. However, existing approaches rely on dedicated sensors that are expensive and cumbersome to be integrated or are vulnerable to ambient noise. Moreover, their performance on the detection of HR does not guarantee a reliable computation of the HR variability (HRV) measure, which is a more applicable metric for inferring mental and physiological status. The accurate computation HRV measure is based on the precise measurement of the beat-to-beat interval, which can only be accomplished by medical-grade devices that attach electrodes to the body. Considering these existing challenges, we proposed a facial expression-based HRV estimation solution. The rationale is to establish a link between facial expression and heartbeat since both are controlled by the autonomic nervous system. To solve this problem, we developed a tree-based probabilistic fusion neural network approach, which significantly improved HRV estimation performance compared to conventional random forest or neural network methods and the measurements from smartwatches. The proposed solution relies only on commodity camera with a lightweighted algorithm, facilitating its ubiquitous deployment in current and future vehicles. Our experiments are based on 3400 km of driving data from nine drivers collected in a naturalistic field study.

A Clinical Decision Support System (CDSS) for Unbiased Prediction of Caesarean Section Based on Features Extraction and Optimized Classification

Nowadays, caesarean section (CS) is given preference over vaginal birth and this trend is rapidly rising around the globe, although CS has serious complications such as pregnancy scar, scar dehiscence, and morbidly adherent placenta. Thus, CS should only be performed when it is absolutely necessary for mother and fetus. To avoid unnecessary CS, researchers have developed different machine-learning- (ML-) based clinical decision support systems (CDSS) for CS prediction using electronic health record of the pregnant women. However, previously proposed methods suffer from the problems of poor accuracy and biasedness in ML. To overcome these problems, we have designed a novel CDSS where random oversampling example (ROSE) technique has been used to eliminate the problem of minority classes in the dataset. Furthermore, principal component analysis has been employed for feature extraction from the dataset while, for classification purpose, random forest (RF) model is deployed. We have fine-tuned the hyperparameter of RF using a grid search algorithm for optimal classification performance. Thus, the newly proposed system is named ROSE-PCA-RF and it is trained and tested using an online CS dataset available on the UCI repository. In the first experiment, conventional RF model is trained and tested on the dataset while in the second experiment, the proposed model is tested. The proposed ROSE-PCA-RF model improved the performance of traditional RF by 4.5% with reduced time complexity, while only using two extracted features through the PCA. Moreover, the proposed model has obtained 96.29% accuracy on training data while improving the accuracy of 97.12% on testing data.

Interaction forests: Identifying and exploiting interpretable quantitative and qualitative interaction effects

Although interaction effects can be exploited to improve predictions and allow for valuable insights into covariate interplay, they are given limited attention in analysis. Interaction forests are a variant of random forests for categorical, continuous, and survival outcomes that explicitly models quantitative and qualitative interaction effects in bivariable splits performed by the trees constituting the forests. The new effect importance measure (EIM) associated with interaction forests allows for ranking of covariate pairs with respect to their interaction effects' importance to prediction. Using EIM, separate importance value lists for univariable effects, quantitative interaction effects, and qualitative interaction effects are obtained. In the spirit of interpretable machine learning, the bivariable split types of interaction forests target easily interpretable and communicable interaction effects. To learn about the nature of the interplay between covariates identified as interacting it is convenient to visualise their estimated bivariable influence. Functions that perform this task are provided in the R package diversityForest, which implements interaction forests. In a large-scale empirical study using 220 data sets, interaction forests tended to deliver better predictions than conventional random forests and competing random forest variants that use multivariable splitting. In a simulation study, EIM delivered considerably better rankings for the relevant quantitative and qualitative interaction effects than competing approaches. These results indicate that interaction forests are suitable tools for the challenging task of identifying and making use of easily interpretable and communicable interaction effects in predictive modelling.