Regression Tree Method Research Articles

Prediction model for bleeding after endoscopic submucosal dissection of gastric neoplasms from a high-volume center.

Bleeding after endoscopic submucosal dissection (ESD) is a main adverse event. To date, although there have been several studies about risk factors for post-ESD bleeding, there has been few predictive model for post-ESD bleeding with large volume cases. We aimed to design a prediction model for post-ESD bleeding using a classification tree model. We analyzed a prospectively established cohort of patients with gastric neoplasms treated with ESD from 2007 to 2016. Baseline characteristics were collected for a total of 5080 patients, and the bleeding risk was estimated using variable statistical methods such as logistic regression, AdaBoost, and random forest. To investigate how bleeding was affected by independent predictors, the classification and regression tree (CART) method was used. The prediction tree developed for the cohort was internally validated. Post-ESD bleeding occurred in 262 of 5080 patients (5.1%). In multivariate logistic regression, ongoing antithrombotic use during the procedure, cancer pathology, and piecemeal resection were significant risk factors for post-ESD bleeding. In the CART model, the decisive variables were ongoing antithrombotic agent use, resected specimen size ≥49mm, and patient age <62years. The CART model accuracy was 94.9%, and the cross-validation accuracy was 94.8%. We developed a simple and easy-to-apply predictive tree model based on three risk factors that could help endoscopists identify patients at a high risk of bleeding. This model will enable clinicians to establish precise management strategies for patients at a high risk of bleeding and to prevent post-ESD bleeding.

Predicting the infiltration characteristics for semi-arid regions using regression trees

Abstract The study of the infiltration process is considered essential and necessary for all hydrology studies. Therefore, accurate predictions of infiltration characteristics are required to understand the behavior of the subsurface flow of water through the soil surface. The aim of the current study is to simulate and improve the prediction accuracy of the infiltration rate and cumulative infiltration of soil using regression tree methods. Experimental data recorded with a double ring infiltrometer for 17 different sites are used in this study. Three regression tree methods: random tree, random forest (RF) and M5 tree, are employed to model the infiltration characteristics using basic soil characteristics. The performance of the modelling approaches is compared in predicting the infiltration rate as well as cumulative infiltration, and the obtained results suggest that the performance of the RF model is better than the other applied models with coefficient of determination (R2) = 0.97 and 0.97, root mean square error (RMSE) = 8.10 and 6.96 and mean absolute error (MAE) = 5.74 and 4.44 for infiltration rate and cumulative infiltration respectively. The RF model is used to represent the infiltration characteristics of the study area. Moreover, parametric sensitivity is adopted to study the significance of each input parameter in estimating the infiltration process. The results suggest that time (t) is the most influencing parameter in predicting the infiltration process using this data set.

Quantitative Precipitation Estimates Using Machine Learning Approaches with Operational Dual-Polarization Radar Data

Traditional radar-based rainfall estimation is typically done by known functional relationships between the rainfall intensity (R) and radar measurables, such as R–Zh, R–(Zh, ZDR), etc. One of the biggest advantages of machine learning algorithms is the applicability to a non-linear relationship between a dependent variable and independent variables without any predefined relationships. We explored the potential use of two supervised machine learning methods (regression tree and random forest) in rainfall estimation using dual-polarization radar variables. The regression tree does not require normalization and scaling of data; however, this method is quite unstable since each split depends on the parent split. Since the random forest is an ensemble method of regression trees, it has less variability in prediction compared with regression trees, but consumes more computer resources. We considered several different configurations for machine learning algorithms with different sets of dependent and independent variables. The random forest model was appropriately tuned. In the test of variable importance, the specific differential phase (differential reflectivity) was the most important variable to predict the rainfall rate (residual that is the difference between the true rainfall rate and the one estimated from the R–Z relationship). The models were evaluated by 10-fold cross-validation. The best model was the random forest model using a residual with the non-classified training set. The results indicated that the machine learning algorithms outperformed the traditional R–Z relationship. Then, we applied the best machine learning model to an S-band dual-polarization radar (Mt. Myeonbong) and validated the result with ground rain gauges. The results of the application to radar data showed that the estimates of the residuals had spatial variability. The stratiform and weak rain areas had positive residuals while convective areas had negative residuals, indicating that the spatial error structure driven by the R–Z relationship was well captured by the model. The rainfall rates of all pixels over the study area were adjusted with the estimated residuals. The rainfall rates adjusted by residual showed excellent agreement with the rain gauge, especially at high rainfall rates.

Classification of Malaria Complication Using CART (Classification and Regression Tree) and Naïve Bayes

Malaria is a disease caused by the Plasmodium parasite that transmitted by female Anopheles mosquitoes. Malaria can become a dangerous disease if late have the medical treatment. The late medical treatment happened because of misdiagnosis and lack of medical staff, especially in the countryside. This problem can cause severe malaria that has complications. This study creates a system prediction to classify the severe malaria disease using Classification and Regression Tree (CART) method and the probability of malaria complication using Naïve Bayes method. The first step of this study is classifying the patients that have symptom are infected severe malaria or not based on the model that has been built. The next step, if the patient classified severe malaria then the data predicted if there any probability of complication by the malaria. There are 8 possibilities of complication malaria which are convulsion, hypoglycemia, hyperpyrexia, and the combinations of these four. The first step will evaluate by using F-score, precision and recall while the second step will evaluate by using accuracy. The highest result F-score, precision and recall are 0.551, 0.471 and 0.717. The highest accuracy 81.2% which predicted the complication is Hypoglycemia.

MR-CART: Multiresponse optimization using a classification and regression tree method

The conventional approach for optimizing multiresponse is fitting multiple response surface models and then analyzing them to obtain optimal settings for the input variables. However, it is difficult to obtain reliable response surface models when dealing with large amounts of data. In this article, a new approach to multiresponse optimization based on a classification and regression tree method is presented. Desirability functions are employed to simultaneously optimize the multiple responses. The case study of steel manufacturing company with large amounts of data shows that the proposed method obtains an optimal region in which multiple responses are simultaneously optimized.

What are the predictors of survival among patients who suffer cardiac arrest in the ED? A single-center chart review

Background: Cardiac arrest in the emergency department (ED) has not been well studied; however, the existing literature shows that there are better survival and neurological outcomes with cardiac arrest in the ED than out-of-hospital cardiac arrest (OHCA) and in-hospital cardiac arrests. The aim of this study was to examine the cardiac arrests managed in a tertiary care center ED and identify the characteristics of patients who had good outcomes. Methods: This was a retrospective chart review of patients who had cardiac arrest in ED from January 2018 to June 2019 Results: There were 209 cardiac arrests reported in the ED. A total of 189 patients were excluded because they presented with OHCA. Twenty patients had cardiac arrests in the ED. The survival to discharge from the hospital for the total patients analyzed was 40% (8/20). 60% (12/20) of patients were male with a median age of 60.5 years. Altered mental status or drowsiness was the presenting complaints of nearly 40% (8/20) of the patients, followed by dyspnea in 20% (4/20). Only 10% (2/20) reported having chest pain at presentation. 45% (9/20) of the patients had peri-intubation arrest. The most common initial rhythm was pulseless electrical activity (PEA) (70%, 14/20), followed by asystole (25%, 5/20) and ventricular fibrillation (VF), which was only 5% (1/20). 95% (19/20) of the patients had a return of spontaneous circulation with an average duration of cardiopulmonary resuscitation (CPR) of only 5 minutes. CPR duration and heart rate (HR) on arrival were the potential predictors of survival as per the classification and regression trees method. Age, gender, HR as a continuous variable, systolic blood pressure, diastolic blood pressure, mean arterial blood pressure >70 mm hg, the Modified Early Warning Score, or National Early Warning Score 2 were not statistically significant by Cox regression analysis as predictors of survival. Conclusion: Cardiac arrests occurring in ED are associated with better outcomes than OHCAs. CPR duration of less than 5 minutes coupled with presenting HR of > 110 b/min might be a predictor of survival. Larger database analyses of cardiac arrest in EDs are needed to validate this finding.

Prediction of Egg Weight from External Egg Traits of Guinea Fowl Using Multiple Linear Regression and Regression Tree Methods

The study was done to predict egg weight from the external traits of the Guinea fowl egg using the statistical methods of multiple linear regression (MLR) and regression tree analysis (RTA). A total of 110 eggs from a flock of 23-week-old Guinea fowl were evaluated. Egg weight (EW) and external traits: eggshell weight (ESW), egg polar diameter (EPD), egg equatorial diameter (EED), egg shape index (ESI), and egg surface area (ESA) were measured. Descriptive statistics, Pearson correlation coefficients, and regression equations using the MLR were obtained; additionally, a RTA was done using the CHAID algorithm with the SPSS software (IBM ver. 22). EW presented positive correlations (p<0.0001) with ESA (r = 0.72), EPD (r = 0.65), and EED (r = 0.49). EW can be predicted through MLR using ESA as a predictor variable (R2 = 72%). Predictive accuracy improves when adding EPD and EED traits to the model (R2 = 75%). The RTA built a diagram using ESA, EED, and EPD as significant independent variables; of these, the most important variable was ESA (F = 50,295, df1 = 4, and df2 = 105; Adj. p<0.000) and the variation explained for EW was 74%. Likewise, the RTA showed that the highest egg weight (41.818 g) is obtained from eggs with a surface area > 59.03 cm2 and a polar diameter > 5.10 cm. The proposed statistical methods can be used to reliably predict the egg weight of Guinea fowl.

Geographically weighted logistic regression modeling on stunting cases in Indonesia

Stunting is a condition of failure to thrive in children as a result of chronic malnutrition, so the child is too short at his/her age. Stunting harms children’s growth and affects the quality of human resources in the future. To reduce the prevalence of stunting in Indonesia, the government determined priority areas for handling stunting cases in Indonesia. This study aims to determine the variables that affect the status of priority areas for handling stunting in Indonesia. The model used in this study is Geographically Weighted Logistic Regression (GWLR) as a development of logistic regression model that considers spatial effect. This study used Maximum Likelihood Estimation (MLE) method to estimate the parameter model. The spatial weighting function used in this study is the Fixed Gaussian and Fixed Bisquare kernel weighting functions. The response and predictor variables in this study contain missing values, so Classification and Regression Tree (CART) method used to handle the missing values. The results showed that the best GWLR model on stunting cases modeling in Indonesia is the GWLR model with Fixed Bisquare kernel weighting function with AIC value of 622.806477 and model classification accuracy of 0.7257.

Application of classification and regression tree (CRT) analysis to identify the agricultural households at risk of financial exclusion

The article deals with the problem of using the classification and regression trees (CRT) method to identify socio-economic features of agricultural households at risk of financial exclusion. Financial exclusion applies to people/households who do not use or use financial services and products to a small extent, both of their own choice and due to barriers in accessing financial products and services. The research used empirical data on agricultural households in Central Pomerania (Poland). Based on the use of the decision tree method, it was shown that nearly half of the surveyed farms do not use financial products and services or use them to a small extent, therefore they are financially excluded. Higher acreage indicates a lower risk of financial exclusion. The age of the household head influences the risk of financial exclusion. The fact of saving confirms the desire to multiply capital, hence in the group of farms where this occurs, the risk of financial exclusion is lower than in the remaining ones.

Environmental factors prediction in preterm birth using comparison between logistic regression and decision tree methods: An exploratory analysis

Environmental factors prediction in preterm birth using comparison between logistic regression and decision tree methods: An exploratory analysis

Analysis of factors affecting employee loyalty of PT X in Jakarta region

Having loyal employees, i.e. those who desire to do their best in their work to achieve the company’s target, is important for the company. This study is conducted to identify factors that explain loyalty of employees in PT X, Jakarta. Furthermore, we also aim to generate profiles of the most loyal employees. Questionnaires were used to collect data from the central office and 12 branches of the company, using purposive sampling scheme, giving 467 respondents in total. Data processing was conducted using Partial Least Square (PLS) and Classification and Regression Tree (CART) methods. The results showed that job satisfaction and gender directly influence employees’ loyalty. Moreover, physiological needs, safety needs, love needs, and self-actualization needs indirectly influence loyalty of employees through job satisfaction. Highly loyal employees are those with either high job satisfaction and low physiological needs or low job satisfaction and low level of love needs.

Environmental Factors Prediction in Preterm Birth Using Comparison between Logistic Regression and Decision Tree Methods:&amp;nbsp; An Exploratory Analysis

Background and aimPreterm birth is one of the major cause of neonatal death in the developing countries and environmental factors are playing vital role in pre term birth. Nowadays Machine learning techniques are very useful for finding the hidden factors and classifications. The purpose of this study is to illustrate the importance of machine learning classification models and to identify the significant environmental factors behind pre-term birth.MethodWe have used 90 pregnant mothers, of whom 40 are preterm and 50 are full-term births. We have checked the model accuracy of the dataset through logistic regression and decision tree classifier model.ResultsThe comparative outcome of the logistic and decision tree model reveals that logistic regression is stronger in terms of metrics (precision = 0.92, F1-score = 0.96 and AUROC = 0.97), while the weak result shows by the decision tree (precision = 0.75, F1-score = 0.86 and AUROC = 0.87).ConclusionsThe conclusion shows that logistic regression is more appropriate as compare to decision tree classification model in the preterm birth data. The most influential factors for preterm birth are variables like α -HCH, total HCH and MDA (Malondialdehyde).

Derivation With Internal Validation of a Multivariable Predictive Model to Predict COVID-19 Test Results in Emergency Department Patients.

ObjectivesThe COVID‐19 pandemic has placed acute care providers in demanding situations in predicting disease given the clinical variability, desire to cohort patients, and high variance in testing availability. An approach to stratifying patients by likelihood of disease based on rapidly available emergency department (ED) clinical data would offer significant operational and clinical value. The purpose of this study was to develop and internally validate a predictive model to aid in the discrimination of patients undergoing investigation for COVID‐19.MethodsAll patients greater than 18 years presenting to a single academic ED who were tested for COVID‐19 during this index ED evaluation were included. Outcome was defined as the result of COVID‐19 polymerase chain reaction (PCR) testing during the index visit or any positive result within the following 7 days. Variables included chest radiograph interpretation, disease‐specific screening questions, and laboratory data. Three models were developed with a split‐sample approach to predict outcome of the PCR test utilizing logistic regression, random forest, and gradient‐boosted decision tree methods. Model discrimination was evaluated comparing area under the receiver operator curve (AUC) and point statistics at a predefined threshold.ResultsA total of 1,026 patients were included in the study collected between March and April 2020. Overall, there was disease prevalence of 9.6% in the population under study during this time frame. The logistic regression model was found to have an AUC of 0.89 (95% confidence interval [CI] = 0.84 to 0.94) when including four features: exposure history, temperature, white blood cell count (WBC), and chest radiograph result. Random forest method resulted in AUC of 0.86 (95% CI = 0.79 to 0.92) and gradient boosting had an AUC of 0.85 (95% CI = 0.79 to 0.91). With a consistently held negative predictive value, the logistic regression model had a positive predictive value of 0.29 (0.2–0.39) compared to 0.2 (0.14–0.28) for random forest and 0.22 (0.15–0.3) for the gradient‐boosted method.ConclusionThe derived predictive models offer good discriminating capacity for COVID‐19 disease and provide interpretable and usable methods for those providers caring for these patients at the important crossroads of the community and the health system. We found utilization of the logistic regression model utilizing exposure history, temperature, WBC, and chest X‐ray result had the greatest discriminatory capacity with the most interpretable model. Integrating a predictive model‐based approach to COVID‐19 testing decisions and patient care pathways and locations could add efficiency and accuracy to decrease uncertainty.

A Generalized Method for Modeling the Adsorption of Heavy Metals with Machine Learning Algorithms

Applications of machine learning algorithms (MLAs) to modeling the adsorption efficiencies of different heavy metals have been limited by the adsorbate–adsorbent pair and the selection of specific MLAs. In the current study, adsorption efficiencies of fourteen heavy metal–adsorbent (HM-AD) pairs were modeled with a variety of ML models such as support vector regression with polynomial and radial basis function kernels, random forest (RF), stochastic gradient boosting, and bayesian additive regression tree (BART). The wet experiment-based actual measurements were supplemented with synthetic data samples. The first batch of dry experiments was performed to model the removal efficiency of an HM with a specific AD. The ML modeling was then implemented on the whole dataset to develop a generalized model. A ten-fold cross-validation method was used for the model selection, while the comparative performance of the MLAs was evaluated with statistical metrics comprising Spearman’s rank correlation coefficient, coefficient of determination (R2), mean absolute error, and root-mean-squared-error. The regression tree methods, BART, and RF demonstrated the most robust and optimum performance with 0.96 ⫹ R2 ⫹ 0.99. The current study provides a generalized methodology to implement ML in modeling the efficiency of not only a specific adsorption process but also a group of comparable processes involving multiple HM-AD pairs.

Which Characteristics of Fecal Incontinence Predispose Incontinence-Associated Dermatitis? A Classification and Regression Tree Analysis.

To investigate which characteristics of fecal incontinence (FI) are predictors for incontinence-associated dermatitis (IAD) using the Classification and Regression Tree method. Data collected from 2014 to 2016 during the Austrian Nursing Quality Measurement 2.0, an annually conducted, cross-sectional, multicenter prevalence study, were merged and analyzed. The duration, frequency, and amount of FI were used as predictors for IAD. Nurses were asked if the participants suffered from IAD based on their clinical judgment (yes/no). In total, 1,513 participants with FI were included in this analysis. More than 75% of the participants with FI were to a great extent or completely care dependent. Of all FI participants, nearly 6% suffered from IAD, and more than 70% received special skin care for IAD prevention. Participants with FI had the highest risk of developing IAD if they experienced FI every day, had FI for less than 3 months, and had developed the FI in their current institution. Nurses face many challenges while helping patients with FI maintain healthy skin. Knowledge of the results of this study and accumulated knowledge about the specific characteristics of FI that are associated with the development of IAD can help healthcare personnel prevent IAD. Based on these results, improving patient education for persons with newly diagnosed FI to prevent IAD is recommended. Research studies should use the definition of FI established by the International Continence Society.

KLASIFIKASI STATUS KEMISKINAN RUMAH TANGGA DENGAN METODE SUPPORT VECTOR MACHINES (SVM) DAN CLASSIFICATION AND REGRESSION TREES (CART) MENGGUNAKAN GUI R (Studi Kasus di Kabupaten Wonosobo Tahun 2018)

The poor are people who have average monthly expenditures per capita below the poverty line. Wonosobo District became the poorest district in Central Java in 2011-2018, although the percentage of poor people has decreased every year. It cannot be separated from the efforts of the Wonosobo District Government to overcome poverty through various programs. This study classified households in Wonosobo District in 2018 as poor and non-poor based on influencing factors. This study used the Support Vector Machines (SVM) method to be compared with the Classification and Regression Trees (CART) method. It used the data from the 2018 National Socio-Economic Survey of Central Java with a total of 795 observations. Result of the research using the SVM method and the RBF kernel, the classification accuracy reaches 89.82% then the classification accuracy using the CART method reaches 87.08%. GUI designed by RShiny package can make easier for users to analyze the SVM and CART with the valid output.

Classification of District Employment Agencies in Terms of Employment and Cost-Effectiveness Using Regression Trees

Abstract Research background: The efficiency of the functioning of District Employment Agencies is often assessed on the basis of the level of employment and cost-effectiveness indices. The values of these indices are influenced by various socio-economic factors, which were grouped into five areas in the paper: unemployment, demography, environment, entities and the human potential of District Employment Agencies (PUPs). The research was conducted in 340 District Employment Agencies in 2017. Purpose: The purpose of the study is to separate groups of District Employment Agencies with similar values of employment and cost-effectiveness indices, with the simultaneous identification of the level of factors that characterize the socio-economic situation and staff potential in each of the separated groups. Research methodology: One of the methods of a multidimensional statistical analysis – the regression trees method was used in the work. Results: The use of regression trees allowed the separation of groups of District Employment Agencies, which differed in terms of the level of employment and cost-effectiveness indices, and characterized these groups due to socio-economic factors and staffing potential. Novelty: The survey covers all District Employment Agencies in Poland and the obtained research results can be useful for labor market institutions to assess the efficiency of PUPs.

Diagnosis of environmental controls on daily actual evapotranspiration across a global flux tower network: the roles of water and energy

Relative contributions from environmental factors to daily actual evapotranspiration (ETa) across a variety of climate zones is a widely open research question, especially regarding the roles played by soil water content ((SWC); water supply) and net radiation ((Rn); energy supply) in controlling ETa. Here, the boosted regression tree method scheme was employed to quantify environmental controls on daily ETa using the global FLUXNET dataset. Similar to the general trend suggested by the Budyko theory at annual scales, the results showed that the relative control of SWC on daily ETa increased with increasing aridity index (Φ); however, Rn played a major role at most FLUXNET sites (roughly Φ < 4), indicating that Rn could be a leading control on daily ETa even at water-limited sites. The variability in the relative controls of SWC and Rn also partly depended on factors affecting water availability for daily ETa (e.g. vegetation characteristics and groundwater depth). Our study showed that other than SWC and Rn, the net effect of environmental controls (particularly leaf area index) on daily ETa was more important at drier sites than at relatively humid sites. This suggests that near-surface hydrological processes are more sensitive to vegetation variations due to their ability to extract deep soil water and enhance ETa, especially under arid and semi-arid climatic conditions. Our findings illustrate how environmental controls on daily ETa change as the climate dries, which has important implications for many scientific disciplines including hydrological, climatic, and agricultural studies.

Statistical comparison of additive regression tree methods on ecological grassland data

Additive tree methods are widely used in ecology. To date most ecologists have used boosted regression tree (BRT) methods. However, Bayesian additive regression tree (BART) models may offer advantages to ecologists previously unexamined.Here we test whether BART has some benefits over the widely used BRT method. To do this we use two grassland data and 13 hydroclimatic and land use predictor variables. The dataset contained data from a period of drought as well as during a recovery phase after the drought. The response variable was the trend in the Enhanced Vegetation Index (EVI), which is an remotely sensed indicator of grassland degradation and recovery.The settable parameters of both methods (BRT and BART) were varied to compare the performance of each method. BRT and BART models were evaluated using three prediction error statistics; root mean square error (RMSE), mean absolute error (MAE), and the coefficient of determination (R2). The best models across the two methods were assessed by inspecting the relative importance of predictor variables and the prediction error statistics.BRT and BART models exhibited similar variable selection abilities, but the BART method generated models with similar or more favourable prediction error statistics than the BRT method (BART explained an additional 10.17% to 11.92% of the variation than BRT models). Our results indicate that BARTs may be more effective at modelling ecological data than BRTs.BARTs also had shorter run times, more reasonable defaults in its software implementation, and greater functionality of said software implementation, beyond model building and prediction functions. Ecologists using additive regression approaches may benefit from using BART approaches and we suggest their use alongside more commonly used BRT methods in ecological studies.

What makes better village economic development in traditional agricultural areas of China? Evidence from 338 villages

Villages are homes for most rural residents and powerhouses of the rural economy, with a vital role in reducing rural-urban disparity. It raises the need to explore the driving forces of village development, which in turn sheds light on strategies to secure rural revitalization. This research aims to dissect the factors contributing to village economic development through quantitative analysis of 338 villages in the traditional agricultural area of northern Jiangsu Province, China. This research leverages an integration of the econometric modeling and machine learning methodologies (multiple regression model and Boosted Regression Trees method) to ensure our findings’ accuracy and comparativeness. This research finds: (1) village economic development depends more on the intensity of land-use than the absolute amount of farmland resources; (2) transportation infrastructure, labor outmigration, and land transfer are determinants of village economic development with increasing marginal effects; non-farm employment, entrepreneurship, and bottom-up partnership are also positively related to village economic development despite their non-linear effects. (3) Large-scale labor outmigration would brings a short-term positive effect but poses a long-term challenge to village development. The authors suggest that priority should be given to the construction of transportation infrastructure, regulation of farmland transfer, industrial integration, promotion of rural entrepreneurship, and land consolidation to achieve sustainable development of the village economy.