An extension of regression trees to generate better predictive models

Introduction: No accurate diagnostic tests exist for exocrine pancreatic insufficiency (EPI). As a result, EPI is misdiagnosed, including in chronic pancreatitis (CP), and there is an unmet need for clinical tools to help clinicians evaluate patients with CP at risk for EPI. The aim of this study was to generate a simple clinical tool to predict EPI in patients with CP. Methods: 49 variables from medical records of CP patients with or without EPI were entered into a Classification and Regression Tree (CART) model using the unimputed full analysis set and 4 other models (Logistic Regression with Least Absolute Shrinkage and Selection Operator regularization, Support Vector Machine, Random Forest, and Gradient Boosting Machine) using imputed full analysis set. EPI misclassification rate (mRate) served as primary metric to train the 5 prediction models. Each model’s generalizability was assessed using 5-fold cross-validation and the model that was best suited for clinician use based on mRate and other characteristics was selected. Results: Records of 274 patients with CP from 6 pancreatitis centers across the US were included. Among the 160 (58%)/114 (42%) patients with EPI/without EPI, 48%/55% were female, mean age was 55/52 years, 76%/69% were white, 29%/52% were never smokers, and 39%/53% were never alcohol users. All 5 prediction models demonstrated similar EPI mRates. The CART model was selected. The final CART decision tree includes 10 variables (in order of importance): smoking, pancreatic atrophy, steatorrhea, alcohol use, hemoglobin, serum vitamin D, pancreatic duct contour change, serum albumin, pancreatic parenchymal calcifications, and use of opioid analgesics. The mRate without and with 5-fold cross-validation of the CART was 0.153 (training error) and 0.314 (prediction error), and the area under the receiver operating characteristic curve was 0.889 and 0.682, respectively (Figure 1). Sensitivity/specificity without/with 5-fold cross validation was 0.888/0.789 and 0.794/0.535, respectively. Conclusion: A parsimonious set of 10 variables associated with EPI in CP was identified. The CART decision tree demonstrated good prediction performance for EPI and was selected as the most practical screening tool to aid clinicians in the assessment of EPI in patients with CP. Further external validation of the EPI screening tool is planned.Figure 1.: AUC-ROC of the CART model trained on the full analysis set and after 5-fold cross-validation. AUC, area under the curve; CART, classification and regression tree; CV, cross-validation; ROC, receiver operating characteristic.

To develop a predictive system for the occurrence of nosocomial pneumonia in patients who had cardiac surgery performed. Retrospective cohort study.Setting. Two cardiologic tertiary care hospitals in Rio de Janeiro, Brazil. Between June 2000 and August 2002, there were 1,158 consecutive patients who had complex heart surgery performed. Patients older than 18 years who survived the first 48 postoperative hours were included in the study. The occurrence of pneumonia was diagnosed through active surveillance by an infectious diseases specialist according to the following criteria: the presence of new infiltrate on a radiograph in association with purulent sputum and either fever or leukocytosis until day 10 after cardiac surgery. Predictive models were built on the basis of logistic regression analysis and classification and regression tree (CART) analysis. The original data set was divided randomly into 2 parts, one used to construct the models (ie, "test sample") and the other used for validation (ie, "validation sample"). The area under the receiver-operating characteristic (ROC) curve was 69% for the logistic regression model and 76% for the CART model. Considering a probability greater than 7% to be predictive of pneumonia for both models, sensitivity was higher for the logistic regression models, compared with the CART models (64% vs 56%). However, the CART models had a higher specificity (92% vs 70%) and global accuracy (90% vs 70%) than the logistic regression models. Both models showed good performance, based on the 2-graph ROC, considering that 84.6% and 84.3% of the predictions obtained by regression and CART analyses were regarded as valid. Although our findings are preliminary, the predictive models we created showed fairly good specificity and fair sensitivity.

ABSTRAKKabupaten Tangerang mengalami pertumbuhan penduduk pesat yang mendorong perubahan tutupan lahan. Rencana Tata Ruang Wilayah (RTRW) 2020 digunakan sebagai panduan pemanfaatan lahan, dengan klasifikasi utama: lahan terbangun, badan air, vegetasi, dan lahan terbuka. Penelitian ini menganalisis perubahan tutupan lahan menggunakan algoritma Classification and Regression Trees (CART) pada citra Sentinel-2A tahun 2019 dan 2023 di Google Earth Engine. Hasil menunjukkan peningkatan luas bangunan dan lahan terbuka masing-masing 19,866 km² dan 17,877 km², sementara vegetasi dan badan air menurun 33,446 km² dan 4,297 km². Akurasi klasifikasi mendapatkan 89,36% (2019) dan 90,29% (2023). Selain itu, kesesuaian tutupan lahan dengan RTRW meningkat 36,71 km² atau 4%. Hasil ini menunjukkan efektivitas metode CART dalam memantau perubahan tutupan lahan serta relevansinya dengan kebijakan tata ruang di Kabupaten Tangerang.Kata Kunci : Tutupan lahan; Rencana Tata Ruang Wilayah (RTRW), Google Earth Engine (GEE),Classification and Regression Trees (CART), Penginderaan jauh.ABSTRACTTangerang Regency is experiencing rapid population growth that is driving land cover change. The 2020 Regional Spatial Plan (RTRW) is used as a guide for land use, with the main classifications: built-up land, water bodies, vegetation, and open land. This study analyzed land cover change using the Classification and Regression Trees (CART) algorithm on 2019 and 2023 Sentinel-2A images in Google Earth Engine. The results showed an increase in building area and open land of 19.866 km² and 17.877 km² respectively, while vegetation and water bodies decreased by 33.446 km² and 4.297 km². The classification accuracy was 89.36% (2019) and 90.29% (2023). In addition, land cover conformity with the RTRW increased by 36.71 km² or 4%. These results demonstrate the effectiveness of the CART method in monitoring land cover change and its relevance to spatial policy in Tangerang District.Keywords: Land cover, Regional Spatial Plan (RTRW), Supervised Classification, Google Earth Engine (GEE), Classification and Regression Trees (CART), Remote sensing

Despite the obvious and significant impacts of climate change and variability on crop yields in the Sahel during the past three decades, few quantitative works on potential climate change impacts on Sahelian agriculture has been conducted. Hence, this chapter presents results of predictive statistical models of peanut yields in Diourbel (Senegal) for the period 2011–2040. Rainfall indices were used as input data for a Classification and Regression Tree (CART) model aimed at forecasting the level of yield which was broken into three classes (normal, above normal, below normal). Cumulative seasonal rainfall, the percentage of days with rain, the maximum cumulative precipitation on three consecutive days, the maximum number of consecutive dry days, the maximum consecutive days of precipitation, the average precipitation intensity by wet days and the 95th percentile of daily precipitation were used to describe rainy season characteristics. The model was successfully calibrated and validated over a control period (1997–2007), then driven by the downscaled projected precipitations of six Regional Circulation Models (RCMs). These RCMs were driven by two Global Circulation Models (GCMs) under the SRES A1B scenario (AMMA-SETS). Two types of downscaling techniques were taken into consideration for the outputs of RCMs (multiple regression and Quantile-Quantile transformation). Eventually, the corrected DMI-HIRHAM5 model was selected, and corrected outputs were used to generate predictors for the CART model for the period 2011–2040. These predictors were then used to estimate future results in Diourbel. The results show yields were split between high and low normal levels for the next 30 years.

Stress is induced in response to any mental, physical or emotional change associated with our daily experiences. While short term stress can be quite beneficial, prolonged stress is detrimental to the heart, muscle tissues and immune system. In order to be proactive against these symptoms, it is important to assess the impact of stress due to various activities, which is initially determined through the change in the sympathetic (SNS) and parasympathetic (PNS) nervous systems. After acquiring physiological data wirelessly through captive electrocardiogram (ECG), galvanic skin response (GSR) and respiration (RESP) sensors, 21 time, frequency, nonlinear, GSR and respiration features were manually extracted from 15 subjects ensuing a baseline phase, virtual reality (VR) roller coaster simulation, color Stroop task and VR Bubble Bloom game. This paper presents a comprehensive physiological analysis of stress from an experiment involving a VR video game Bubble Bloom to manage stress levels. A personalized classification and regression tree (CART) model was developed using a novel Gini index algorithm in order to effectively classify binary classes of stress. A novel K-means feature was derived from 11 other features and used as an input in the Decision Tree (DT) algorithm, strong learners Ensemble Gradient Boosting (EGB) and Extreme Gradient Boosting (XGBoost (XGB)) embedded in a pipeline to classify 5 classes of stress. Results obtained indicate that heart rate (HR), approximate entropy (ApEN), low frequency and high frequency ratio (LF/HF), low frequency (LF), standard deviation (SD1), GSR and RESP all reduced and high frequency (HF) increased following the VR Bubble Bloom game phase. The personalized CART model was able to classify binary stress with 87.75% accuracy. It proved to be more effective than other related studies. EGB was able to classify binary stress with 100% accuracy, which outperformed every other related study. XGBoost and DT were able to classify five classes of stress with 72.22% using the novel K-means feature. This feature produced less error and better model performance in comparison to using all the features. Results substantiate that our proposed methods were more effective for stress classification than most related studies.

Although retaining new nurses is imperative for the future of the nursing profession, it remains a challenging task in the healthcare industry. Understanding the career journey of new graduates as they transition from students to nurses is vital. However, longitudinal studies investigating the factors influencing retention during this period are lacking. The aim of this study is to identify the influencing factors and develop a longitudinal prediction model for new graduate nurse retention. A secondary data analysis was conducted using the New Nurse e-Cohort Study dataset from two survey periods, November-December 2020 and February-March 2022. The participants were categorized into either retention or turnover groups based on their turnover experiences. A decision tree based on classification and regression tree (CART) analysis was utilized. Of the total 586 participants, 463 (79%) were in the retention group. The CART model highlighted that new nurses' retention was significantly associated with younger age, higher readiness for practice (clinical problem-solving) during the nursing program, lower transition shock (such as confusion in professional values, loss of social support, and conflicts between theory and practice), and a higher person-environment fit (person-job fit). The predictive accuracy of the CART model was 79.7%. To retain new nurses, nursing educators and hospital managers should collaborate to prepare nursing students for actual practice, offer support during organizational socialization, and foster healthy professional values for competence in the workplace. Implications for Nursing Management. Transforming the educational strategies of nursing programs and hospital management policies is imperative to ultimately enhance the retention of new graduate nurses.

Improving the prediction of ground motion parameters based on an efficient bagging ensemble model of M5′ and CART algorithms

False Financial Statements (FFS) have long been a serious problem in China and other Asian countries, which significantly dampen the confidence of the investors. Regardless of listed companies or non-listed companies, the percentage of financial statements that contained false information is quite high, which is one of the major reasons why China stock markets moved in the opposite direction towards its wonderful economic growth over the past few years. The objective of this research is to introduce one statistical technique — Classification and Regression Tree (CART), to identify and predict the impacts of FFS. We survey financial statements manipulation tricks, FFS indicators and FFS detection techniques from both China and international perspective, and further look into ten listed companies with known FFS history in China; combining these findings, we propose key indicators to work with CART. Our analysis includes 24 false financial reports, and 124 non-false financial reports. We use CART to develop two FFS detecting models: CART without industry benchmark and CART with industry benchmark. For supporting comparison, we also build a Logit regression which is a commonly used technique in FFS detecting. We find that CART is effective in distinguishing FFS from non-FFS. Both CART models achieve better accuracy in identifying fraud cases and making predictions than Logit regression does, and CART with industry benchmark is slightly better than CART without benchmark, but it does not always have superior performance. Our CART model also tries to capture the indicators and their combinations that could reflect firms with high possibility of FFS in China.

In this paper, a hybrid model consisting of the fuzzy ARTMAP (FAM) neural network and the classification and regression tree (CART) is formulated. FAM is useful for tackling the stability---plasticity dilemma pertaining to data-based learning systems, while CART is useful for depicting its learned knowledge explicitly in a tree structure. By combining the benefits of both models, FAM---CART is capable of learning data samples stably and, at the same time, explaining its predictions with a set of decision rules. In other words, FAM---CART possesses two important properties of an intelligent system, i.e., learning in a stable manner (by overcoming the stability---plasticity dilemma) and extracting useful explanatory rules (by overcoming the opaqueness issue). To evaluate the usefulness of FAM---CART, six benchmark medical data sets from the UCI repository of machine learning and a real-world medical data classification problem are used for evaluation. For performance comparison, a number of performance metrics which include accuracy, specificity, sensitivity, and the area under the receiver operation characteristic curve are computed. The results are quantified with statistical indicators and compared with those reported in the literature. The outcomes positively indicate that FAM---CART is effective for undertaking data classification tasks. In addition to producing good results, it provides justifications of the predictions in the form of a decision tree so that domain users can easily understand the predictions, therefore making it a useful decision support tool.

The comparative analysis of three methods for objective grid-based bias removal in mesoscale numerical weather prediction models is considered. The first technique is the local observation-based (LOB) method that extends further the approaches of several recent studies and is focused on utilizing the information obtained from meteorological stations or neighbor grid points in the proximity of a site of interest. The bias at a site of interest might then be considered as a spatiotemporal function of the weighted information on the past biases observed in the cluster of neighbors during a certain time window. The second method is an extension of model output statistics (MOS), combining several modern multiple regression techniques such as the classification and regression trees (CARTs) and the alternative conditional expectation (ACE) and, therefore, is named the CART–ACE method. The CART–ACE method allows representing possible nonlinear aspects of the bias in a parsimonious linearized statistical model. Finally, the third considered method is a natural combination of the LOB and CART–ACE methods in which the information provided by the LOB method is interpreted as an extra predictor in the regression model of the CART–ACE method. The proposed methods are illustrated by a case study of an observation-based verification and bias correction of fifth-generation Pennsylvania State University–National Center for Atmospheric Research Mesoscale Model (MM5) 48-h surface temperature, that is, 2-m temperature, forecasts over the Pacific Northwest.

Decision trees are a fundamental statistical learning tool for addressing classification and regression problems through a recursive partitioning approach that effectively accommodates numerical and categorical data [1, 2]. The Classification and regression tree (CART) algorithm underlies modern Boosting methodologies such as Gradient boosting machine (GBM), Extreme gradient boosting (XGBoost), and Light gradient boosting machine (LightGBM). However, the standard CART algorithm may require improvement due to its inability to learn from unlabeled data. This study proposes several modifications to incorporate test data into the training phase. Specifically, we introduce a method based on Graph-based semi-supervised learning called “Distance-based Weighting,” which calculates and removes irrelevant records from the training set to accelerate the training process and improve performance. We present Semi-supervised classification and regression tree (Semi-Cart), a new implementation of CART that constructs a decision tree using weighted training data. We evaluated its performance on thirteen datasets from various domains. Our results demonstrate that Semi-Cart outperforms standard CART methods and contributes to statistical learning.

About two-thirds of patients with minor strokes are discharged home. However, these patients may have difficulties returning to their usual living activities. To investigate the factors associated with successful home discharge, our aim was to provide a decision tree (based on clinical data) that could identify if a patient discharged home could return to pre-stroke activities and to perform an external validation of this decision tree on an independent cohort. Two cohorts of patients with minor strokes gathered from stroke registries at the Hôpital Pitié-Salpêtrière and University Hospital Bern were included in this study (n = 105 for the construction cohort coming from France; n = 100 for the second cohort coming from Switzerland). The decision tree was built using the classification and regression tree (CART) analysis on the construction cohort. It was then applied to the validation cohort. Accuracy, sensitivity, specificity, false positive, and false-negative rates were reported for both cohorts. In the construction cohort, 60 patients (57%) returned to their usual, pre-stroke level of independence. The CART analysis produced a decision tree with the Montreal Cognitive Assessment (MoCA) as the first decision point, followed by discharge NIHSS score or age, and then by the occupational status. The overall prediction accuracy to the favorable outcome was 80% in the construction cohort and reached 72% accuracy in the validation cohort. This decision tree highlighted the role of cognitive function as a crucial factor for patients to return to their usual activities after a minor stroke. The algorithm may help clinicians to tailor planning of patients' discharge.

Blasting is a widely used technique for rock fragmentation in surface mines and tunneling projects. The ground vibrations produced by blasting operations are the main concern for the industries undertaking blasting operations, which can damage the surrounding structures, adjacent rock masses, roads and slopes in the vicinity. Therefore, proper prediction of blast-induced ground vibrations is essential to demarcate the safety area of blasting. In this research, classification and regression tree (CART) as a rule-based method was used to predict the peak particle velocity through a database comprising of 51 datasets with results of maximum charge per delay and distance from the blast face were fixed as model inputs. For comparison, the empirical and multiple regression (MR) models were also applied and proposed for peak particle velocity prediction. Performance of the proposed models were compared and evaluated using three statistical criteria, namely coefficient of correlation (R2), root mean square error (RMSE) and variance account for (VAF). Comparison of the obtained results demonstrated that the CART technique is more reliable for predicting the peak particle velocity than the MR and empirical models and it can be introduced as a new technique in this field.

Comparison of CART- and MLP-based power system transient stability preventive control

ABSTRACTIn order to achieve the accurate estimation of state of charge (SOC) of the battery in a hybrid electric vehicle (HEV), this paper proposed a new estimation model based on the classification and regression tree (CART) which belongs to a kind of decision tree. The basic principle and modelling process of the CART decision tree were introduced in detail in this paper, and we used the voltage, current, and temperature of the battery in an HEV to estimate the value of SOC under the driving cycle. Meanwhile, we took the energy feedback of the HEV under the regenerative braking into consideration. The simulation data and experimental data were used to test the effectiveness of the estimation model of CART, and the results indicate that the proposed estimation model has high accuracy, the relative error of simulation is within 0.035, while the relative error of experiment is less than 0.05.