Classification And Regression Tree Research Articles

Using Machine Learning CART Decision Trees for Predicting the Causes of Delays in Projects from the Construction Industry

Construction projects are complex endeavours, with potential obstacles that can cause delays which can have particularly profound implications potentially impacting on company’s financial health, business continuity and reputation. It is becoming increasingly recognised that delays are context-specific and multifaceted, requiring more industry-oriented perceptions. This work proposes the exploratory use of Machine Learning based on Classification and Regression Trees (CART) Decision Trees (DT) to assess the predictive analysis of these approaches, considering surveys (primary data) collected from 100 specialists with different backgrounds and experiences in the construction industry. Survey responses are discussed, followed by the CART DTs, which are used as predictor for clarifying underneath relationship among different variables in a project environment. The major issue presented is related to Project Design, with “The firm is not allowed to apply for an extension of contract period”, with two possible predictors, firstly, as the main factor it is found “Mistakes, inconsistencies, and ambiguities in specification and drawing”, while other aspect highlights “Poor site supervision and management by the contractor”. The results indicate that the correct use of Artificial Intelligence techniques with relevant data are potential tools to support the analysis of scenarios and avoidance of project delays in Project Management.

Optimizing Photocatalytic Dye Degradation: A Machine Learning and Metaheuristic Approach for Predicting Methylene Blue in Contaminated Water

Dye contamination in water sources has severe environmental and public health risks; therefore, it needs effective monitoring and remediation strategies. The aim of the study is to use machine learning techniques to develop predictive models that may be used to evaluate methylene blue dye degradation capacity in contaminated water. Ten different machine learning models, including AdaBoost, Bagging, CatBoost, Decision Tree, Extra Trees, Gradient Boosting, HistGradientBoosting, LightGBM, Random Forest, and XGBoost, were evaluated using CuWO₄@TiO₂ as a photocatalyst. R², MSE, RMSE, MAE, and MedAE were used to assess the performance of the models. Among all models, HistGradientBoosting had a very well-balanced performance. It reached a very high R² of 0.9998 on the training set and 0.9915 on the test set, coupled with low error metrics, showcasing its strong generalization capability. However, Gradient Boosting and CatBoost exhibited impressive predictive performance, while AdaBoost and Decision Tree models suffered from overfitting. The maximum prediction obtained in the case of the integral approach for MB dye degradation is 98.99%. Experimental validation indicated that the effectiveness under optimized conditions reached 98.5%. In the case of initial MB concentration at 10 mg/L, a dosage of CuWO4@TiO2 photocatalyst at 200.33 mg/L, light intensity at 150 mW/cm², contact time at 88.6 minutes at room temperature, and near-neutral pH 7.0. The final model metrics included an R² score of 0.9915, MedAE of 1.171, MSE of 5.634, MAE of 1.735, and RMSE of 2.374. This work points out the possibility of taking complete advantage of advanced machine learning algorithms along with metaheuristics optimization in improving photocatalytic processes, hence opening a bright avenue for real applications in water treatment.

Random Forest and Decision Tree Facies Classification Models for Well Log Data of the Mishrif Formation from Basrah Oil Company, Southern Iraq

Facies collected from wells drilled in the study area were interpreted manually by using cores at every 10 meters of depth during well drilling. This depth of cores does not give true facies of all wells because the cores every 10 meters are considered very large. Extracting cores is financially expensive and takes a long time. The methodology of machine learning consists of four steps (Data gathering, Data preprocessing, Model training, and Model evaluation). This work intends to apply two of the supervised machine learning techniques random forest and decision tree models. (1) Data gathering, this dataset was collected from the Basrah Oil Company. It contains of ten wells (B-3, B-4, B-5, B-15, B-17, B-18, B-19, B-34, B-39, and B-40). Every well contains six features (logs): Sonic log, Resistivity Deep, Micro Spherically Focused Log, Neutron porosity, Density log, and Gamma-ray. Also, the dataset contains ten facies labels: mudstone, wack stone, packstone, roundstone, floatstone, shale, mud and wack stone, wack and pack stone, pack and grain stone and pack and float stone. These logs cover all the thickness of the Mishrif Formation, which is the goal of our study. (2) Preprocess, the data must be cleaned of outlier values; these values reduce the accuracy of the model during training. It is necessary at this stage to understand the relationships between all the features, because the highly correlated relationship between any two features, the more useless it will be in machine learning. The well B-5 blinded it for training to demonstrate the ability of machine learning models to predict lithofacies, then splitting randomly the dataset into 70% for training, validation 10%, and 20% for testing to verify the performance. (3) Training two models machine learning models decision tree and random forest. (4) Four statistics are computed for two models from the confusion matrix (accuracy of classification, recall, precision, and F1-score) showing the random forest was more accurate than the decision tree because the random forest model deals very well with this amount of dataset. Receiver Operating Characteristics curves of the random forest model have obtained the largest Area Under the Curve than the decision tree, is positive and above the main diagonal for all lithotypes, and the values for all classes reached more than 95%, except class 6 (shale), because of the 100% accuracy of classification. Facies classification by machine learning approach has two benefits (1) Increased accuracy of describing oil reservoirs and (2) Reducing the time, which a geologist needs to interpret logs data.

Research on parameter regionalization based on decision tree algorithm: a case study of 16 catchments in northeast China

ABSTRACT In order to improve the accuracy of hydrological forecasting in ungauged catchments, this study explores a new machine learning (ML) approach and applies it to the regionalization of hydrological model parameters in ungauged catchments. Taking 16 catchments located in Northeast China as the research objects, based on the Spatiotemporal Variable Source-mixed Runoff (SVSMR) Model, the splitting conditions of Classification and Regression Tree (CART) are used to quantify the rules for parameter transplantation of the model, and suitable donor basins are identified for the target basin. The results show that the SVSMR model has good applicability in the research area, with an average NSE (Nash-Sutcliffe efficiency coefficient) of 0.85 for the 16 basins and an average EQP (error of peak flow) of 5.98%. The criteria for similar watershed discrimination formulated using CART can improve the effect of hydrological model parameter transplantation, with suitable donor watersheds found for 87.5% of the target watersheds.

Predicting complete concentric collapse at the palatal level during drug-induced sleep endoscopy: an analysis of 1761 cases.

Drug-induced sleep endoscopy (DISE) helps select patients with obstructive sleep apnea (OSA) for surgery by identifying upper airway collapse patterns. This study aimed to predict the probability of complete concentric collapse at the palatal level (CCCp) during DISE based on patient characteristics, specifically body mass index (BMI). We retrospectively reviewed records of OSA patients who underwent DISE from January 2018 to July 2023. Logistic regression with receiver operating characteristic (ROC) analysis and classification and regression tree (CART) analysis were used to assess the diagnostic efficiency of BMI and other predictors for CCCp. A consecutive series of 1761 eligible patients was included for analysis. CCCp was observed in 22.3% of cases. The CCCp group had significantly higher BMI, neck circumference, apnea-hypopnea index (AHI) and height as well as increase in partial and complete collapse at oropharyngeal level. The ROC analysis for predicting CCCp for a BMI cut-off was similar for males and females 29.4kg/m² and 29.5kg/m², with an area under the curve (AUC) of 0.65 and 0.73, respectively. Adding predictors like tonsils, AHI, height and neck circumference improved the model's performance. Although confirming an association between increasing BMI and presence of CCCp, we were unable to define an accurate BMI cut-off value for predicting CCCp. The multifactorial nature of this collapse pattern challenges BMI's efficacy as a sole predictor. Our findings underscore the continued importance of DISE in evaluating CCCp and other collapse patterns for clinical decision-making in patients considered for hypoglossal nerve stimulation therapy.

Analyzing and predicting short-term substance use behaviors of persons who use drugs in the great plains of the U.S.

Substance use induces large economic and societal costs in the U.S. Understanding the change in substance use behaviors of persons who use drugs (PWUDs) over time, therefore, is important in order to inform healthcare providers, policymakers, and other stakeholders toward more efficient allocation of limited resources to at-risk PWUDs. This study examines the short-term (within a year) behavioral changes in substance use of PWUDs at the population and individual levels. 237 PWUDs in the Great Plains of the U.S. were recruited by our team. The sample provides us longitudinal survey data regarding their individual attributes, including drug use behaviors, at two separate time periods spanning 4-12 months. At the population level, we analyze our data quantitatively for 18 illicit drugs; then, at the individual level, we build interpretable machine learning logistic regression and decision tree models for identifying relevant attributes to predict, for a given PWUD, (i) which drug(s) they would likely use and (ii) which drug(s) they would likely increase usage within the next 12 months. All predictive models were evaluated by computing the (averaged) Area under the Receiver Operating Characteristic curve (AUROC) and Area under the Precision-Recall curve (AUPR) on multiple distinct sets of hold-out sample. At the population level, the extent of usage change and the number of drugs exhibiting usage changes follow power-law distributions. At the individual level, AUROC's of the models for the top-4 prevalent drugs (marijuana, methamphetamines, amphetamines, and cocaine) range 0.756-0.829 (+2.88-7.66% improvement with respect to baseline models using only current usage of the respective drugs as input) for (i) and 0.670-0.765 (+4.34-18.0%) for (ii). The corresponding AUPR's of the said models range 0.729-0.947 (+2.49-13.6%) for (i) and 0.348-0.618 (+26.9-87.6%) for (ii). The observed qualitative changes in short-term substance usage and the trained predictive models for (i) and (ii) can potentially inform human decision-making toward efficient allocation of appropriate resources to PWUDs at highest risk.

Spatiotemporal variations of cropland phosphorus runoff loss in China

Quantitative assessment of cropland phosphorus (P) loss via surface runoff is essential for developing effective pollution mitigation strategies. In this study, we compiled 812 datasets from 114 peer-reviewed papers for cropland P loss across China. We then developed machine learning (ML) approaches to estimate temporal and spatial variations in P runoff loss across China from 1990 to 2020. Four prevalent ML models were considered, namely, multiple linear regression (MLR), random forest (RF), classification and regression trees (CART), and boosted regression trees (BRT). Among these four models, RF exhibited the highest predictive accuracy for both uplands (calibration: R2 = 0.86, n = 293; validation: R2 = 0.61, n = 96) and paddy fields (calibration: R2 = 0.88, n = 137; validation: R2 = 0.60, n = 44). According to RF, China’s croplands are estimated to have lost an average of 148 ± 27 Gg P yr−1 from 1990 to 2020, with uplands and paddy fields contributing 114 ± 26 Gg P yr−1 and 34 ± 4 Gg P yr−1, respectively. There was a significant increase in upland TP runoff loss over the study period (p < 0.001), whereas paddy field TP loss remained relatively constant. Regions in southern, eastern, and southwestern China, notably in Hainan, Guangxi, and Fujian provinces, were identified as hotspots of cropland TP runoff loss. Improved cropland management scenarios were predicted to reduce TP runoff loss by 1.4–11.8 %, with the best results obtained by minimizing runoff depth. To effectively mitigate TP runoff loss, an integrated management approach involving water, soil, and fertilizer is recommended. This study enhances quantitative understanding of cropland TP runoff loss in China, providing crucial insights for efficient cropland P management, which is key to managing nonpoint source pollution on a national level.

Identification and characterization of alopecia areata subgroups regarding quality of life impairment based on demographic, clinical, psychological, and social factors

ABSTRACT The benign dermatological hair condition Alopecia Areata (AA) is known to impair Quality of Life (QoL), especially mental and social health, due to the accompanying visible appearance changes. Previous studies have identified demographic, clinical, social, and psychological variables related to QoL. Yet, the novelty of this study lies in examining how QoL differences in AA relate to (combinations of) these variables. The aim of the current study is to identify and characterize subgroups of AA persons with less or more QoL impairment by means of (combinations of) demographic, clinical, psychological, and social factors, including both potential risk (perceived stigmatization, avoidant coping, physical identity definition) and protective factors (disclosure, social support, emotion-focused coping and non-physical identity definition). An online questionnaire was filled out by 322 persons with AA, including the Dermatology Life Quality Index (DLQI) as QoL measure, Perceived Stigmatization Questionnaire (FSQ), brief COPE, Social Support Survey, a newly developed identity-definition measure, clinical characteristics, and demographics. Classification and Regression Tree (CART) analysis identified subgroups based on QoL outcome. Lowest QoL impairment was found in persons with low feelings of being flawed combined with low secretiveness. Low QoL impairment was also found in persons feeling flawed combined with low avoidant coping, low sensitivity to others’ opinions and older age. QoL impairment was intermediate in persons perceiving more social support in those with a younger age, and defining identity less on physical appearance in those sensitive to others’ opinions. Highest QoL impairment was characterized by feeling flawed combined with an avoidant coping style. Current findings provide indications for the identification of risk and protective profiles for QoL impairment and which factors to address in interventions to improve QoL in AA.

Study on predicting new onset heart failure events in patients with hypertrophic cardiomyopathy using machine learning algorithms based on clinical and magnetic resonance features

Objective: To explore the value of predicting new-onset heart failure events in patients with hypertrophic cardiomyopathy (HCM) using clinical and cardiac magnetic resonance (CMR) features based on machine learning algorithms. Methods: The study was a retrospective cohort study. Patients with a confirmed diagnosis of HCM who underwent CMR examinations at Beijing Anzhen Hospital from May 2017 to March 2021 were selected and randomly divided into the training set and the validation set in a ratio of 7∶3. Clinical data and CMR parameters (including conventional parameters and radiomics features) were collected. The endpoint events were heart failure hospitalization and heart failure death, with follow-up ending in January 2023. Features with high stability and P value<0.05 in univariate Cox regression analysis were selected. Subsequently, three machine learning algorithms-random forest, decision tree, and XGBoost-were used to build heart failure event prediction models in the training set. The model performance was then evaluated using the independent validation set, with the performance assessed based on the concordance index. Results: A total of 462 patients were included, with a median age of 51 (39, 62) years, of whom 332 (71.9%) were male. There were 323 patients in the training set and 139 in the validation set. The median follow-up time was 42 (28, 52) months. A total of 44 patients (9.5% (44/462)) experienced endpoint events (8 cases of heart failure death and 36 cases of heart failure hospitalization), with 31 events in the training set and 13 in the validation set. Univariate Cox regression analysis identified 39 radiomic features, 4 conventional CMR parameters (left ventricular end-diastolic volume index, left ventricular end-systolic volume index, left ventricular ejection fraction, and late gadolinium enhancement ratio), and 1 clinical feature (history of non-sustained ventricular tachycardia) that could be included in the machine learning model. In the prediction models built with the training set, the concordance indices for the random forest, decision tree, and XGBoost models were 0.966 (95%CI 0.813-0.995), 0.956 (95%CI 0.796-0.992), and 0.973 (95%CI 0.823-0.996), respectively. In the validation set, the concordance indices for the random forest, decision tree, and XGBoost models were 0.854 (95%CI 0.557-0.964), 0.706 (95%CI 0.399-0.896), and 0.703 (95%CI 0.408-0.890), respectively. Conclusion: Integrating clinical and CMR features of HCM patients through machine learning aids in predicting heart failure events, with the random forest model showing superior performance.

Diabetes Eye Disease Sufferers and Non-Sufferers Are Differentiated by Sleep Hours, Physical Activity, Diet, and Demographic Variables: A CRT Analysis

Introduction: Diabetic eye disease is the most common microvascular complication of diabetes mellitus. This complication has some direct impact on an individual’s well-being and health. Some lifestyle habits have been associated with the incidence of these co-morbidities. Objective: To classify the diabetic population into sufferers or non-sufferers of diabetes eye disease according to lifestyle and demographic variables, and to identify which of these variables are significant for this classification. Methods: The present cross-sectional study based on the NHANES 2011–2020 used the Classification and Regression Tree (CRT) analysis for classifying the diabetic population into sufferers and non-sufferers of diabetes eye disease. The odds ratio (OR) and relative risks (RR) of suffering this diabetes complication of the subgroups formed by the model were studied. The final sample formed 2657 individuals (1537 males and 1120 females). Results: A 79.4% accuracy was found for the CRT model. The independent variables of sleep hours (100.0%), physical activity (PA) group (92.8%), gender (76.2%), age (46.4%), education level (38.4%), sedentary time (38.1%), and diet (10.0%) were found to be significant for the classification of cases. The variable high alcohol consumption was not found significant. The analysis of the OR and RR of the subgroups formed by the model evidenced greater odds of suffering diabetes eye disease for diabetes sufferers from the inactive and walk/bicycle PA group compared to those from the Low, Moderate, and High PA groups (OR: 1.48 and RR: 1.36), for males compared to females (OR: 1.77 and RR: 1.61), for those sleeping less than 6 h or more than 9 compared to those who sleep between 6 and 8 h (OR: 1.61 and RR: 1.43), and for diabetes sufferers aged over 62 compared to younger ones (OR: 1.53 and RR: 1.40). Conclusions: sleep hours, PA group, gender, age, education level, sedentary time, and diet are significant variables for classifying the diabetic population into sufferers and non-sufferers of diabetes eye disease. Additionally, being in the inactive or walk/bicycle PA group, being a male, sleeping less than 6 or more than 9 h, and being aged over 62 were identified as risk factors for suffering this diabetes complication.

Investigation of factors affecting fresh herbage yield in pea (Pisum arvense L.) using data mining algorithms.

This study was carried out to determine the factors affecting the wet grass yield of pea plants grown in Turkey. Wet grass yield was predicted using parameters such as genotype, crude protein, crude ash, acid detergent fiber (ADF), and neutral detergent fiber (NDF) with some data mining algorithms. These techniques provided easily interpretable data trees and precise cutoff values. This led to a comparison of the predictive abilities of data mining methods, including multivariate adaptive regression spline (MARS), Chi-square automatic interaction detection (CHAID), classification and regression tree (CART), and artificial neural network (ANN). To test the compatibility of the data mining algorithms, seven goodness-of-fit criteria were used. The predictive abilities of the fitted models were assessed using model fit statistics such as the coefficient of determination (R 2), adjusted R 2, root mean square error (RMSE), mean absolute percentage error (MAPE), standard deviation ratio (SD ratio), Akaike information criterion (AIC), and corrected Akaike information criterion (AICc). With the greatest R 2 and adjusted R 2 values (0.998 and 0.986) and the lowest values of RMSE, MAPE, SD ratio, AIC, and AICc (10.499, 0.7365, 0.047, 268, and 688, respectively), the MARS method was determined to be the best model for quantifying plant fresh herbage yield. In estimating the fresh herbage production of the pea plant, the results showed that the MARS method was the most appropriate model and a good substitute for other data mining techniques.

Analysis of the possibility of using exploration and learning algorithms in the production of castings

The research presented in the article indicates the process of building models based on machine learning algorithms, linear regression, decision trees, ensemble learning, random forest, ensemble averaging, Boosting, stacking, and support vector regression (SVR) algorithms. The basis for building these models are experimental data collected during research conducted at the Łukasiewicz Research Network-Krakow Institute of Technology. An analysis of the initial state and the analysis of the state of correlation in the set were performed, which facilitated the development of models. To increase the amount of data, augmentation was performed using the Bootstrapping. For selected results, castings were made and tested in real conditions. The research results indicate the possibility of identifying the most appropriate input parameters for specific production processes of austempered ductile iron (ADI), the possibility of predicting the expected mechanical parameters based on the indicated parameters of the production process, chemical composition, specific parameters of the heat treatment process, and the thickness of the target product. A set of such models constitutes the basis of the system, enabling the end user to estimate the final parameters of the casting planned to be produced.

Optimizing random forests to detect intrusion in the Internet of Things

The Internet of Things (IoT) has created new security challenges by connecting billions of smart devices to each other. One of these challenges is detecting attacks in IoT networks. Traditional attack detection methods are usually not suitable for large and complex networks such as IoT networks. In this research, a new model for detecting intrusion in IoT networks using Software-Defined Networking (SDN) is introduced. The main goal of the current research was to improve the stability of IoT networks against various attacks using an optimized machine learning model in a distributed manner. The presented approach uses the advantages of SDN, such as flexibility and centralized control, to improve intrusion detection performance. The proposed method includes two phases: first, the topology of the network is divided into a set of subdomains, and a controller node is assigned to each subdomain. Then, in the second phase, an ensemble classification model based on a random forest is utilized for detecting intrusion in each subdomain. This learning model is a forest of classification and regression trees (CARTs), each component of which is optimized by genetic algorithm (GA). Controller nodes can use this classification model to identify intrusion independently or cooperatively. The main novelty of the current work lies in optimizing multiple learning models and cooperatively utilizing them for intrusion detection goals. In an experimental environment based on MATLAB software, the effectiveness of this model for detecting intrusions on two databases, NSW-NB15 and NSLKDD, was evaluated. The findings of the experiments showed that this model can identify the attacks in these two databases with 98.06 % and 99.67 % accuracy respectively, which is significantly higher than the compared models.

Wasserstein distributionally robust learning for predicting the cycle time of printed circuit board production

This paper proposes a Wasserstein distributionally robust learning (WDRL) model to predict the production cycle time of highly mixed printed circuit board (PCB) orders on multiple production lines. The PCB production cycle time is essential for optimizing production plans. However, the design of the PCB, production line configuration, order combinations, and personnel factors make the prediction of the PCB production cycle time difficult. In addition, practical production situations with significant disturbances in feature data make traditional prediction models inaccurate, especially when there is new data. Therefore, we establishe a WDRL model, derive a tight upper bound for the expected loss function, and reformulate a tractable equivalent model based on the bound. To demonstrate the effectiveness of this method, we collected data related to surface mounted technology (SMT) production lines from a leading global display manufacturer for our computational experiments. In addition, we also designed experiments with perturbations in the training and testing datasets to verify the WDRL model’s ability to handle perturbations. This proposed method has also been compared with other machine learning methods, such as the support vector regression combined with symbiotic organism search, decision tree, and kernel extreme learning machine, among others. Experimental results indicate that the WDRL model can make robust predictions of PCB cycle time, which helps to effectively plan production capacity in uncertain situations and avoid overproduction or underproduction. Finally, we implement the WDRL model for the actual SMT production to predict the production cycle time and set it as the target for production. We observed a 98–103 % achievement rate in the last 20 months since the implementation in September 2022.

Spatiotemporal Dynamics and Driving Factors of Soil Salinization: A Case Study of the Yutian Oasis, Xinjiang, China

Soil salinization is a critical global environmental issue, exacerbated by climatic and anthropogenic factors, and posing significant threats to agricultural productivity and ecological stability in arid regions. Therefore, remote sensing-based dynamic monitoring of soil salinization is crucial for timely assessment and effective mitigation strategies. This study used Landsat imagery from 2001 to 2021 to evaluate the potential of support vector machine (SVM) and classification and regression tree (CART) models for monitoring soil salinization, enabling the spatiotemporal mapping of soil salinity in the Yutian Oasis. In addition, the land use transfer matrix and spatial overlay analysis were employed to comprehensively analyze the spatiotemporal trends of soil salinization. The geographical detector (Geo Detector) tool was used to explore the driving factors of the spatiotemporal evolution of salinization. The results indicated that the CART model achieved 5.3% higher classification accuracy than the SVM, effectively mapping the distribution of soil salinization and showing a 26.76% decrease in salinized areas from 2001 to 2021. Improvements in secondary salinization and increased vegetation coverage were the primary contributors to this reduction. Geo Detector analysis highlighted vegetation (NDVI) as the dominant factor, and its interaction with soil moisture (NDWI) has a significant impact on the spatial and temporal distribution of soil salinity. This study provides a robust method for monitoring soil salinization, offering critical insights for effective salinization management and sustainable agricultural practices in arid regions.

The relative importance of herbicide use for conservation tillage adoption by U.S. corn and soybean producers.

Herbicide use is widespread in agricultural production to control weeds prior to and after planting and to "burndown" weeds in the spring for conservation tillage. Whether conservation tillage adoption leads to higher herbicide usage has been a question of policy relevance for decades in the United States. Older U.S. studies using standard statistical and economic techniques have not consistently demonstrated higher herbicide usage levels among producers practicing conservation tillage, but these studies did not fully account for other practices, economic, or agronomic factors. To provide a more timely and comprehensive understanding of the importance of herbicides to conservation tillage, this study achieves two objectives with the most recent, nationally representative data from the US Department of Agriculture. First, it describes trends and compares conservation tillage and herbicide usage among field corn and soybean producers, similar to previous studies using standard economic techniques. Second, a Classification and Regression Tree (CART) model is employed-a novel methodology relative to previous studies that offers distinct advantages over traditional regression modeling-to evaluate the importance of herbicide use for conservation tillage adoption while accounting for other factors. Pairwise mean comparisons for field corn and soybeans indicated that herbicide usage pre-emergence was significantly higher with conservation tillage, but there was no consistent, significant differences in herbicide usage post-emergence. The CART analysis (with prediction accuracy ranging from 68-72%) also showed that pre-emergent use of glyphosate was the strongest predictor (with predicted probabilities from 0.83-0.86) of conservation tillage for field corn in 2016 and soybeans in 2018. Other factors such as the use of crop rotations, highly erodible land, region, and farm size were also strong predictors of conservation tillage. These findings highlight the importance and complexity of herbicide use in the adoption of conservation tillage for U.S. field corn and soybeans.

Using artificial intelligence in education: decision tree learning results in secondary school students based on cold and hot executive functions

Improving educational quality is a universal concern. Despite efforts made in this regard, learning outcomes have not improved sufficiently. Therefore, further investigation is needed on this issue, adopting new perspectives (conceptual and analytical) to facilitate the understanding and design of effective actions. The objective of this study was to determine the influence of executive functions (considering both cognitive and affective processes) and their interactions on learning outcomes in Language and Literature and Mathematics in Spanish students, through the use of artificial intelligence, based on the machine learning approach, and more specifically, the decision tree technique. A total of 173 students in compulsory secondary education (12–17 years old) from the same educational institution participated. The school’s educational counsellor provided information on student executive function levels by completing the BRIEF2 questionnaire for each participant. She also reported on the learning outcomes achieved by students in the subjects of interest for this research (Language and Literature and Mathematics). R software was used to model the regression trees. The results revealed groups of students characterised by different profiles, i.e., by different combinations of difficulties in various executive functions and varying levels of learning outcomes in each academic area. However, regardless of the academic area considered (Language and Literature or Mathematics), working memory was identified as the most relevant executive function in all of the students’ learning outcomes. Understanding the combination of executive functions that predict learning outcomes in each group of students is important since it enables teachers and other educational professionals, policymakers and researchers to provide individualised educational resources according to the diverse student profiles and needs. It constitutes an effective mechanism to improve students’ learning results and, ultimately, to enhance an equitable and more effective educational system.

Ralationship between polymorphisms and diplotypes of HLA-G 3’UTR and fetuses with abnormal chromosomes or unexplained pregnancy loss (UPL)

ObjectivesHuman leukocyte antigen G (HLA-G) plays a crucial role in pregnancy. Pregnancy loss (PL) is caused by a variety of causes, such as fetal chromosomal abnormalities, maternal hypertension and diabetes, immune causes, spontaneous immune diseases, infections, unknown causes, etc. This study reports on the association of fetal HLA-G 3’UTR polymorphisms and diplotypes with chromosomally abnormal fetuses (CAF) or unexplained pregnancy loss (UPL).MethodsA total of 552 specimens were collected and grouped by next-generation sequencing technology (NGS) and fetal survival: UPL (112 cases), CAF (170 cases) and control (258 cases). The polymorphisms of HLA-G 3’UTR in all samples were detected by Sanger sequencing. The genotypes, haplotypes and diplotypes of HLA-G 3’UTR were analyzed. The classification and regression tree (CART) analysis was used to evaluate the role of HLA-G diplotypes in predicting fetal outcomes. The correlations between CAF or UPL and maternal age, paternal age, times of miscarrage, times of delivery were analyzed by logistic regression.ResultsThe frequencies of HLA-G + 2960del/del and + 3035CC genotypes were remarkablly increased in CAF than those in control group. The frequencies of HLA-G + 2960ins/del, + 3010CC, + 3035TC, + 3142GG, + 3187AA in CAF were significantly lower than those in normal fetuses. Through genetic models and logistic regression analysis, the dominant model of HLA-G 3’UTR genotypes [such as + 2960 (OR = 1.27, 95% CI = 1.05–1.54, p = 0.016), + 3010 (OR = 0.78, 95% CI = 0.63–0.97, p = 0.026), + 3035 (OR = 1.22, 95% CI = 1.00–1.49, p = 0.047), + 3142 (OR = 0.76, 95% CI = 0.62–0.95, p = 0.014) and + 3187 (OR = 0.80, 95% CI = 0.65–0.99, p = 0.041)] were dramatically associated with CAF. However, the frequencies of HLA-G + 3010GC, + 3142GC and + 3187AG in fetuses with UPL were memorably decreased than those in normal fetuses. No significant difference was found in the frequencies of HLA-G haplotypes in all groups. However, the frequency of UTR-1 positive specimens in CAF was significantly higher than that in UPL and control group. At the same time, the frequency of UTR-1/UTR-3 diplotypes in CAF was observably higher than that in UPL and control group, while the UTR-1/UTR-7 frequency in UPL was signally lower than that in control group. Multivariate logistic regression analysis indicated that positive HLA-G UTR-1 (OR = 1.8, 95% CI = 1.16–2.81, p = 0.009), times of abortion (OR = 1.23, 95% CI = 1.02–1.50, p = 0.035), and times of delivery (OR = 0.31, 95% CI = 0.20–0.48, p < 0.001) were correlated with CAF.ConclusionsThis study suggests that HLA-G 3’UTR polymorphisms and diplotypes play an important role in the process of successful pregnancy of the embryos with abnormal chromosomes after fertilization. At the same time, Different alleles or diplotypes also affect the development of embryos with UPL.

Fear of movement interacts with trunk mobility, and pain intensity to predict disability in patients with chronic low back pain: a classification and regression tree (CART) analysis.

Low back pain significantly impacts global health, particularly in low-income areas where primary care challenges are prevalent. Assessing both biophysical and psychological factors associated with low back pain-related disability is crucial for developing effective screening and management strategies. This study aimed to examine the association between fear of movement, trunk mobility, pain intensity, physical activity, and low back pain-related disability, and to identify key factors contributing to higher disability levels in patients with chronic low back pain. This cross-sectional study analyzed data from 381 chronic low back pain patients aged over 18, collected at a primary care ambulatory between 2015 and 2019. Variables measured included pain intensity, fear of movement, physical activity, trunk mobility, and disability. A Classification and Regression Tree approach was used to model disability predictors. Fear of movement was the main driver of high disability (OR = 17.3, 95%CI 8.9-33.7). Two profiles were particularly prone to higher disability: patients with high fear of movement (>47.5) and pain intensity > 1.5; and those with lower fear of movement but poor trunk mobility (>36.5 cm). Patients with better trunk mobility and low pain intensity (≤3.5) showed significantly less disability (OR = 0.06, 95%CI 0.02-0.17). The findings highlight the need for integrating better psychological and biophysical assessments in managing chronic low back pain. Understanding how fear of movement, trunk mobility, and pain interaction can improve screening accuracy and optimize care in resource-limited settings, using valid and feasible tools for these scenarios.

A comparative analysis of prediction problems utilizing Interval type-2 fuzzy and machine learning models

This paper compares the effectiveness of Interval Type 2 Fuzzy Logic (IT2FL) and Machine Learning (ML) models in addressing real-world challenges. It employs four non-parametric ML algorithms (Support Vector Machine (SVM), K-Nearest Neighbor (KNN), (Random Forest (RF) and Classification and Regression Tree (CART)) and evaluates their performance against IT2FL using parameters generated through Gaussian membership functions. Initially, the IT2FL algorithm preprocesses the dataset by predicting missing values, leveraging IT2F membership to optimize decision-making and mitigate uncertainties. The study assesses the predictive performance, robustness, and interpretability of IT2F-ML models, analyzing datasets from cardiovascular disease patients to predict shock levels. After transforming the dataset using IT2FL, it is divided into 60% training and 40% testing sets to train and test four ML algorithms, aimed at predicting shock levels in patients. The models’ performance is evaluated using various metrics, demonstrating the superior learning enhancement and performance of the IT2F-SVR method compared to other predictive approaches on the same dataset. Moreso, it is observed that the integration of the IT2-FL algorithm with machine learning models offers a promising approach for addressing data incompleteness and improving prediction accuracy. Furthermore, the study provides valuable guidance for researchers and practitioners in selecting suitable methodologies for prediction tasks.