Decision Tree Model Research Articles

In-Motion, Non-Contact Detection of Ties and Ballasts on Railroad Tracks

This study aims to develop a robust and efficient system to identify ties and ballasts in motion using a variety of non-contact sensors mounted on a robotic rail cart. The sensors include distance LiDAR sensors and inductive proximity sensors for ferrous materials to collect data while traversing railroad tracks. Many existing tie/ballast health monitoring devices cannot be mounted on Hyrail vehicles for in-motion inspection due to their inability to filter out unwanted targets (i.e., ties or ballasts). The system studied here addresses that limitation by exploring several approaches based on distance LiDAR sensors. The first approach is based on calculating the running standard deviation of the measured distance from LiDAR sensors to tie or ballast surfaces. The second approach uses machine learning (ML) methods that combine two primary algorithms (Logistic Regression and Decision Tree) and three preprocessing methods (six models in total). The results indicate that the optimal configuration for non-contact, in-motion differentiation of ties and ballasts is integrating two distance LiDAR sensors with a Decision Tree model. This configuration provides rapid, accurate, and robust tie/ballast differentiation. The study also facilitates further sensor and inspection research and development in railroad track maintenance.

Classifiers of anterior cruciate ligament status in female and male adolescents using return-to-activity criteria.

A lack of standardization exists for functional tasks in return-to-activity (RTA) guidelines for adolescents with anterior cruciate ligament injury (ACLi). Identifying the variables that discern ACLi status among adolescents is a first step in the creation of such guidelines following surgical reconstruction. This study investigated the use of classification models to discern ACLi status of adolescents with and without injury using spatiotemporal variables from functional tasks typically used in RTA guidelines for adults. Sixty-four adolescents with ACLi and 70 uninjured adolescents completed single-limb hops, lunges, squats, countermovement jumps and drop-vertical jumps. Jumping distances, heights, and depths were collected. Decision trees (DTs) were used to classify ACLi status and were evaluated using the F-measure (F1), kappa statistic (ĸ) and area under the precision-recall curve (PRC). Independent t tests and effect sizes were calculated for each important classifier of the DT models. A five-variable model classified ACLi status with an accuracy of 67.5% (F1 = 0.6842; ĸ = 0.350; PRC = 0.491) with sex as a classifier. Significant differences were found in three of the four spatiotemporal variables (p ≤ 0.002). Separate models then classified ACLi status in males and females with an accuracy of 53.3% (F1 = 0.5882; ĸ = 0.0541; PRC = 0.476) and 76.9% (F1 = 0.7692; ĸ = 0.541; PRC = 0.528), respectively, with significant differences for all variables (p ≤ 0.013). Among the DT models, females were better able to classify ACLi status compared to males, highlighting the importance of sex-specific rehabilitation guidelines for adolescents. Level III.

Battery Condition Monitoring of Quadrotor UAV Using Machine Learning Classification Algorithm

Unmanned aerial vehicle flight performance and efficiency rely on various factors. Flight instabilities can happen due to malfunctions inside the system and disturbances from the external environment. Battery status plays a significant role in healthy flight conditions. A weak battery will affect the performance of propellers and motors, and the presence of wind disturbance can contribute towards inefficient flying capabilities. Therefore, investigation of fault at the early stage is crucial to maintain the great performance of the UAV. This paper aims to investigate the best prediction system from the existing machine learning algorithm such as Decision Tree (DT), Linear Discriminant (LD), Naïve Bayes (NB), Support Vector Machine (SVM), K-Nearest Neighbors (KNN) and Neural Network (NN) to classify the battery condition of the quadrotor by extracting the features from the displacement time series dataset. By using recorded flight data, it will be statistically analyzed to extract the flying condition features. The extracted features are the Euclidian distance (ED), speed, acceleration, Periodogram Power spectral density (PSD) and Fast Fourier Transform (FFT) of the signal. The result shows that the two best classifier algorithms are the Decision Tree and Neural Network models with training accuracy of 98% and 93% in Set A and B, respectively.

Using Supervised and Unsupervised Machine Learning Models to Analyze Students Academic Performance

Examination result repositories generated by most universities can serve as machine learning datasets for training various models to gain insights from the data. These datasets can train multiple linear regression models to determine a student's cumulative grade point average (CGPA), or the score that a student will get in specific courses. Additionally, classification-based supervised machine learning models can use these datasets to provide insights into the class result that a student will obtain. These insights can be invaluable for academic advising and early intervention. Moreover, these datasets can train clustering-based unsupervised machine learning models, such as the K-means clustering model, to understand how student results are grouped into various clusters. This information can be crucial for planning and evaluating the quality of the university. This paper uses the dataset of undergraduate students' examination results from the Department of Computer Science at the University of Nigeria, Nsukka, to train three supervised machine learning models and one unsupervised machine learning model, utilizing Jupyter Notebook as the Python IDE. The training results showed acceptable accuracies of 91.5% for the Naïve Bayes model and 95.1% for the Decision Tree model. The linear regression model demonstrated a negligible root mean square error of 8.23×10−18, while the K-means clustering model exhibited an acceptable Silhouette metric of 0.12.

Fast CU decision method based on texture characteristics and decision tree for depth map intra-coding

As the demand for higher quality 3D videos continues to grow, the 3D extensions of the Efficient Video Coding (3D-HEVC) standard are gradually unable to meet the needs of users. Versatile Video Coding Standard (H. 266/VVC), as an advanced video coding standard, adopts the nested Multi-Type Tree quadtree (QTMT) partitioning structure that current fast depth map coding unit (CU) partitioning methods cannot apply. Therefore, we have designed a fast intra-frame CU partitioning algorithm for VVC 3D video depth maps. Our proposed algorithm in this article consists of two steps, including two sub-algorithms. First, by analyzing the relationship between image entropy and variance and depth map CU division, we establish a bi-criterion decision algorithm to determine whether the texture complexity of the current CU is low enough to terminate its partitioning process. Then, for CUs that have been determined by the first algorithm to need further partitioning, we use a decision tree model based on Light Gradient Boosting Machine (LGBM) to predict which direction of Rate-Distortion Optimization (RDO) calculation can be skipped, which can avoid some unnecessary RDO calculations in a certain direction. The final experiment demonstrated the effect of the proposed algorithm, which can reduce 47.65% complexity of VVC 3D video intra-coding with negligible 0.23% Bjøntegaard Delta Bitrate (BDBR) increase, superior to other advanced methods.

Inverse design of high-strength medium-Mn steel using a machine learning-aided genetic algorithm approach

To develop medium-Mn steels with an ultimate tensile strength (UTS) exceeding 2 GPa and excellent ductility, we created a highly accurate UTS prediction machine learning (ML) model using a boosted decision tree model and 1520 dataset of tensile properties of medium-Mn steels with micro-alloying elements. We also optimized the hyper-parameters of a genetic algorithm (GA) using the Shannon diversity index to enhance search efficiency while retaining diversity. In a high-dimensional search space with millions of potential combinations, the ML-GA approach efficiently identified diverse chemical compositions and austenitizing conditions to achieve UTSs above 2 GPa. The k-means clustering method then grouped them into five distinct specimens based on similarities. These five specimens, fabricated using inversly designed chemical compositions and austenitizing temperatures, successfully exhibited UTSs exceeding 2 GPa and greater ductility compared to hot-stamped C steels. These excellent tensile properties were attributed to grain refinement resulting from the low austenitizing temperature and the pinning effect of micro-alloying element carbides, such as TiC and VC.

Predictors of preparedness for well-dying among middle-aged and older adults.

This study investigated psychosocial factors related to preparedness for well-dying among middle-aged and older adults and verified a model that can predict it along with psychosocial factors and demographic profiles to provide useful information for intervention. The participants were 340 middle-aged and older Korean adults aged 40-79 years. The predictive models were verified using stepwise regression and decision tree analyses. The results revealed that personality, meaning of life, hardiness, intrinsic religiosity, death anxiety, family support, subjective well-being, and expectations for future life were significantly correlated with preparedness for well-dying among middle-aged and older adults. A stepwise regression analysis revealed that meaning of life accounted for the greatest variance in preparedness for well-dying. A decision tree model predicting preparedness for well-dying included search for meaning, the presence of a spouse, family support, having a job, tenacity, conscientiousness, and positive emotions. These findings suggest that search for meaning and tenacity in middle-aged and older adults as well as the presence of a spouse or occupation and family support may play important roles in preparing for well-dying.

Quantum-Enhanced Generalized Pattern Search Optimization

While the development of quantum computers promises a myriad of advantages over their classical counterparts, care must be taken when designing algorithms that substitute a classical technique with a potentially advantageous quantum method. The probabilistic nature of many quantum algorithms may result in new behavior that could negatively impact the performance of the larger algorithm. The purpose of this work is to preserve the advantages of applying quantum search methods for generalized pattern search algorithms (GPSs) without violating the convergence criteria. It is well known that quantum search methods are able to reduce the expected number of oracle calls needed for finding the solution to a search problem from O(N) to O(N) However, the number of oracle calls needed to determine that no solution exists with certainty is exceedingly high and potentially infinite. In the case of GPS, this is a significant problem since overlooking a solution during an iteration will violate a needed assumption for convergence. Here, we overcome this problem by introducing the quantum improved point search (QIPS), a classical–quantum hybrid variant of the quantum search algorithm QSearch. QIPS retains the O(N) oracle query complexity of QSearch when a solution exists. However, it is able to determine when no solution exists, with certainty, using only O(N) oracle calls.

Optimization scheme design in production process based on multi-stage decision-making

This paper studies the control of defective rate in the electronic industry chain, designs a small sample binomial distribution sampling detection scheme to verify the supplier 's defective rate statement and optimize the detection cost. Based on the defective rate, the linear programming and dynamic programming / decision tree model are used to optimize the multi-stage production process, including parts inspection, finished product inspection and unqualified product treatment, so as to maximize the profit of the enterprise. A phased optimization strategy is proposed for complex multi-process situations. Finally, according to the sampling test results, the defect rate estimation is adjusted, the problem model is re-optimized, and the cost control reference scheme is provided.

Cost-Effectiveness of Influenza Vaccination in Healthy Children: A 10-Year Population-Based Study

Background/Objectives: Seasonal influenza annually puts a significant burden on the pediatric population, especially the youngest, causing severe illness and death. Additionally, associated healthcare costs cause a significant financial strain on healthcare systems. While vaccination is the most effective prevention method, its cost-effectiveness for healthy children remains unassessed. Methods: Using the Pedianet database spanning from 2009 to 2019, we analyzed influenza cases among 6-month-olds to 14-year-olds in Italy. Data included influenza-related medical visits, prescriptions, exams, emergency visits, hospitalizations, and costs. Adverse events and quality-adjusted life years (QALYs) were considered from the existing literature. A static decision-tree model compared annual vaccination strategies, assessing probabilities for influenza or influenza-like illnesses by vaccination status. Incremental cost-effectiveness ratios (ICERs) were calculated, along with sensitivity analyses and cost-effectiveness acceptability curve generation. Results: Mean total influenza costs for vaccinated children averaged EUR 18.6 (range 0–3175.9, including EUR 15.79 for the influenza vaccination), whereas costs for unvaccinated children were consistently lower at around EUR 4.6 (range 0–3250.1). The average ICER for years where vaccine and virus strains are matched was EUR 29,831 per QALY, which is below the EUR 40,000 threshold set by the Italian National Health Services. The ICER values range from EUR 13,736 (2017/2018) to EUR 72,153 (2013/2014). Averted influenza costs averaged EUR 23 per case, with fluctuations over the years. In most observed years, influenza vaccination was cost-effective from the healthcare providers’ standpoint. The exception was 2009–2010, due to a mismatch between vaccine and virus strains. Conclusions: This study highlights the economic viability of influenza vaccination, especially when virus and vaccine strains align. It demonstrates the potential of vaccination programs in preserving children’s health and well-being while managing healthcare costs.

Development of a decision tree model for predicting the malignancy of localized gingival enlargements based on clinical characteristics.

The present study aimed to determine the prevalence of localized gingival enlargements (LGEs) and their clinical characteristics in a group of Thai patients, as well as utilize this information to develop a clinical diagnostic guide for predicting malignant LGEs. All LGE cases were retrospectively reviewed during a 20-year period. Clinical diagnoses, pathological diagnoses, patient demographic data, and clinical information were analyzed. The prevalence of LGEs was determined and categorized based on their nature, and concordance rates between clinical and pathological diagnoses among the groups were evaluated. Finally, a diagnostic guide was developed using clinical information through a decision tree model. Of 14,487 biopsied cases, 946 cases (6.53%) were identified as LGEs. The majority of LGEs were reactive lesions (72.62%), while a small subset was malignant tumors (7.51%). Diagnostic concordance rates were lower in malignant LGEs (54.93%) compared to non-malignant LGEs (80.69%). Size, consistency, color, duration, and patient age were identified as pivotal factors to formulate a clinical diagnostic guide for distinguishing between malignant and non-malignant LGEs. Using a decision tree model, we propose a novel diagnostic guide to assist clinicians in enhancing the accuracy of clinical differentiation between malignant and non-malignant LGEs.

Understanding the Relationship Between Comorbidities, Medication Nonadherence, Activities of Daily Living, and Heart Condition Status Among Older Adults in the United States: A Regression Analysis and Machine Learning Approach.

Nonadherence to medication among patients with heart disease poses serious risks, including worsened heart failure and increased mortality rates. This study aims to explore the complex interplay between comorbidities, medication nonadherence, activities of daily living, and heart condition status in older American adults, using both traditional statistical methods and machine learning. Data from 326 older adults with heart conditions, drawn from the Health and Retirement Study, were analyzed. Descriptive statistics characterized demographic profiles and comorbidities, whereas logistic regression, multiple regression analyses, and decision tree models were used to address our research inquiries. In addition, a machine learning approach, specifically decision tree models, was integrated to enhance predictive accuracy. Our analysis showed that factors like age, gender, hypertension, and stroke history were significantly linked to worsening heart conditions. Notably, depression emerged as a robust predictor of medication nonadherence. Further adjusted analyses underscored significant correlations between stroke and challenges in basic activities such as dressing, bathing, and eating. Depression correlated significantly with difficulties in dressing, bed mobility, and toileting, whereas lung disease was associated with bathing hindrances. Intriguingly, our decision tree model revealed that patients experiencing dressing challenges, but not toileting difficulties, were more prone to report no improvement in heart condition status over the preceding 2 years. Blending traditional statistics with machine learning in this study reveals significant implications for crafting personalized interventions to improve patients' depression, leading to increased activities of daily living, medication adherence, reduced severity of comorbidities, and ultimately better management of heart conditions.

Predictive modeling of arginine vasopressin deficiency after transsphenoidal pituitary adenoma resection by using multiple machine learning algorithms

This study aimed to predict arginine vasopressin deficiency (AVP-D) following transsphenoidal pituitary adenoma surgery using machine learning algorithms. We reviewed 452 cases from December 2013 to December 2023, analyzing clinical and imaging data. Key predictors of AVP-D included sex, tumor height, preoperative and postoperative changes in sellar diaphragm height and pituitary stalk length, preoperative ACTH levels, changes in ACTH levels, and preoperative cortisol levels. Six machine learning algorithms were tested: logistic regression (LR), support vector classification (SVC), random forest (RF), decision tree (DT), k-nearest neighbors (KNN), and extreme gradient boosting (XGBoost). After cross-validation and parameter optimization, the random forest model demonstrated the highest performance, with an accuracy (ACC) of 0.882 and an AUC of 0.96. The decision tree model followed, achieving an accuracy of 0.843 and an AUC of 0.95. Other models showed lower performance: LR had an ACC of 0.522 and an AUC of 0.54; SVC had an ACC of 0.647 and an AUC of 0.67; KNN achieved an ACC of 0.64 and an AUC of 0.70; and XGBoost had an ACC of 0.794 and an AUC of 0.91. The study found that a shorter preoperative pituitary stalk length, significant intraoperative stretching, and lower preoperative ACTH and cortisol levels were associated with a higher likelihood of developing AVP-D post-surgery.

The Effect of Dataset Limitation on the Diabetes Risk Factors

Diabetes research is an important topic in the medical domain. However, the research on diabetes risk factors always restricts objects to a certain category. This research wants to figure out the effect of such limitations. The result of this study could give guidance in this field. A dataset built by the Center of Disease Control and Prevention, which has been cleaned is used in this study. The binomial logistic regression analysis is applied to this research and the decision tree model is used for testing. The results suggest that many environmental factors are associated with diabetes. However, taking these risk factors together did not accurately predict diabetes. By analyzing and comparing, limitations on the dataset could increase the accuracy of prediction. The result states that limitations during the study on diabetes risk factors could lead to high-accuracy correlation results. This study provides a good guide for other research on diabetes risk factors to investigate people with more common points in their daily lifestyle.

Clinical strategy study on prenatal screening and diagnostic model for Down syndrome

Exploring efficient and easily implementable prenatal screening strategies aims at birth defect prevention and control. However, there have been limited economic evaluations of non-invasive prenatal screening (NIPS) strategies in China. Furthermore, these studies were predominantly confined to local or geographically proximate provinces and lacked universality and representativeness. This study assesses the health economics of current prenatal screening strategies and NIPS as first-line screening programs, analyzing their efficacy to determine an optimal strategy. From the perspective of health economics, cost-effectiveness, cost-benefit, and single-factor sensitivity were conducted for five different screening strategies using a decision tree model. Among pregnant women aged < 35 years who underwent only one screening for foetal Down syndrome (DS), the detection rate, false positive rate and positive predictive value of NIPS for foetuses with DS were superior to those of the other four serological screening methods. Although applying NIPS as first-line screening method yields the highest efficacy and benefits, it currently lacks cost-effectiveness when compared to serological screening and sequential NIPS screening strategies.

An efficient approach for EMG controlled pattern recognition system based on MUAP identification and segregation

An Electromyography (EMG) based pattern recognition system constitutes various steps of signal processing and control engineering from signal acquisition to real-time control. Efficient control of external devices largely depends on the signal processing steps executed before the final output. This work presents a new approach to signal processing using Motor Unit Action Potential (MUAP) based signal decomposition and segmentation. An MUAP is a neurological response during muscle contraction. Due to the higher contact area of surface electrodes, MUAPs from multiple muscles are captured. An MUAP generated from a single muscle usually has identical waveshapes and similar discharging rates and usually lasts for 8–15 ms. These are known as primary MUAPs. The proposed algorithm identifies and uses the primary observed MUAPs for feature extraction and classification. Firstly, noise signals are eliminated by a determined noise margin, which also separates the active muscle movement signals. Next, a novel MUAP identification algorithm is implemented to detect the MUAP trains. Then, identified primary MUAPs are used to make segments with variable widths to extract feature vectors. Based on the correlation score of all the primary MUAPs, the segmentation is performed, which results in segmentation width varying from 110–200 ms. The achieved segmentation width is lesser than the conventional overlapping and non-overlapping methods — the proposed approach results in a 20 to 50% reduction in the segmentation width. Four different classifiers are tested during the machine learning stage to investigate the performance of the proposed approach. The obtained feature sets are then used to train the Linear Discriminant Analysis (LDA), K-Nearest Neighbor (kNN), Decision Tree (DT), and Random Forest (RF) classifiers. The classifiers are tested with precision, recall, F1 score, and accuracy. The kNN and DT classifiers performed better than the LDA and RF classifiers. The maximum precision and recall are 100% while the maximum achieved accuracy is 98.56%. The comparative results show higher accuracy even at lower segmentation widths than the conventional constant window scheme. The kNN and DT classifiers provide a 5% to 15% increment in accuracy compared to the constant window segmentation-based approach.

Regional Impact of the COVID-19 Pandemic on Cancer Screening.

Cancer screening is the most cost-effective prevention strategy that enables improved cancer mortality and morbidity rates. However, there was a significant decline in cancer screening during the COVID-19 pandemic worldwide. This study aimed to measure regional differences in cancer screening by comparing screening rates before and during COVID-19 and to examine factors explaining the regional difference in cancer screening. All variables including the differences in cancer screening rate before and during the COVID-19 pandemic were collected from 229 administrative subregions in South Korea and joined with each community in ArcGIS environment. Factors explaining regional difference in cancer screening were analyzed through descriptive statistics, correlation analysis, and interactive decision tree modeling. Cancer screening rates decreased in all regions during the COVID-19 pandemic in South Korea. The findings revealed that national health insurance premiums, the number of cancer screening facilities, livelihood benefit recipients, and the population aged 65+ could be classified as variables highly affecting the decrease in cancer screening rate. These findings highlight the efforts to appraise different barriers to cancer screening in each community. In addition, it is important to promote sustainable cancer screening strategies from regional perspectives based on regional vulnerabilities and their complex impact on cancer screening.

Enhancing hydrological predictions: optimised decision tree modelling for improved monthly inflow forecasting

ABSTRACT The utilisation of modelling tools in hydrology has been effective in predicting future floods by analysing historical rainfall and inflow data, due to the association between climate change and flood frequency. This study utilised a historical dataset of monthly inflow and rainfall for the Terengganu River in Malaysia, and it is renowned for its hydrological patterns that exhibit a high level of unpredictability. The evaluation of the predictive precision and effectiveness of the Optimised Decision Tree ODT model, along with the RF and GBT models, in this study involved analysing several indicators. These indicators included the correlation coefficient, mean absolute error, percentage of relative error, root mean square error, Nash-Sutcliffe efficiency, and accuracy rate. The research results indicated that the ODT and RF models performed better than the GBT model in predicting monthly inflows. The ODT model, as well as the RF and GBT models, showed validation results with average accuracies of 94%, 91%, and 92%, respectively. The R² values were 90.2%, 84.8%, and 96.0%, respectively, and the NES values ranged from 0.92 to 0.94. The results of this research have greater implications, extending beyond the forecasting of monthly inflow rates to encompass other hydro-meteorological variables that depend exclusively on historical input data.

Psychological Predictors of Attitude toward Integrated Arts Education among Chinese College Students Majoring in the Arts.

This study investigated the psychological factors related to the attitudes of Chinese arts college students toward integrated arts education. It also examined predictive models incorporating psychological variables, demographic profiles, and art education-related characteristics to offer valuable insights for future research and art education practices. The sample comprised 303 Chinese college students majoring in arts and aged 18-22 years. The predictive models were examined using stepwise regression and decision tree analyses. The results indicated positive correlations between attitudes toward integrated arts education and several psychological variables, including extraversion, openness, conscientiousness, agreeableness, the Behavioral Inhibition System (BIS), hardiness, creativity, self-efficacy, and purpose orientation for personal growth. Neuroticism and the Behavioral Activation System (BAS) were negatively correlated with attitudes toward integrated arts education. Further, extraversion accounted for the greatest variance in attitudes toward integrated arts education. Extraversion, self-efficacy, purpose orientation for personal growth, BIS, and commitment accounted for approximately 38.3% of the variance. The decision tree model, predicting the attitudes of college students majoring in the arts toward integrated arts education, included extraversion, self-efficacy, teaching experience in their major, and academic year. This study contributes to a better understanding of the psychological and educational factors that shape the attitudes of Chinese arts students toward integrated arts education and provides a predictive framework that can inform future research and educational practices.

Assessing genotypic diversity by multivariate analysis and predicting useful selection ranges using decision tree in soybean (Glycine max L.)

The existence of sufficient genetic variability and knowing the suitable selection criterion ranges are indispensable requirements of breeding studies. This study aimed to explore genetic diversity in ten F2 populations (heterozygous–heterogeneous), five parents, and three commercial check cultivars by multivariate methods and to predict useful selection ranges using a decision tree model for further breeding studies. Hybridizations were made in 2019, F1 populations were grown, and F2 seeds were obtained in 2020. Genotypes were grown according to an Augmented Randomized Complete-Block Design in 2021 at Aydın Adnan Menderes University Faculty of Agriculture. Results revealed considerable genetic diversity and transgressive segregation in key characteristics across the F2 populations. Seed yields of F2 plants are higher than those of the parents. The seed protein and oil content of the F2 plants were not different from the parents and commercial checks. Higher phenotypic than genotypic variance indicated a high influence of environment on the inheritance of all studied characteristics. The highest broad heritability was observed in the number of shattered pods per plant (94.81 %) and the lowest in infertile pods per plant (2.16 %). Seed yield showed a significant and positive genotypic correlation with the number of fertile pods per plant (0.61**) and a significant negative association with the number of fertile branches per plant (−0.53**). Principal component analysis (PCA) of 14 quantitative characteristics showed that soybean genotypes separated by the first four components contributed 71.12 % of total genetic variation, with important yield-attributing characteristics present in PC1 and PC3, while important seed quality characteristics were noted in PC2. The decision tree model indicated that genotypes with high yield and optimum quality could be obtained by selecting the number of fertile pods 87–119 per plant and a hundred-seed weight greater than 14.55 g.