Application Of Decision Tree Research Articles

Comparison of Extremely Randomized Survival Trees and Random Survival Forests: A Simulation Study

Abstract. Purpose: This simulation study investigates the Extremely Randomized Survival Trees (EST) model, a machine learning technique expected to handle survival analysis, particularly in large survival datasets, effectively. The study compares the performance of the EST model with that of the Random Survival Forest (RSF) model, focusing on the C-index value to determine which model performs better. Methods: The analysis begins with the generation of 540 simulated datasets, created by combining three levels of sample sizes, two levels of censoring proportions, three types of hazard functions, and 30 repetitions for each scenario. The simulation data were split into 80% training and 20% testing data. The training data were used to build the EST and RSF models, while the test data were used to evaluate their performance. The model with the highest C-index value was deemed the best performer, as a higher C-index indicates superior model performance. Result: The results indicate that the sample size, type of hazard function, and the method used influence that model performance. The EST model significantly outperformed the RSF model when the sample size was large, though no significant difference was observed when the sample size was small or medium. Additionally, the EST model consistently demonstrated faster computation times across all simulation scenarios. Novelty: This study provides a pioneering exploration into applying decision tree algorithms, specifically EST and RSF, in survival analysis. While these methods have been extensively studied in regression and classification contexts, their application in survival analysis remains relatively unexplored.

Identifying Sustainability Efforts in Company’s Reports Using Text Mining and Machine Learning

This study delves into the utilization of text mining to scrutinize social and environmental reports of companies, showcasing its effectiveness in evaluation. It explores various text mining techniques and practically applies decision tree, k-nearest neighbors, and naïve Bayes methods. The paper offers guidance on extracting pertinent terms related to four CSR dimensions: Environment, Employee, Social responsibility, and Human rights. Results demonstrate the successful differentiation of text based on these dimensions, leveraging a CSR-relevant dictionary by Pencel and Malascue. Employing document classification techniques, the study constructs four models using distinct text mining approaches for comparative analysis. Through this research, the valuable role of text mining in assessing social and environmental disclosures is underscored, providing insights into optimizing these techniques for evaluations and emphasizing their potential to enhance understanding and decision-making in corporate social responsibility assessments. Keywords: sustainability, text mining, machine learning, Corporate Social Responsibility - CSR, environmental reports

Taekwondo win-loss determining factors using data mining-based decision tree analysis: focusing on game analysis for evidence-based coaching

BackgroundPurpose In this study, the purpose of this study is to identify the determinants of winning and losing in taekwondo by applying decision tree analysis, one of the data mining techniques, based on the 2022 World taekwondo championships women’s competition.Methods272 women’s games in the taekwondo championships in Guadalajara held by the WT in 2022 were used. For data processing, an independent sample t-test was performed for differences in game content variables according to the win/lose group, and a decision tree analysis was performed to confirm game content variables affecting the win/lose group. To check the predictive power of the model, classification accuracy, standard error, and misclassification estimates were calculated. All statistical significance levels were set at 0.05.ResultsFirst, it was found that there was no statistically significant difference only in body attack (attempt) and number of kicking variables according to the winning and losing groups(p > .05), and there were differences in all other game content variables(p < .05). Second, as a result of conducting a decision tree analysis to confirm the determinants of winning and losing in taekwondo sparring, winning situation, tie situation, and number of kicks were identified as important variables.ConclusionThe World taekwondo championships are analyzed in the currently changed taekwondo competition rules to identify important factors, and at the same time, based on this, data-based coaching is expected to improve performance.

A Machine-Learning-Assisted Classification Algorithm for the Detection of Archaeological Proxies (Cropmarks) Based on Reflectance Signatures

The detection of subsurface archaeological remains using a range of remote sensing methods poses several challenges. Recent studies regarding the detection of archaeological proxies like those of cropmarks highlight the complexity of the phenomenon. In this work, we present three different methods, and associated indices, for identifying stressed reflectance signatures indicating buried archaeological remains, based on a dataset of measured ground spectroradiometric reflectance. Several spectral profiles between the visible and near-infrared parts of the spectrum were taken in a controlled environment in Cyprus during 2011–2012 and are re-used in this study. The first two (spectral) methods are based on a suitable analysis of the spectral signatures in (1) the visible part of the spectrum, in particular in the neighborhood of 570 nm, and (2) the red edge part of the spectrum, in the neighborhood of 730 nm. Machine learning (decision trees) allows for the deduction of suitable wavelengths to focus on in order to formulate the proposed indices and the associated classification criteria (decision boundaries) that can enhance the detection probability of stressed vegetation. Noise in the signal is taken into account by simulating reflectance signatures perturbed by white noise. Applying decision tree classification on the ensemble of simulations and basic statistical analysis, we refine the formulation of the indices and criteria for the noisy signatures. The success rate of the proposed methods is over 90%. The third method rests on the estimation of vegetation/canopy reflectance parameters through inversion of the physical-based PROSAIL reflectance model and the associated classification through machine learning methods. The obtained results provide further insights into the formation of stress vegetation that occurred due to the presence of shallow buried archaeological remains, which are well aligned with physical-based models and existing empirical knowledge. To the best of the authors’ knowledge, this is the first study demonstrating the usefulness of radiative transfer models such as PROSAIL for understanding the formation of cropmarks. Similar studies can support future research directions towards the development of regional remote sensing methods and algorithms if systematic observations are adequately dispersed in space and time.

Predicting Barrier Island Shrub Presence Using Remote Sensing Products and Machine Learning Techniques

AbstractBarrier islands are highly dynamic coastal landforms that are economically, ecologically, and societally important. Woody vegetation located within barrier island interiors can alter patterns of overwash, leading to periods of periodic‐barrier island retreat. Due to the interplay between island interior vegetation and patterns of barrier island migration, it is critical to better understand the factors controlling the presence of woody vegetation on barrier islands. To provide new insight into this topic, we use remote sensing data collected by LiDAR, LANDSAT, and aerial photography to measure shrub presence, coastal dune metrics, and island characteristics (e.g., beach width, island width) for an undeveloped mixed‐energy barrier island system in Virginia along the US mid‐Atlantic coast. We apply decision tree and random forest machine learning methods to identify new empirical relationships between island geomorphology and shrub presence. We find that shrubs are highly likely (90% likelihood) to be present in areas where dune elevations are above ∼1.9 m and island interior widths are greater than ∼160 m and that shrubs are unlikely (10% likelihood) to be present in areas where island interior widths are less than ∼160 m regardless of dune elevation. Our machine learning predictions are 90% accurate for the Virginia Barrier Islands, with almost half of our incorrect predictions (5% of total transects) being attributable to system hysteresis; shrubs require time to adapt to changing conditions and therefore their growth and removal lags changes in island geomorphology, which can occur more rapidly.

Research on the Application of Decision Tree and Correlation Analysis Algorithm in College Students’ Physical Fitness Analysis

With the advent of the big data era, data-driven decision-making and analysis are increasingly valued in various fields. Especially in the field of education, how to use big data technology to better understand student needs, optimize the education process, and improve education quality has become an important research topic. This paper will explore the application of decision trees and related analysis algorithms in the analysis of college students’ physical fitness, in order to provide scientific basis for improving the physical health level of college students. This paper studies the application of DT (decision tree) and correlation analysis algorithm in the analysis of college students’ physical fitness. In this paper, the method of big data and DM (data mining) is proposed to extract the rules contained in the data information, so as to directly provide auxiliary decision-making for physical fitness test and analysis. The research results show that through training the training set, a good classification accuracy rate is achieved, and through optimizing the depth, the accuracy rate can reach more than 85.033%. Using DM technology as a carrier, this paper digs into the rules behind the new knowledge of college students’ physical fitness, and digs out the previously unknown, implied and potentially useful information and knowledge.

Physics-assisted machine learning for THz time-domain spectroscopy: sensing leaf wetness

Signal processing techniques are of vital importance to bring THz spectroscopy to a maturity level to reach practical applications. In this work, we illustrate the use of machine learning techniques for THz time-domain spectroscopy assisted by domain knowledge based on light–matter interactions. We aim at the potential agriculture application to determine the amount of free water on plant leaves, so-called leaf wetness. This quantity is important for understanding and predicting plant diseases that need leaf wetness for disease development. The overall transmission of 12,000 distinct water droplet patterns on a plastized leaf was experimentally acquired using THz time-domain spectroscopy. We report on key insights of applying decision trees and convolutional neural networks to the data using physics-motivated choices. Eventually, we discuss the generalizability of these models to determine leaf wetness after testing them on cases with increasing deviations from the training set.

Using asymptotic homogenization and machine learning to derive a 3D trace theory for GFRP

Tsai’s modulus was proposed as a trace theory for CFRP, where only one property is measured experimentally and the other four properties are computed using the normalized trace relation. An extension of this theory is proposed for GFRP. Laminae elastic properties are generated using the asymptotic homogenization and a machine learning training is performed applying decision trees algorithm to define normalized trace relations for GFRP. The results are validated with experimental data of 12 GFRP laminae, indicating average errors around 5% for longitudinal and transversal elastic moduli and in-plane Poisson’s ratio, and around 10% for in-plane and out-of-plane shear moduli.

The Application of Immersive Virtual Reality and Machine Learning for the Assessment of Unilateral Spatial Neglect

Abstract Even subtle forms of hemispatial neglect after stroke negatively affect the performance of daily life tasks, increase the risk of injury, and are associated with poor rehabilitation outcomes. Conventional paper-and-pencil tests, however, often underestimate the symptoms. We aimed to identify relevant neglect-specific measures and clinical decision rules based on machine learning techniques on behavioral data generated in a new Virtual Reality (VR) application, the immersive virtual road-crossing task. In total, 59 participants were included in our study: two right-hemispheric stroke groups with left neglect (N = 20) or no neglect (N = 19), classified based on conventional tests and medical diagnosis, and healthy controls (N = 20). A neuropsychological test battery and the VR task were administered to all participants. We applied decision trees and random forest models to predict the respective groups based on the results of the VR task. Our feature selection procedure yielded six features as suitable predictors, most of which involved lateral time-related measures, particularly reaction times, and head movements. Our model achieved a high training accuracy of 96.6% and estimated test accuracy of 76.8%. These results confirm previous reports that temporal behavioral patterns are key to detecting subtle neglect in patients with chronic stroke. Our results indicate that VR combined with machine learning has the potential to achieve higher test accuracies while being highly applicable to clinical practice.

Applying decision tree algorithms to early differential diagnosis between different clinical forms of acute Lyme borreliosis and tick-borne encephalitis

Background: Tick-borne encephalitis and Lyme borreliosis are the most common natural focal infections in Russia often arising as a mixed infection, which is often clinically difficult to distinguish from a monoinfection at the onset of the disease that is a result of delayed laboratory verification of the diagnosis and it requires further searching for fundamentally new approaches to the issue of early differential diagnosis of tick-borne infections.Aims: is to develop decision tree algorithms for early differential diagnosis between the mono- and mixed forms of acute Lyme borreliosis and tick-borne encephalitis with prevailing febrile syndrome in clinical picture based on clinical and laboratory data.Materials and methods: We retrospectively analyzed 55 clinical and laboratory parameters obtained from 291 hospitalized tick-borne infections patients with or without erythema migrans at the site of ixodid tick bites in the first week of the disease, who were included in the single-center study from 2010 to 2023. In 211 patients without erythema, the analysis was carried out between three classes depending on the diagnosis: the mixed infection of non-erythematous Lyme borreliosis and tick-borne encephalitis, the mono-infection of non-erythematous Lyme borreliosis or the monoinfection of tick-borne encephalitis. The other two classes, which included 80 patients with erythema, had the mixed infection of acute erythematous Lyme borreliosis and tick-borne encephalitis or erythematous Lyme borreliosis monoinfection. Python programming language was applied to develop two decision tree models. Feature importance was assessed for all predictors. Each patient class was randomly divided into training (70%) and testing (30%) datasets. Accuracy evaluation of the models was based on ROC analysis.Results: The decision tree algorithm for early differential diagnosis among the tick-borne infection patients without erythema migrans included the following most important predictors: maximal fever rise, chills, neutrophil-to-monocyte ratio, ESR, absolute number of reactive lymphocytes and immature granulocytes, and percentage of eosinophils. The model for differential diagnosis between the patients with erythema migrans included the following predictors: maximal fever rise, the absolute number of reactive lymphocytes and immature granulocytes, and the percentage of basophils. Both decision tree models showed excellent predictive values based on sensitivity, specificity, precision, accuracy, and F1 scores, as well as areas under the ROC curve, which were higher than 0.90.Conclusions: Based on clinical and laboratory parameters, two decision tree models with high sensitivity have been developed, which can be easily applied in clinical practice for early differential diagnosis of the tick-borne infections with prevailing fever syndrome.

Analysing the determinants of the Turkish household electricity consumption using gradient boosting regression tree

Residential buildings are the second largest electricity consumer in Turkey. Thus, the goal here is to detect the factors that determine the electricity consumption of the households in Turkey using the Household Budget Survey (HBS). This study applies Decision Tree (DT), Random Forest (RF) and Gradient Boosted Regression Tree (GBRT) methods. Since the GBRT method provides the lowest Root Mean Squared Error (RMSE), the impact of each variable on the electricity consumption is analysed with this method. The most critical determinant is found to be the household size, while income level and heating type are discovered as 2nd and 3rd most prominent determinants for household electricity demand. With the help of the Partial Dependence Plots (PDP) provided by the GBRT method, the impact of each categorical and continuous variable is presented. Using the results of PDPs, the monetary values of both electricity generation and the social cost of CO2 emissions emitted into the atmosphere due to electricity generation are calculated for the most important determinants.

Optimasi Decision Tree Menggunakan Teknik Boosting pada Prediksi Penyakit Diabetes

Early detection of diabetes is very important to reduce the consequences caused by the disease. Diabetes is influenced by many factors, so to make a diagnosis requires a complex analysis. The dataset used to analyze the prediction of diabetes is using machine learning algorithm. The machine learning algorithm is used to classify someone with diabetes or not based on the factors that have been set as input. The results of the diagnosis/prediction that are not perfect are caused by many misclassifications. To reduce classification errors, it is proposed to apply decision tree and boosting techniques. The classification algorithm used in this study is Random Forest. The experimental results show that decision tree and boosting techniques and a combination of the two can reduce misclassification in diabetes prediction.

Detection of offensive terms in resource-poor language using machine learning algorithms.

The use of offensive terms in user-generated content on different social media platforms is one of the major concerns for these platforms. The offensive terms have a negative impact on individuals, which may lead towards the degradation of societal and civilized manners. The immense amount of content generated at a higher speed makes it humanly impossible to categorise and detect offensive terms. Besides, it is an open challenge for natural language processing (NLP) to detect such terminologies automatically. Substantial efforts are made for high-resource languages such as English. However, it becomes more challenging when dealing with resource-poor languages such as Urdu. Because of the lack of standard datasets and pre-processing tools for automatic offensive terms detection. This paper introduces a combinatorial pre-processing approach in developing a classification model for cross-platform (Twitter and YouTube) use. The approach uses datasets from two different platforms (Twitter and YouTube) the training and testing the model, which is trained to apply decision tree, random forest and naive Bayes algorithms. The proposed combinatorial pre-processing approach is applied to check how machine learning models behave with different combinations of standard pre-processing techniques for low-resource language in the cross-platform setting. The experimental results represent the effectiveness of the machine learning model over different subsets of traditional pre-processing approaches in building a classification model for automatic offensive terms detection for a low resource language, i.e., Urdu, in the cross-platform scenario. In the experiments, when dataset D1 is used for training and D2 is applied for testing, the pre-processing approach named Stopword removal produced better results with an accuracy of 83.27%. Whilst, in this case, when dataset D2 is used for training and D1 is applied for testing, stopword removal and punctuation removal were observed as a better preprocessing approach with an accuracy of 74.54%. The combinatorial approach proposed in this paper outperformed the benchmark for the considered datasets using classical as well as ensemble machine learning with an accuracy of 82.9% and 97.2% for dataset D1 and D2, respectively.

Applying Decision Tree Algorithm Classification and Regression Tree (CART) Algorithm to Gini Techniques Binary Splits

Decision tree study is a predictive modelling tool that is used over many grounds. It is constructed through an algorithmic technique that is divided the dataset in different methods created on varied conditions. Decisions trees are the extreme dominant algorithms that drop under the set of supervised algorithms. However, Decision Trees appearance modest and natural, there is nothing identical modest near how the algorithm drives nearby the procedure determining on splits and how tree snipping happens. The initial object to appreciate in Decision Trees is that it splits the analyst field, i.e., the objective parameter into diverse subsets which are comparatively more similar from the viewpoint of the objective parameter. Gini index is the name of the level task that has applied to assess the binary changes in the dataset and worked with the definite object variable “Success” or “Failure”. Split creation is basically covering the dataset values. Decision trees monitor a top-down, greedy method that has recognized as recursive binary splitting. It has statistics for 15 statistics facts of scholar statistics on pass or fails an online Machine Learning exam. Decision trees are in the class of supervised machine learning. It has been commonly applied as it has informal implement, interpreted certainly, derived to quantitative, qualitative, nonstop, and binary splits, and provided consistent outcomes. The CART tree has regression technique applied to expected standards of nonstop variables. CART regression trees are an actual informal technique of understanding outcomes.

Measuring dynamic inefficiency through machine learning techniques

This paper contributes by developing new models for assessing dynamic inefficiency that incorporate machine learning techniques. In particular, the new approaches apply decision trees models for the estimation of dynamic production technologies that account for investment adjustment costs. Methodologically, the new models build on the recently developed techniques of Efficiency Analysis Trees (EAT) and Convexified Efficiency Analysis Trees (CEAT) and extend them even further to a dynamic framework comprising dynamic EAT and CEAT models. The study compares dynamic inefficiency scores estimated assuming the new models against the traditional dynamic free disposal hull (FDH) and dynamic data envelopment analysis (DEA). Our empirical application focuses on dairy manufacturing firms in the main dairy processing countries in the European Union for the years 2014 and 2018. The results show that inefficiency related to the dynamic CEAT or EAT is higher than their corresponding values calculated through the dynamic DEA or FDH. The discriminating power of dynamic DEA (dynamic FDH) drastically improves when switching to dynamic CEAT (dynamic EAT). Finally, the differences between countries are observed regarding the development of dynamic inefficiency in the period associated with milk quota abolition.

Applying decision tree to pseudo-observations for mixture cure models

Applying decision tree to pseudo-observations for mixture cure models

Decision Tree-Based Ensemble Model for Predicting National Greenhouse Gas Emissions in Saudi Arabia

Greenhouse gas (GHG) emissions must be precisely estimated in order to predict climate change and achieve environmental sustainability in a country. GHG emissions are estimated using empirical models, but this is difficult since it requires a wide variety of data and specific national or regional parameters. In contrast, artificial intelligence (AI)-based methods for estimating GHG emissions are gaining popularity. While progress is evident in this field abroad, the application of an AI model to predict greenhouse gas emissions in Saudi Arabia is in its early stages. This study applied decision trees (DT) and their ensembles to model national GHG emissions. Three AI models, namely bagged decision tree, boosted decision tree, and gradient boosted decision tree, were investigated. Results of the DT models were compared with the feed forward neural network model. In this study, population, energy consumption, gross domestic product (GDP), urbanization, per capita income (PCI), foreign direct investment (FDI), and GHG emission information from 1970 to 2021 were used to construct a suitable dataset to train and validate the model. The developed model was used to predict Saudi Arabia’s national GHG emissions up to the year 2040. The results indicated that the bagged decision tree has the highest coefficient of determination (R2) performance on the testing dataset, with a value of 0.90. The same method also has the lowest root mean square error (0.84 GtCO2e) and mean absolute percentage error (0.29 GtCO2e), suggesting that it exhibited the best performance. The model predicted that GHG emissions in 2040 will range between 852 and 867 million tons of CO2 equivalent. In addition, Shapley analysis showed that the importance of input parameters can be ranked as urbanization rate, GDP, PCI, energy consumption, population, and FDI. The findings of this study will aid decision makers in understanding the complex relationships between the numerous drivers and the significance of diverse socioeconomic factors in defining national GHG inventories. The findings will enhance the tracking of national GHG emissions and facilitate the concentration of appropriate activities to mitigate climate change.

Research on Civil Engineering Cost Prediction Based on Decision Tree Algorithm

Civil engineering includes not only all engineering construction on the ground, but also the maintenance, exploration and design of equipment and materials used in the whole construction process. During the construction of civil engineering, there is no scientific and reasonable supervision system for the supervision of construction, which greatly affects the quality of engineering and the control and management of engineering cost. In this paper, through the application of DT (Decision tree) algorithm, the research of civil engineering cost prediction is carried out. Aiming at the application of the algorithm in civil engineering cost management, this paper tries to improve the C4.5 algorithm. DT civil engineering life-cycle cost analysis and prediction model is trained by training set, and the optimal result of civil engineering life-cycle cost analysis is obtained by inputting sample data for model prediction. The application results show that the above algorithm has lower computational complexity and higher prediction efficiency in civil engineering cost prediction, which can better meet the needs of actual civil engineering cost analysis and prediction.

Applying decision trees on road traffic accident data for predicting the survival chance of patients

Applying decision trees on road traffic accident data for predicting the survival chance of patients

Applying Decision Trees on Road Traffic Accident Data for Predicting Survival Chance of Patients

Applying Decision Trees on Road Traffic Accident Data for Predicting Survival Chance of Patients