Decision Tree Algorithm Research Articles

Machine learning algorithm approach to complete blood count can be used as early predictor of COVID-19 outcome.

Although the SARS-CoV-2 infection has established risk groups, identifying biomarkers for disease outcomes is still crucial to stratify patient risk and enhance clinical management. Optimal efficacy of COVID-19 antiviral medications relies on early administration within the initial five days of symptoms, assisting high-risk patients in avoiding hospitalization and improving survival chances. The complete blood count can be an efficient and affordable option to find biomarkers that predict the COVID-19 prognosis due to infection-induced alterations in various blood parameters. This study aimed to associate hematological parameters with different COVID-19 clinical forms and utilize them as disease outcome predictors. We performed a complete blood count in blood samples from 297 individuals with COVID-19 from Belo Horizonte, Brazil. Statistical analysis, as well as ROC Curves and machine learning Decision Tree algorithms were used to identify correlations, and their accuracy, between blood parameters and disease severity. In the initial four days of infection, traditional hematological COVID-19 alterations, such as lymphopenia, were not yet apparent. However, the monocyte percentage and granulocyte-to-lymphocyte ratio proved to be reliable predictors for hospitalization, even in cases where patients exhibited mild symptoms that later progressed to hospitalization. Thus, our findings demonstrate that COVID-19 patients with monocyte percentages lower than 7.7% and a granulocyte-to-lymphocyte ratio higher than 8.75 are assigned to the hospitalized group with a precision of 86%. This suggests that these variables can serve as important biomarkers in predicting disease outcomes and could be used to differentiate patients at hospital admission for managing therapeutic interventions, including early antiviral administration. Moreover, they are simple parameters that can be useful in minimally equipped health care units.

Securing constrained IoT systems: A lightweight machine learning approach for anomaly detection and prevention

With the advent of advanced technological developments such as IoT, edge, and fog computing, cyber attacks have become increasingly sophisticated. IoT networks facilitate collaborative and intelligent tasks across various domains, including Industry 4.0, digital healthcare, and home automation. However, the proliferation of IoT devices has raised concerns about severe attacks, particularly those targeting resource constraints such as energy and memory. In response to these challenges, Tiny Machine Learning (TinyML) has emerged as a new research area, focusing on machine learning techniques tailored for embedded and IoT systems. This study proposes an ML detection mechanism designed to categorize and detect resource-constrained attacks in IoT devices. We consider IoT devices to be integral components within the continuum of edge and cloud computing, leveraging EdgeML and CloudML for detection purposes. Our paper conducts a comparative analysis of ML models, with a specific focus on energy consumption and memory usage in IoT applications. We compare various ML methodologies, including cloud-based, edge-based, and device-based strategies for both training and detection. The evaluation encompasses the application of these ML techniques to petite IoT devices, utilizing TinyML, as well as cloud and edge devices. Our findings reveal that the Decision Tree algorithm deployed on smart devices surpasses other approaches in terms of training efficiency, resource utilization, and the ability to detect resource-constrained attacks on IoT devices. We demonstrate a high level of accuracy, exceeding 96.9%, across all presented ML models in detecting resource constraint attacks within IoT systems. In summary, this research serves as a guide for implementing effective security measures in the dynamic landscape of IoT and associated technologies.

ADVANCEMENTS IN AIRLINE SECURITY: EVALUATING MACHINE LEARNING MODELS FOR THREAT DETECTION

This study assessed the performance of four machine learning algorithms—Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), and Neural Network (NN)—for predicting airline security threats using a dataset of 100,000 entries with 30 features. The models were evaluated based on accuracy, precision, recall, F1-Score, and AUC-ROC. The Neural Network achieved the highest performance, with an accuracy of 88%, precision of 86%, recall of 85%, F1-Score of 85.5%, and AUC-ROC of 0.90, demonstrating superior capability in capturing complex, non-linear patterns. The Random Forest model followed, with an accuracy of 85%, precision of 83%, recall of 82%, F1-Score of 82.5%, and AUC-ROC of 0.87, offering a robust and generalizable solution. The SVM model attained an accuracy of 81%, precision of 80%, recall of 78%, F1-Score of 79%, and AUC-ROC of 0.84, showing effective binary classification but with higher computational costs. The Decision Tree model, while interpretable, had the lowest performance metrics: accuracy of 78%, precision of 76%, recall of 72%, F1-Score of 74%, and AUC-ROC of 0.79. The results indicate that Neural Networks and Random Forests are the most effective models for airline security threat detection, with Neural Networks providing the highest overall accuracy and AUC-ROC.

Rainfall Prediction Using Random Forest and Decision Tree Algorithms

Weather

Enhancing freight train delay prediction with simulation‐assisted machine learning

AbstractBoosting the rail freight modal share is an ambitious target in Europe and North America. Yards, where freight trains are arranged, can be crucial in realizing this target by reliable dispatching to the network. This paper predicts freight train departures by developing a simulation‐assisted machine learning model with two concepts: general (adding all predictors at once) and step‐wise (adding predictors as they become available in sub‐yard operations) for hump yards with the conventional layout to provide a generalized model for European and North American contexts. The developed model is a decision tree algorithm, validated via 10‐fold cross‐validation. The model's performance on three data sets—a real‐world European yard, a baseline simulation, and an ultimate randomness simulation for a comparable North American yard—shows a respective of 0.90, 0.87, and 0.70. Step‐wise inclusion of the predictors results differently for the real‐world and simulation data. The global feature importance highlights maximum planned length, departure weekday, the number of arriving trains, and minimum arrival deviation as key predictors for the real‐world data. For the simulation data, the most significant predictors are departure yard predictors, the number of arriving trains, and the maximum hump duration. Additionally, utilization rates—except for the receiving yard—enhance the predictions.

Prediction of civil aircraft hard landing risk based on GBDT-GS method

Hard landing may cause symptom events such as structural damage to the aircraft, or even cause fatal flight accidents. In view of the lack of physical analysis in the current hard landing risk assessment, in order to effectively implement the hard landing risk identification and grade judgment criteria, so as to improve the quality of pilots' landing operations, a hard landing risk prediction model was constructed based on the gradient boosting decision tree algorithm (GBDT) and grid search (GS) combined with flight state analysis. Firstly, the five flight state parameters closely related to the hard landing were determined by analyzing the aircraft force and constructing the landing flight kinematic equation; then, the flight state data was extracted from the data recorded by the onboard Quick Access Recorder (QAR) to construct a data set, and according to the QAR parameter characteristics, the hard landing risk prediction model was constructed by the GBDT algorithm, and the model parameters were optimized by GS; finally, taking the "Chengdu-Shenyang" route of an airline as an example, 530 QAR data were selected to train and test the model, and the results were compared with those of random forest, logistic multivariate regression, recurrent neural network (RNN) and other algorithms. The results show that the performance of GBDT-GS algorithm in predicting hard landing risk is better than other algorithms, and the prediction accuracy reaches , which 92%，verifies the objective effectiveness of the model.

Predicting immune risk in treatment-naïve HIV patients using a machine learning algorithm: a decision tree algorithm based on micronutrients and inversion of the CD4/CD8 ratio.

Micronutrients have significant functional implications for the human immune response, and the quality of food is a major factor affecting the severity and mortality caused by HIV in individuals undergoing antiretroviral therapy. A decrease in CD4 lymphocyte count and an increase in CD8 lymphocyte count are the hallmarks of HIV infection, which causes the CD4/CD8 ratio to invert from a normal value of >1.6 to <1.0. In this study, we tried to analyze whether the nutritional status of HIV-positive patients has an impact on the CD4/CD8 ratio inversion by utilizing a machine learning (ML) algorithm. In this study, 55 confirmed HIV-positive patients who had not started their anti-retroviral therapy were included after obtaining their informed, written consent. Moreover, 55 age-and sex-matched relatives and caregivers of the patients who tested negative in the screening were enrolled as controls. All individual patient data points were analyzed for model development with an 80-20 train-test split. Four trace elements, zinc (Zn), phosphate (P), magnesium (Mg), and calcium (Ca), were utilized by implementing a random forest classifier. The target of the study was the inverted CD4/CD8 ratio. The data of 110 participants were included in the analysis. The algorithm thus generated had a sensitivity of 80% and a specificity of 83%, with a likelihood ratio (LR+) of 4.8 and LR-of 0.24. The utilization of the ML algorithm adds to the limited evidence that currently exists regarding the role of micronutrients, especially trace elements, in the causation of immune risk. Our inherent strength lies in the fact that this study is one of the first studies to utilize an ML-based decision tree algorithm to classify immune risk in HIV patients. Our study uniquely corroborated the nutritional data to the immune risk in treatment-naïve HIV patients through the utilization of a decision tree ML algorithm. This could subsequently be an important classification and prognostic tool in the hands of clinicians.

Estimation Stand Volume, Basal Area and Quadratic Mean Diameter Using Landsat 8 OLI and Sentinel‐2 Satellite Image With Different Machine Learning Techniques

ABSTRACTThe data required for sustainable forest planning is provided by traditional forest inventories, which are labor, time, and cost‐intensive. Providing this data quickly, reliably, and accurately is crucial for planners and researchers. The objective of this study was to predict stand basal area (BA), stand volume (V), and quadratic mean diameter (dq) by leveraging vegetation indices (VIs) and reflectance (R) derived from Landsat 8 OLI and Sentinel 2 satellite images, along with topographic (T) data obtained from ALOS‐PALSAR satellite imagery. Forest inventory data for a total of 250 sample plots were used for modeling in the study. Stand parameters were estimated using support vector machines (SVM), multiple linear regression (MLR), decision tree (DT), and random forest (RF) algorithms. In modeling V, BA, and dq, both individual and combinations of R, VIs, and T values obtained from satellite imagery were used as independent variables. Using the generated datasets, each of the stand parameters was modeled separately with MLR, SVM, RF, and DT algorithms, and the success of the models was compared to determine the modeling technique and dataset with the highest success for the relevant parameter. The results showed that for each stand parameter, the highest model success was achieved in the combined dataset, which was created by combining all datasets. However, in terms of modeling techniques, the highest success for each stand parameter was achieved with different modeling techniques. The highest success for V is obtained in the model using the SVM method (R2 = 0.78; RMSE = 0.28 m3/ha), the RF method yielded the highest model performance for BA (R2 = 0.70; RMSE = 2.53 m2/ha), and finally, the highest success for dq was obtained in the DT method (R2 = 0.74; RMSE = 0.02 cm). In general, the datasets obtained from Sentinel 2 images showed higher model success than the datasets obtained from Landsat 8 OLI images.

Concrete carbonation depth prediction model based on a gradient-boosting decision tree and different metaheuristic algorithms

Carbonation is a significant factor contributing to the instability of concrete structures. This study proposes two prediction models based on the gradient boosting decision tree (GBDT) to address the challenge of predicting concrete carbonation depth. This study integrates GBDT with two metaheuristic algorithms, particle swarm optimization algorithm (PSO) and sparrow search optimization algorithm (SSA), forming two hybrid models, PSO-GBDT and SSA-GBDT, aimed at enhancing prediction accuracy. Six influencing parameters (FA, t, w/b, B, RH, and CO2) were selected as input features to train and evaluate the hybrid models, with the concrete carbonation depth was used as the output. A database containing 883 groups of cases was established. Finally, three classic models, GBDT, ANN and SVR, were compared against the two hybrid models. To enhance model generalization, all models were subjected to five-fold cross validation. Four evaluation indicators (RMSE, R2, MAE, and VAF) and Taylor diagrams were employed to comprehensively assess these models. The results indicated that the two hybrid models exhibited superior prediction performance in the dataset (training set: R2, VAF>0.98, testing set: R2, VAF>0.96). The SSA-GBDT model achieved the highest prediction performance (RMSE= 2.7008, R2= 0.9639, MAE= 1.7691, VAF= 0.9627). Additionally, the SSA-GBDT model identified exposure time (t) as the feature with the greatest impact. This study demonstrates the applicability of the SSA-GBDT models in predicting concrete carbonation depth, offering a novel approach for improving accuracy in such predictions.

Predicting photosynthetic bacteria-derived protein synthesis from wastewater using machine learning and causal inference

Causal inference-assisted machine learning was used to predict photosynthetic bacterial (PSB) protein production capacity and identify key factors. The extreme gradient boosting algorithm effectively predicted protein content, while the gradient boosting decision tree algorithm excelled in predicting protein production, protein productivity, and protein energy yields. Driving factors were identified, with suitable ranges: protein content (pH 6.0–7.5, hydraulic retention time (HRT) < 3.8 d), protein production (biomass > 1.7 g, organic loading rate (OLR) > 9.2 gL–1d–1, temperature 26.7–35.0 °C), protein productivity (HRT < 3.5 d, biomass > 1.6 g, OLR > 10.0 gL–1d–1), and protein energy yields (light energy 0.1–4.4 kWh, biomass 1.7–65.0 g, chemical oxygen demand (COD) 0.1–2.5 gL–1). Illuminance, dissolved oxygen, COD, and COD/total nitrogen ratio were causal factors influencing protein production. Two-dimensional partial dependence plot revealed the interaction between two driving factors. This study enhances information on PSB protein production and offers insights for wastewater treatment and sustainable resource development.

Analysis of investment behavior among Filipinos: Integration of Social exchange theory (SET) and the Theory of planned behavior (TPB)

Despite the emergence of more accessible and modern forms of investment, the ever competitive and volatile market remains subject to anomalous irrationalities caused by investors. To this day, predicting their behavior remains difficult with lacking information, and poses a problem for investment platforms to effectively adjust to their predispositions. Therefore, this study aimed to comprehensively analyze the factors that have influenced investors’ behaviors using the integrated construct of the Social Exchange Theory and the Theory of Planned Behavior. With consideration of convenience sampling, a total number of 10,725 data points were collected and analyzed through machine learning algorithms of decision tree and neural network. Specifically, the comparison between long short-term memory (LSTM) and neural network, and random forest classifier and LightGBM were considered. It was found that the investor’s attitude, accessibility to financial services, and perceived economic benefits were the most influential predictors to their behavior, while six other factors also showed varying levels of significance. This study aimed to provide a unique framework which could be utilized by investment platforms to cater to the different behavioral factors expressed by investors. In line with these findings, it is recommended that platforms create flexible solutions that are based on their intentions and preferences, and more user-friendly through the implementation of new technologies. In addition, they are suggested to appeal to novice investors by reducing the burden of costs, promising future benefits, and promoting financial education. The results of this study proved the reliability of the integrated model as a social and behavioral framework, and consequently, LSTM overpowering other tools on accurate forecast made, followed by neural network, and random forest.

Using Decision Tree Technique to Predict the Financial Collapse of Iraqi Private Banks: An Applied Study

Predicting the Financial Collapse (FC) using modern technologies and scientific methods for Iraqi private banks is considered an early warning to detect their failure in the future. Therefore, these proposals, represented by scientific methods and modern technologies, are submitted to the specialized bodies represented by (the Central Bank of Iraq) in order to present them to the banks in the research sample to raise their capacity through the use of modern technologies and in order to adopt them as a scientific basis for predicting the FC. The following will be followed: Evaluating the performance of banks and classifying them based on the Camels standard (CS) and predicting the FC based on modern technologies (by relying on the decision tree algorithm). The research relied on the deductive approach to complete the theoretical aspect and the applied (experimental) approach to complete the practical aspect. The researcher also processed and prepared the data and then evaluated the performance of banks and classified them based on the CS equations consisting of five indicators and on the averages of these results as inputs for prediction using DT. Different computer programs were also adopted, as the Excel program was used to unload the data, the (SPSS) program to apply the equations, and the (Matlab) program to predict modern technologies. The most prominent conclusions reached by the research are as follows: Modern technologies represented by decision trees are characterized by providing more accurate information

Agricultural greenhouse extraction using multi-temporal spectral feature with decision tree algorithms in Jilin Province, China

Agricultural greenhouse extraction using multi-temporal spectral feature with decision tree algorithms in Jilin Province, China

Application of the Decision Tree Method for Differentiating Human Groups

One of the tasks of modern biological anthropology is to develop a system that could objectively classify humanity on the basis of measurements. Here, the decision tree algorithm was chosen to create a classification of groups. The method helps to evaluate the differentiating power of specific dimensions for separating samples and to assess the composition of clusters at each step of the analysis. Standard cranial measurements were used, and the entropy index was chosen as a heterogeneity measure. Classification units were 39 ethno-territorial groups from 13 major regions of the Old World. At the first step, differentiation is made between broad-faced and narrow-faced groups, demonstrating the classificatory value of this trait. The first cluster includes only Mongoloids, admixed Southern Siberian populations, and Ainu. The second cluster is heterogeneous, but its further subdivision is more in line with the traditional classification. Traits underlying the branching of the tree may be the same in different branches, evidencing their taxonomic importance. Capabilities of the decision tree method proved sufficient to construct a system largely similar to the traditional one. Certain traits separate large groups of populations, while others are efficient at the regional level. The method, therefore, can be recommended as a supplementary tool at the intraspecific level.

Potholes Detection in Doppler Radar Signal’s Power Spectral Density using Decision Tree

Potholes are defects on the surface of roads, streets, or pavements brought on by depressions or holes, which are hazardous for vehicles and pedestrians, whether small divots or huge craters. Various methods have been explored to improve the accuracy of potholes detection. Existing approaches have advantages and disadvantages. This paper presents the proposed method of pothole detection utilising Doppler Radar signal's Power Spectrum Density (PSD) together with the Decision Tree classification algorithm. While continuous waveform (CW) radar is able to identify moving targets, it cannot localise the exact depth of the reflector, which is the prominent characteristic of potholes. In addition, the target's reflected signal is likely to be masked by nearby harmonics. Since the radar is moving despite the target of interest, mounting it on a moving vehicle offers a different perspective. This paper explores the potential of Doppler radar's signal for pothole detection while comparing two Machine Learning (ML) techniques. A commercially over-the-shelf (COTS) K-LC2 Doppler radar was employed to acquire pothole and non-pothole raw datasets. Doppler signal was hardly distinguished between pothole and non-pothole, either in the time or frequency domain. Hence, Doppler signals were converted to power spectral density (PSD), and PSD's features were extracted. Extracted features were applied with the coarse Decision Tree (DT) and K-Nearest Neighbours (KNN) classification algorithms. The result exhibits a better accuracy of 91.2% for 80:20 distribution by using the Decision Tree.

Gender-Based Analysis of Online Shopping Patterns on Shopee in Malaysia: A J48 Decision Tree Approach

The purpose of this study is to investigates the gender differences of Shopee platform for online shopping behavior by using the J48 decision tree algorithm to classify and predict shopping frequency among male and female consumers for Malaysia context. WEKA software was used in this study to analyze the datasets. From the experiments, the majority of Shopee user were female consumers. The findings shows that female consumer behavior is more complicated and more varied regarding purchasing behavior. The study's findings demonstrate the potential of gender specific insights to enhance e-commerce strategies, particularly in product recommendations and targeted marketing. Although the J48 model performed well in predicting male shopping patterns, it was less effective for females, indicating the need for more advanced modeling techniques is used to better capture the complexities of female consumer behavior. This research also emphasizes the significance of using machine learning tools like the J48 decision tree to analyze consumer data, providing valuable insights for improving customer satisfaction and business performance. However, limitations such as sample size and the focus on a single platform suggest that further research is needed, including the exploration of alternative algorithms and broader demographic factors.

9220 Machine Learning Methods In Differential Diagnosis Of Genetically Confirmed Multiple Endocrine Neoplasia Type 1 And Its Phenocopies

Abstract Disclosure: D. Trukhina: None. K. Voronov: None. E. Mamedova: None. A. Solodovnikov: None. Z. Belaya: None. Background. MEN-1 is a rare autosomal dominant disorder caused by mutations in the MEN1 gene encoding the menin protein (gMEN1). This syndrome is characterised by the occurrence of parathyroid tumours, gastroenteropancreatic neuroendocrine tumours, pituitary adenomas, and other endocrine and non-endocrine neoplasms. If a patient with the MEN-1 phenotype does not have any mutation in the MEN1 gene, the condition is considered a phenocopy of the syndrome (phMEN1). The goal of this study was to develop a machine learning algorithm to estimate the probability of gMEN-1 vs phMEN-1 based on easily available clinical features at first line differential diagnostics. Materials and methods. A single-center, one-stage, cohort, retrospective study was carried out. Patients were randomly stratified into 2 cohorts: training (80%) and test (20%). Eight machine learning algorithms were used to develop predictive models: Logistic Regression, k-nearest neighbors (kNN), Naive Bayes, binary decision tree (CART), C5.0 decision tree algorithms, Bagged CART, Random Forest, Gradient Boosting (Stochastic Gradient Boosting, GBM). Results. The study included 95 patients with genetically confirmed gMEN-1 syndrome and 60 patients with its phenocopies. As a result of preliminary data processing and selection for the most informative features, the final variables for the classification and prediction of gMEN-1 syndrome were identified: the number of affected parathyroid glands, age at diagnostics, pancreatic tumour existence, heredity, type of pituitary adenoma secretion, and gender. The kNN algorithm demonstrated the best diagnostic performance in a training sample (ROC-AUC - 0.96, sensitivity - 84%, specificity - 100%), which was confirmed in a test sample (ROC-AUC - 1.0; sensitivity - 94.4%, specificity - 100%) to determine genetically confirmed MEN-1 syndrome. Conclusion. The k-nearest neighbours (kNN) algorithm may be used as a first-line differential diagnostics method to select patients to be refered for genetic testing for gMEN-1. Presentation: 6/1/2024

7845 Machine Learning Methods In Differential Diagnosis Of Genetically Confirmed Multiple Endocrine Neoplasia Type 1 And Its Phenocopies

Abstract Disclosure: D. Trukhina: None. K. Voronov: None. E. Mamedova: None. A. Solodovnikov: None. Z. Belaya: None. Background. MEN-1 is a rare autosomal dominant disorder caused by mutations in the MEN1 gene encoding the menin protein (gMEN1). This syndrome is characterised by the occurrence of parathyroid tumours, gastroenteropancreatic neuroendocrine tumours, pituitary adenomas, and other endocrine and non-endocrine neoplasms. If a patient with the MEN-1 phenotype does not have any mutation in the MEN1 gene, the condition is considered a phenocopy of the syndrome (phMEN1).The goal of this study was to develop a machine learning algorithm to estimate the probability of gMEN-1 vs phMEN-1 based on easily available clinical features at first line differential diagnostics. Materials and methods. A single-center, one-stage, cohort, retrospective study was carried out. Patients were randomly stratified into 2 cohorts: training (80%) and test (20%). Eight machine learning algorithms were used to develop predictive models: Logistic Regression, k-nearest neighbors (kNN), Naive Bayes, binary decision tree (CART), C5.0 decision tree algorithms, Bagged CART, Random Forest, Gradient Boosting (Stochastic Gradient Boosting, GBM). Results. The study included 95 patients with genetically confirmed gMEN-1 syndrome and 60 patients with its phenocopies. As a result of preliminary data processing and selection for the most informative features, the final variables for the classification and prediction of gMEN-1 syndrome were identified: the number of affected parathyroid glands, age at diagnostics, pancreatic tumour existence, heredity, type of pituitary adenoma secretion, and gender. The kNN algorithm demonstrated the best diagnostic performance in a training sample (ROC-AUC - 0.96, sensitivity - 84%, specificity - 100%), which was confirmed in a test sample (ROC-AUC - 1.0; sensitivity - 94.4%, specificity - 100%) to determine genetically confirmed MEN-1 syndrome. Conclusion. The k-nearest neighbours (kNN) algorithm may be used as a first-line differential diagnostics method to select patients to be refered for genetic testing for gMEN-1. Presentation: 6/1/2024

Psychological Analysis and Psychological Training of Piano Teaching Based on the Analysis of Emotional Characteristics

Modern piano teaching theory attempts to explore the basic laws of modern piano teaching by using emotional feature recognition algorithm from the emotional level and in the analysis of emotional characteristics, deeply reveal the important value of emotion in piano teaching, and effectively improve the quality of piano teaching, which is also what this paper tries to clarify. This paper studies the psychology of piano teaching based on the emotional feature recognition method. In the proportion of piano teaching, when the number of experiments is 6, the average proportion of machine learning algorithm is 56.5%, the average proportion of decision tree algorithm is 55.1%, the average proportion of data mining algorithm is 58.7%, and the average proportion of this method is 61.8%. It can be seen that the proportion of this method in piano teaching is the highest. This method is used to cultivate students' learning enthusiasm, constantly overcome all kinds of difficulties, change restlessness, irritability and other bad mentality, so as to achieve better learning effect.

B-123 Prediction of Opioid Use Duration in Post-Surgical Orthopedic Patients Using CYP2D6 Inhibitor Index and Routine Laboratory Markers

Abstract Background Hydrocodone, the most commonly prescribed opioid in the U.S., is typically used for pain management following surgery and is a known potential drug of abuse with non-negligible risk of addiction. Significant variability in patients’ response to hydrocodone therapy have been reported, owing to both iatrogenic and endogenous sources. The objective of this study is to evaluate the feasibility of predicting post-discharge hydrocodone use in patients who have undergone orthopedic surgery using patient demographics, genotyping, concurrently prescribed medications, and routinely collected laboratory test results. Methods Targeted genotyping was performed on each patient with genes known to affect hydrocodone pharmacokinetics or pharmacodynamics, including CYP2D6, CYP3A4, CYP3A5, CYP2B6, COMT, OPRM1, and ABCB1. An inhibition index for CYP2D6 was calculated by ranking concurrently prescribed medications as weak, moderate, or strong inhibitors of enzyme activity. Pain index, average hydrocodone dose per day during hospital stay, and CYP2D6 estimated activity scores based on allele functionality were included as predictors. In addition, a retrospective chart review of electronic hospital records was performed where basic metabolic panel and complete blood count results were screened within 10 days of each patient's surgery date. Two types of classification and decision tree algorithms were tuned and validated as prediction models using 5-fold cross validation, including a ‘Fast and Frugal Decision Tree’ (FFTree) and an extreme gradient boosting machine (xgBoost). Clinical predictive features were ranked according to the Shapley Additive Explanations (SHAP) values using the xgBoost model. Results Out of 79 total patients with hydrocodone use duration information, 25 were categorized as “Short” and 54 categorized as “Long” using a 1-week threshold. FFTree model selected the top three features of CYP2D6 inhibition index, blood sodium level, and blood creatinine level and obtained a sensitivity and specificity of 0.80 and 0.76, respectively. The xgBoost model, using the same three features, achieved a sensitivity and specificity of 0.872 and 0.628, respectively, an AUROC of 0.814, PRAUC of 0.912, net benefit of 0.359, and was well-calibrated with a Brier score of 0.161. Consistent with previously published results, the predictive capacity of all genotypes analyzed were unremarkable and consistently ranked as the least important features by the xgBoost model SHAP values, due to their low predictive value. Conclusions Both the FFTree and xgBoost models were able to adequately classify the patients according to a 1-week stratification criteria with good to excellent classification performance. Patients with larger CYP2D6 inhibition index were more likely to be classified as “Long” duration. Sodium and creatinine level, which may reflect hydration status and renal function, and therefore hydromorphone elimination rate, were key predictors of post-discharge hydrocodone use duration when combined with the CYP2D6 inhibition index. Further refinement of the proposed models and validation of model robustness and stability using external cohorts is necessary. These results may have implications for providers when prescribing hydrocodone for pain management, where the ability to risk-stratify patients according to their predicted probability of developing opioid use disorder is of clinical importance.