Binary Decision Tree Research Articles

EXPLORING VITAMIN D-RELATED KAP INFLUENCING FACTORS IN MIDDLE-AGED AND ELDERLY PATIENTS WITH ANKLE FRACTURES: A LOGISTIC REGRESSION AND DECISION TREE MODEL-BASED STUDY

This study aimed to analyze the factors influencing vitamin D (VD)-related knowledge, attitude, and practice (KAP) in middle-aged and elderly patients with ankle fractures using the binary logistic regression and decision tree models, thereby providing scientific evidence for VD-related recovery strategies. A Chinese version of the nutrition KAP questionnaire was administered in a Grade III A Hospital using random sampling for the investigation. Meanwhile, the factors influencing KAP were analyzed using the aforementioned two models. The results of the binary logistic regression model showed that the patient age [odds ratio [Formula: see text]; 95% confidence interval (CI): 0.117–0.295], spouse education ([Formula: see text]; 95% CI: 0.267–0.422), and spouse occupation ([Formula: see text]; 95% CI: 1.078–1.693) were the factors influencing VD-related KAP. The decision tree chi-squared automatic interaction detector (CHAID) algorithm showed that spouse education, patient age, and living room lighting were the influencing factors, among which spouse education was the most important. VD in middle-aged and elderly patients with ankle fractures was affected by the age and the education level of the spouse. These two models can analyze the influencing factors of knowledge and practice, and can be used together to complement each other.

Fuzzy Neural Logic Reasoning for Robust Classification

The efficacy of neural networks is widely recognized across a multitude of machine learning tasks, yet their black-box nature impedes the understanding of their decision-making processes. Such lack of explainability limits their use in high-stake fields such as medicine and finance, where transparent decision-making is essential. In contrast, traditional rule-based models offer clear input-output mappings, but often lag in performance when compared to their neural network counterparts. To address this challenge, this study introduces Fuzzy Neural Logic Reasoning (FNLR), a novel architecture that combines the best of both rule-based and deep learning models to achieve performance, interpretability, and noise robustness simultaneously. At its core, FNLR employs a “Symbolic Pre-training + Neural Fine-tuning” paradigm. Initially, the model adapts a pre-fitted binary decision tree. It then performs a “neuralization” process, replacing each node of the tree with a corresponding neural network equivalent. This transformation is facilitated through three shallow MLP modules, which are trained to emulate the relational operators intrinsic to decision trees. The model architecture is also extensible, allowing it to further boost expressiveness. Furthermore, FNLR incorporates fuzzy logic by proposing novel fuzzy relational operators, accounting for satisfaction degrees of propositions and thus eliminating rigid decision boundaries. This approach enhances model flexibility, enabling all paths of the decision tree to contribute to the target prediction in a weighted manner. Empirical evaluations on tabular datasets from various domains demonstrate that FNLR performs comparably to, or better than, state-of-the-art deep learning models designed for tabular data, while also exhibiting strong robustness to noise.

Unsupervised Decision Trees for Axis Unimodal Clustering

The use of decision trees for obtaining and representing clustering solutions is advantageous, due to their interpretability property. We propose a method called Decision Trees for Axis Unimodal Clustering (DTAUC), which constructs unsupervised binary decision trees for clustering by exploiting the concept of unimodality. Unimodality is a key property indicating the grouping behavior of data around a single density mode. Our approach is based on the notion of an axis unimodal cluster: a cluster where all features are unimodal, i.e., the set of values of each feature is unimodal as decided by a unimodality test. The proposed method follows the typical top-down splitting paradigm for building axis-aligned decision trees and aims to partition the initial dataset into axis unimodal clusters by applying thresholding on multimodal features. To determine the decision rule at each node, we propose a criterion that combines unimodality and separation. The method automatically terminates when all clusters are axis unimodal. Unlike typical decision tree methods, DTAUC does not require user-defined hyperparameters, such as maximum tree depth or the minimum number of points per leaf, except for the significance level of the unimodality test. Comparative experimental results on various synthetic and real datasets indicate the effectiveness of our method.

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

Using ensemble learning method and binary decision tree algorithm for drought intensity level classification

This study applies machine learning methods such as Decision Tree (CART) and Random Forest to classify drought intensity based on meteorological data. The goal of the study was to evaluate the effectiveness of these methods for drought classification and their use in water resource management and agriculture. The methodology involved using two machine learning models that analyzed temperature and humidity indicators, as well as wind speed indicators. The models were trained and tested on real meteorological data to assess their accuracy and identify key factors affecting predictions. Results showed that the Random Forest model achieved the highest accuracy of 94.4% when analyzing temperature and humidity indicators, while the Decision Tree (CART) achieved an accuracy of 93.2%. When analyzing wind speed indicators, the models’ accuracies were 91.3% and 93.0%, respectively. Feature importance revealed that atmospheric pressure, temperature at 2 m, and wind speed are key factors influencing drought intensity. One of the study’s limitations was the insufficient amount of data for high drought levels (classes 4 and 5), indicating the need for further data collection. The innovation of this study lies in the integration of various meteorological parameters to build drought classification models, achieving high prediction accuracy. Unlike previous studies, our approach demonstrates that using a wide range of meteorological data can significantly improve drought classification accuracy. Significant findings include the necessity to expand the dataset and integrate additional climatic parameters to improve models and enhance their reliability.

Genetic Algorithms for Feature Selection in the Classification of COVID-19 Patients.

Severe Acute Respiratory Syndrome CoronaVirus-2 (SARS-CoV-2) infection can cause feared consequences, such as affecting microcirculatory activity. The combined use of HRV analysis, genetic algorithms, and machine learning classifiers can be helpful in better understanding the characteristics of microcirculation that are mainly affected by COVID-19 infection. This study aimed to verify the presence of microcirculation alterations in patients with COVID-19 infection, performing Heart Rate Variability (HRV) parameters analysis extracted from PhotoPlethysmoGraphy (PPG) signals. The dataset included 97 subjects divided into two groups: healthy (50 subjects) and patients affected by mild-severity COVID-19 (47 subjects). A total of 26 parameters were extracted by the HRV analysis and were investigated using genetic algorithms with three different subject selection methods and five different machine learning classifiers. Three parameters: meanRR, alpha1, and sd2/sd1 were considered significant, combining the results obtained by the genetic algorithm. Finally, machine learning classifications were performed by training classifiers with only those three features. The best result was achieved by the binary Decision Tree classifier, achieving accuracy of 82%, specificity (or precision) of 86%, and sensitivity of 79%. The study's results highlight the ability to use HRV parameters extraction from PPG signals, combined with genetic algorithms and machine learning classifiers, to determine which features are most helpful in discriminating between healthy and mild-severity COVID-19-affected subjects.

Using machine learning techniques for exploration and classification of laboratory data

Abstract Objectives The study aims to acquaint readers with six widely used machine learning (ML) techniques (Principal Component Analysis (PCA), Uniform Manifold Approximation and Projection (UMAP), k-means, hierarchical clustering and the decision tree models (rpart and random forest)) that might be useful for the analysis of laboratory data. Methods Utilizing a recently validated data set from lung cancer diagnostics, we investigate how ML can support the search for a suitable tumor marker panel for the differentiation of small cell (SCLC) and non-small cell lung cancer (NSCLC). Results The ML techniques used here effectively helped to gain a quick overview of the data structures and provide initial answers to the clinical questions. Dimensionality reduction techniques such as PCA and UMAP offered insightful visualization and impression of the data structure, suggesting the existence of two tumor groups with a large overlap of largely inconspicuous values. This impression was confirmed by a cluster analysis with the k-means algorithm, indicative of unsupervised learning. For supervised learning, decision tree models like rpart or random forest demonstrated their utility in differential diagnosis of the two tumor types. The rpart model, which constructs binary decision trees based on the recursive partitioning algorithm, suggests a tree involving four serum tumor markers (STMs), which were confirmed by the random forest approach. Both highlighted pro-gastrin-releasing peptide (ProGRP), neuron specific enolase (NSE), cytokeratin-19 fragment (CYFRA 21-1) and cancer antigen (CA) 72-4 as key tumor markers, aligning with the outcomes of the initial statistical analysis. Cross-validation of the two proposals showed a higher area under the receiver operating characteristic (AUROC) curve of 0.95 with a 95 % confidence interval (CI) of 0.92–0.97 for the random forest model compared to an AUROC curve of 0.88 (95 % CI: 0.83–0.93). Conclusions ML can provide a useful overview of inherent medical data structures and distinguish significant from less pertinent features. While by no means replacing human medical and statistical expertise, ML can significantly accelerate the evaluation of medical data, supporting a more informed diagnostic dialogue between physicians and statisticians.

Explainable finite mixture of mixtures of bounded asymmetric generalized Gaussian and Uniform distributions learning for energy demand management

We introduce a mixture of mixtures of bounded asymmetric generalized Gaussian and uniform distributions. Based on this framework, we propose model-based classification and model-based clustering algorithms. We develop an objective function for the minimum message length (MML) model selection criterion to discover the optimal number of clusters for the unsupervised approach of our proposed model. Given the crucial attention received by Explainable AI (XAI) in recent years, we introduce a method to interpret the predictions obtained from the proposed model in both learning settings by defining their boundaries in terms of the crucial features. Integrating Explainability within our proposed algorithm increases the credibility of the algorithm’s predictions since it would be explainable to the user’s perspective through simple If-Then statements using a small binary decision tree. In this paper, the proposed algorithm proves its reliability and superiority to several state-of-the-art machine learning algorithms within the following real-world applications: fault detection and diagnosis (FDD) in chillers, occupancy estimation and categorization of residential energy consumers.

Generative AI enhanced with NCCN clinical practice guidelines for clinical decision support: A case study on bone cancer.

e13623 Background: Bone cancer is a complex and challenging disease to diagnose and treat in clinical practice. Recently, generative AI, especially large language models (LLMs), has demonstrated potential as a decision support tool for cancer. However, most implementations have overlooked the integration of available cancer guidelines, such as the NCCN Bone Cancer Guidelines, in fine-tuning the outputs of generative AI models. Incorporating these guidelines into LLMs presents an opportunity to harness the extensive clinical knowledge they contain and improve the decision-support capabilities of the model. Methods: In this study, the aim is to enhance the LLM with cancer clinical guidelines to enable accurate medical decisions and personalized treatment recommendations. Therefore, we introduce a novel method for incorporating the NCCN Bone Cancer Guidelines into LLMs using a Binary Decision Tree (BDT) approach. The approach involves constructing a BDT based on NCCN Bone Cancer Guidelines, where internal nodes represent decision points from the Guidelines, and leaf node signify final treatment suggestions. Then the LLM makes decision at each internal node, considering a given patient's characteristics, and guides toward a treatment recommendation in the leaf node. To assess the efficacy of Guideline-enhanced LLMs, an oncologist from our team created 11 hypothetical osteosarcoma patients’ medical progress notes. Each note contains their demographics, medical history, current illness, physical exams, diagnostic tests. We tested three LLMs in the implementation (GPT-4, GPT-3.5, and PaLM 2) and compared the LLM-generated treatment recommendations with the gold standard treatment across four runs with different random seeds (random seeds is a setting to control the LLM outputs). The results are reported as the average of four runs. The original LLMs are used as baseline methods for comparison. Results: The table below provides a comparison between the performance of original LLMs and those augmented with cancer guidelines for osteosarcoma treatment recommendations. We can observe that the PaLM 2 model demonstrated superior performance compared to its counterparts, underscoring the effectiveness of integrating cancer guidelines into LLMs for decision support. Conclusions: The clinical decision support capabilities of the LLMs are promising when enhanced by NCCN Bone Cancer Guidelines using our approach. To fully exhibit the potential of our proposed method as a clinical decision support tool, further investigation into other subtypes of bone cancer should be conducted in the future study. [Table: see text]

A novel fusion-based approach for the classification of packets in wireless body area networks

This abstract focuses on the significance of wireless body area networks (WBANs) as a cutting-edge and self-governing technology, which has garnered substantial attention from researchers. The central challenge faced by WBANs revolves around upholding quality of service (QoS) within rapidly evolving sectors like healthcare. The intricate task of managing diverse traffic types with limited resources further compounds this challenge. Particularly in medical WBANs, the prioritization of vital data is crucial to ensure prompt delivery of critical information. Given the stringent requirements of these systems, any data loss or delays are untenable, necessitating the implementation of intelligent algorithms. These algorithms play a pivotal role in expediting diagnosis and treatment processes during medical emergencies. This study introduces an innovative protocol termed collaborative binary Naive Bayes decision tree (CBNBDT) designed to enhance packet classification and transmission prioritization. Through the utilization of this protocol, incoming packets are categorized based on their respective classes, enabling subsequent prioritization. Thorough simulations have demonstrated the superior performance of the proposed CBNBDT protocol compared to baseline approaches.

Remote sensing monitoring of changes in forest cover in the Volyn region: a cross section for the first two decades of the 21st century

The aim of the article. This article highlights the significance of forest cover as an important indicator of the state of the environment. It discusses the findings of the Food and Agriculture Organization of the United Nations (FAO) Forest Resources Assessment (FRA) 2020 report, which states that the world's forest area has decreased by 178 million hectares since 1990. The case study of Volyn region shows how cloud processing and vegetation classification can help quantify forest dynamics from 2000 to 2020, allowing local authorities and decision makers to monitor and analyze trends in near real time. Overall, this work provides insights into the importance of monitoring forest dynamics and the potential for remote sensing technology to facilitate this process. Data & Methods. Remote sensing is an effective tool for monitoring forest ecology and management, and Google Earth Engine (GEE) is an online platform that combines data from various agencies to analyze environmental data. The article presents a case study of the Volyn region and how cloud processing and vegetation classification were used to assess forest dynamics from 2000 to 2020. The study used data from Landsat 7 Collection 1 Tier 1 composites and the CART algorithm for binary decision tree building. The study was based on information provided by the Main Department of Statistics in the Volyn region on the area of forests and areas where logging was carried out during the specified period. Research results. It is interesting to note that despite the decrease in logging activities, there is an increase in forest cover loss within forest ranges. This could be due to various reasons, such as illegal logging or natural disturbances like fires or disease outbreaks. The use of machine learning methods like CART classification can help to identify and monitor these changes, which can then be used to inform policy decisions and management practices to reduce forest cover loss. In general, in the Volyn region, there is a gradual decrease in the areas where various kinds of logging are carried out from 524 km2 in 2003 to 239 km2 in 2020. In contrast, forest cover loss within forest ranges increased rapidly from 37.85 km2 in 2015 to 84.01 km2 in 2017 and beyond from 5.53 km2 to 10.80 km2 in 2015 and 2017 respectively. In this study, the accuracy assessment was performed using 30% of the control points obtained initially, based on data on the reliability of the land cover. The manufacturer's accuracy and user accuracy were calculated to evaluate error omissions and possibilities of a pixel being categorized in a certain category. The spatial resolution of Landsat 7 data used in this study was 30 m, with a minimum calculation area of 0.337 hectares. The overall accuracy and the coefficient κ are the most representative measures of accuracy, with an average accuracy of classification of OAav=98.82% and κav=0.9764.

Pengesanan Denyutan Pi2 Hampir Masa Nyata Berdasarkan Model Pembelajaran Mesin Terautomatik

The impact of geomagnetic substorms on space weather and various technology on the Earth urges the need to detect the phenomenon. However, the brief period of substorm occurrences makes it more difficult to directly detect them, therefore, their detection by proxy in the form of precursors need to be done. Pi2 pulsations, which is a damped irregular signal in the horizontal component of the geomagnetic field ( as a reliable precursor to substorms. The cessation of Hz), is seen index that previously was the main detection indicator of Pi2 pulsations motivates this study to introduce a new detection method by implementing machine learning. In this study, several features were extracted from geomagnetic field data based on several statistical parameters, impulsive and signal metrics to be used in the development of classification model. The development was performed based on the automated machine learning (AutoML) approach that determines the best algorithm and optimizes the hyperparameters automatically. AutoML tries various algorithms like ensemble, neural network, Naïve Bayes, support vector machine, k-nearest neighbour and binary decision tree that are optimized by Asynchronous Successive Halving Algorithm. An ensemble classification model was determined to be the best performing model with an accuracy of 99.15%. It was then used in the development of near-real tine detection system. The system streams the geomagnetic field data continuously and reports the Pi2 detections on an open cloud repository, The detection system is envisioned to be one of the reference sources to inquire the occurrences of Pi2 pulsations globally.

Ai solutions for digital diagnostics of diseases of grain crops (using the example of <i>Pyrenophora teres</i> of winter barley)

The aim of the research is to justify the feasibility of using digital intelligent technologies in forecasting the development of spot blotch in winter barley. The AI solution developed is a binary decision tree capable of predicting scenarios of net blotch development: depressive, moderate, and epiphytotic. To configure the algorithm’s parameters from 2021 to 2023, field and laboratory experiments were conducted at the Federal Scientific Center for Biological Plant Protection. The data preparation involved several stages, including setting up field plots to create an artificial infection background, preparing an inoculum, sowing highly susceptible and resistant winter barley varieties, and artificial inoculation. The selected input factors included the observed degree of leaf damage, type of variety resistance, vegetation phase at the time of primary infection, average relative air humidity during the vegetation phase of infection. The total sample size comprised 144 observations. The trained model demonstrated high classification accuracy on both the training and test datasets, with an accuracy rate exceeding 96 %. Based on a statistical assessment of the factors influencing the development of spot blotch in barley, it is shown that the most influential factor is the current degree of leaf infection (74,3 %), followed by the average relative air humidity (11,9 %), the variety’s resistance to the disease (10,4 %), and the stage development during which infection occurred (3,4 %). The proposed solution holds significant practical importance as it provides new opportunities for the diagnostic process of spot blotch in winter barley. Among these are high diagnostic speed, accuracy in forecast predictions, and applicability in field conditions.

Exploring multisite musculoskeletal symptoms among sewing machine operators in a tunisian leather and footwear industry using decision tree models

Background/objectivesSewing machine operators (SMO) are the most likely workers to experience a high prevalence of musculoskeletal disorders in the textile, clothing, and footwear industries. We conducted a cross-sectional and exhaustive study among SMO working in the leather and footwear industry to describe the prevalence of multi-site musculoskeletal symptoms (MMS) and evaluate factors associated with their occurrence. MethodsMusculoskeletal symptoms declared by these operators were assessed through the modified Nordic questionnaire. The psychosocial work environment was assessed using the Karasek model. The variables associated with MMS were issued from binary logistic regression and decision tree using R software. ResultsOf 145 operators, 65.5 % of men and 72.4 % of women had MMS. Based on binary logistic regression, a history of musculoskeletal disorders (MSDs) increased the risk of developing MMS by 8 folds. The binary decision tree identified five main nodes: history of MSDs, professional seniority, often finding the pace of work restrictive and male gender. ConclusionIdentifying homogeneous profiles of MMS's occurrence will help the implementation of an effective and targeted preventive strategy.

Classification of LED Packages for Quality Control by Discriminant Analysis, Neural Network and Decision Tree.

This study investigates supervised learning to improve LED classification. A hardware system for testing was built. The data for learning were acquired and then analyzed to show their characteristics. An LED was tested, and the results were categorized into three defective LED groups and one normal LED group. Before classification, electrical and optical data were examined to identify their characteristics. To find out the best way for quality control, an ensemble of methods was used. First, the discriminant analysis using the validation data achieved a 77.9% true positive rate for normal products, inadequate for quality control. Second, neural network-based learning boosted this rate to 97.8%, but the 2.2% false negative rate remained problematic. Finally, a binary decision tree was constructed, achieving a 99.4% true positive rate from just 14 splits, proving highly effective in product classification. The training time was measured as 8.1, 18.2 and 8.2 s for discriminant analysis, neural network and decision tree, respectively. This work has found the binary decision tree is advantageous considering both learning and classification efficiencies.

Rational satisficing heuristics as determinants of online search behavior

We design a set of satisficing heuristic algorithms that mimic the online information retrieval behavior of rational decision makers (DMs) as reflected in their click through rates (CTRs). We illustrate how basic heuristic algorithms formalizing binary decision trees composed by 21 nodes and requiring DMs to observe one satisficing alternative lack the structural capacity to characterize the retrieval process. The algorithm requiring DMs to observe two satisficing alternatives –formalizing binary decision trees composed by 111 nodes – provides a sufficient approximation to their CTRs. Adding a third alternative – accounting for 351 nodes – delivers an almost identical set of CTRs to those displayed by DMs. The mimicking quality of the heuristic algorithms prevails as alternatives are added up to include the ten ranked within the first page of search results, incorporating 2,047 nodes to formalize the corresponding retrieval process. The set of algorithms bridges the gap between purely rational decision theory and heuristic behavior, illustrating how the CTRs observed can be generated by rational DMs performing a sequential search process while aiming to observe two or three satisficing alternatives. The decision-tree algorithmic structures presented are sufficiently malleable to introduce any potential modification to the beliefs and preferences of DMs and study its consequences in terms of CTRs.

Machine learning methods in differential diagnosis of ACTH-dependent hypercortisolism

To develop a noninvasive method of differential diagnosis of ACTH-dependent hypercortisolism, as well as to evaluate the effectiveness of an optimal algorithm for predicting the probability of ectopic ACTH syndrome (EAS) obtained using machine learning methods based on the analysis of clinical data. As part of a single-center, one-stage, cohort study, a retrospective prediction of the probability of EAS among patients with ACTH-dependent hypercortisolism was carried out. Patients were randomly stratified into 2 samples: training (80%) and test (20%). Eleven machine learning algorithms were used to develop predictive models: Linear Discriminant Analysis, Logistic Regression, elastic network (GLMNET), Support Vector machine (SVM Radial), 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). The study included 223 patients (163 women, 60 men) with ACTH-dependent hypercortisolism, of which 175 patients with Cushing's disease (CD), 48 - with EAS. As a result of preliminary data processing and selection of the most informative signs, the final variables for the classification and prediction of EAS were selected: ACTH level at 08:00 hours, potassium level (the minimum value of potassium in the active stage of the disease), 24-h urinary free cortisol, late-night serum cortisol, late-night salivary cortisol, the largest size of pituitary adenoma according to MRI of the brain. The best predictive ability in a training sample of all trained machine learning models for all three final metrics (ROC-AUC (0.867), sensitivity (90%), specificity (56.4%)) demonstrated a model of gradient boosting (Generalized Boosted Modeling, GBM). In the test sample, the AUC, sensitivity and specificity of the model in predicting EAS were 0.920; 77.8% and 97.1%, respectively. The prognostic model based on machine learning methods makes it possible to differentiate patients with EAS and CD based on basic clinical results and can be used as a primary screening of patients with ACTH-dependent hypercortisolism.

Empowering In-Network Classification in Programmable Switches by Binary Decision Tree and Knowledge Distillation

Given the high packet processing efficiency of programmable switches (e.g., P4 switches of Tbps), several works are proposed to offload the decision tree (DT) to P4 switches for in-network classification. Although the DT is suitable for the match-action paradigm in P4 switches, the range match rules used in the DT may not be supported across devices of different P4 standards. Additionally, emerging models including neural networks (NNs) and ensemble models, have shown their superior performance in networking tasks. But their sophisticated operations pose new challenges to the deployment of these models in switches. In this paper, we propose Mousikav2 to address these drawbacks successfully. First, we design a new tree model, i.e., the binary decision tree (BDT). Unlike the DT, our BDT consists of classification rules in the form of bits, which is a good fit for the standard ternary match supported by different hardware/software switches. Second, we introduce a teacher-student knowledge distillation architecture in Mousikav2, which enables the general transfer from other sophisticated models to the BDT. Through this transfer, sophisticated models are indirectly deployed in switches to avoid switch constraints. Finally, a lightweight P4 program is developed to perform classification tasks in switches with the BDT after knowledge distillation. Experiments on three networking tasks and three commodity switches show that Mousikav2 not only improves the classification accuracy by 3.27%, but also reduces the switch stage and memory usage by 2.00 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\times$</tex-math> </inline-formula> and 28.67%, respectively. Code is available at https://github.com/xgr19/Mousika.

Determination of the most significant rubber components influencing the hardness of natural rubber (NR) using various statistical methods

Manufacturers use a large number of components in the production of modern rubber products. The selection of the constituents of the rubber recipe is primarily determined by the purpose of use. The different fields of applications of rubbers require the presence of appropriate mechanical properties. In this respect, it can be useful to know which substances forming the rubber recipe have significant influence on the different mechanical properties. In this study, the statistical analysis of the influence of rubber components on the hardness of natural rubber (NR) is proposed based on literature review. Based on the literature data, various statistical analyses, like linear regression, constrained linear regression, Ridge regression, Ridge sparse regression and binary classification decision trees were performed to determine which rubber components have the most significant effect on the hardness. In the statistical analyses, the effect of a total of 42 constituents of rubber compound on hardness was investigated. Most of the applied statistical methods confirmed that the traditional frequently used rubber components, such as carbon black and sulfur, have a primary effect on the hardness. However, the substances forming the rubber compound that are not widely used in practice or newly developed components appear differently in the lists of significant additives obtained by the different statistical methods.