Decision Tree Learning Model Research Articles

Design and performance analysis of decision tree learning model for classification of dry and cooked rice samples

Design and performance analysis of decision tree learning model for classification of dry and cooked rice samples

Adapting video-based programming instruction: An empirical study using a decision tree learning model

Adapting video-based programming instruction: An empirical study using a decision tree learning model

Interpretable land cover classification with modal decision trees

ABSTRACT Land cover classification (LCC) refers to the task of classifying each pixel in satellite/aerial imagery by predicting a label carrying information about its nature. Despite the importance of having transparent, symbolic decision models, in the recent literature, LCC has been mainly approached with black-box functional models, that are able to leverage the spatial dimensions within the data. In this article, we argue that standard symbolic decision models can be extended to perform a form of spatial reasoning that is adequate for LCC. We propose a generalization of a classical decision tree learning model, based on replacing propositional logic with a modal spatial logic, and provide a CART-like learning algorithm for it. We evaluate its performance at five different LCC tasks, showing that this technique leads to classification models whose performances are superior to those of their propositional counterpart, and at least comparable with those of non-symbolic ones. Ultimately, we show that spatial decision trees and random forests are able to extract complex, but interpretable spatial patterns.

Construction of a Recommendation Method for Financial Insurance Products Based on Machine Learning

In the era of the rise of big data, if insurance companies can effectively and reasonably use existing data to tap more potential information, it can not only improve work efficiency but also achieve precise marketing to customers, thereby saving costs. At the same time, China pays more and more attention to financial insurance, so it is of great significance to use customer information to explore the purchase behavior of financial insurance. This paper mainly analyzes the factors that affect the purchase of financial insurance from the aspects of individuals and families and provides references for insurance marketing. This paper selects the results of the China Comprehensive Social Survey in 2020 as the empirical sample data of this paper and sets whether to purchase financial insurance as the target variable. The characteristic variables were screened, and the 10 variables selected after screening were finally used to build the model. Next, the data are divided into training set and test set, and a decision tree learning model is established on the two data sets at the same time. The classification results of the model are evaluated according to the classification evaluation indicators. Finally, the decision tree analysis results provide suggestions and strategies for the construction of insurance marketing recommendation methods.

Analysis of an Economic Coupling Relationship Model of the Coastal Ecological Fragile Zone Based on a Machine Learning Model

This study is based on a decision tree learning model and integrated system. In order to improve the classification accuracy of the base learner, the diversity of the results is analyzed. Accuracy and diversity are the focus of decision tree classifier research. The balance between decision tree classifiers is a key issue in creating a good ensemble classifier. By reducing the redundant features and training the model with the optimal feature subset, the accuracy of the model can be improved and the running time can be reduced. Normal distribution and generalized error distribution are better. The clustering algorithm based on the K -means partition is a number of compact and independent classification clusters based on the K -means partition algorithm. In order to objectively and accurately treat the machine learning model, an analysis model and economic coupling relationship model were used to evaluate the coastal ecological fragile zone. The error of the result of the classifier after training is smaller.

Analysis of Failure Features of High-Speed Automatic Train Protection System

An automatic train protection (ATP) system is the core to ensure operation safety of high-speed railway. At present, failure rate change rules of the system are not well understood and the maintenance strategy is not refined. In order to improve the protection capability and maintenance level of high-speed trains, this paper proposes a decision tree machine learning model for failure feature extraction of ATP systems. First, system type, mean operation mileage, mean service time, etc. are selected as ATP failure feature parameters, and cumulative failure rate is selected as a model output label. Second, support vector machine, AdaBoost, artificial neural networks and decision tree model are adopted to train and test practical failure data. Performance analysis shows that decision tree learning model has better generalization ability. The accuracy of 0.9761 is significantly greater than the other machine learning models. Therefore, it is most suitable for failure features analysis. Third, interpretability analysis reveals the quantitative relationship between system failure and features. Finally, an intelligent maintenance system for ATP systems is built, which realize the refined maintenance throughout life cycle.

P-Wave Area Predicts New Onset Atrial Fibrillation in Mitral Stenosis: A Machine Learning Approach.

Introduction: Mitral stenosis is associated with an atrial cardiomyopathic process, leading to abnormal atrial electrophysiology, manifesting as prolonged P-wave duration (PWD), larger P-wave area, increased P-wave dispersion (PWDmax—PWDmin), and/or higher P-wave terminal force on lead V1 (PTFV1) on the electrocardiogram.Methods: This was a single-center retrospective study of Chinese patients, diagnosed with mitral stenosis in sinus rhythm at baseline, between November 2009 and October 2016. Automated ECG measurements from raw data were determined. The primary outcome was incident atrial fibrillation (AF).Results: A total 59 mitral stenosis patients were included (age 59 [54–65] years, 13 (22%) males). New onset AF was observed in 27 patients. Age (odds ratio [OR]: 1.08 [1.01–1.16], P = 0.017), systolic blood pressure (OR: 1.03 [1.00–1.07]; P = 0.046), mean P-wave area in V3 (odds ratio: 3.97 [1.32–11.96], P = 0.014) were significant predictors of incident AF. On multivariate analysis, age (OR: 1.08 [1.00–1.16], P = 0.037) and P-wave area in V3 (OR: 3.64 [1.10–12.00], P = 0.034) remained significant predictors of AF. Receiver-operating characteristic (ROC) analysis showed that the optimum cut-off for P-wave area in V3 was 1.45 Ashman units (area under the curve: 0.65) for classification of new onset AF. A decision tree learning model with individual and non-linear interaction variables with age achieved the best performance for outcome prediction (accuracy = 0.84, precision = 0.84, recall = 0.83, F-measure = 0.84).Conclusion: Atrial electrophysiological alterations in mitral stenosis can detected on the electrocardiogram. Age, systolic blood pressure, and P-wave area in V3 predicted new onset AF. A decision tree learning model significantly improved outcome prediction.

An Identification Decision Tree Learning Model for Self-Management in Virtual Radio Access Network: IDTLM

Along with the blowout of new applications and the integration of the heterogeneous networks platforms in the future Internet of everything, the self-management of virtual radio access network is of significant importance. The urgent problem needed to be solved for the self-management in virtual radio access network is the match of application and virtual service (tailored service of virtual functions for the application). In this paper, an identification decision tree learning model (IDTLM) based on transfer learning has been proposed. First, we do research on the redundant problem of the traditional packet decision trees and reduce the dimensionality of features by a proximal gradient descent method and clustering the features by Lagrange’s multiplier, so as to improve the online matching speed between applications and virtual service. And then, in consideration of the independent and non-identical distribution among online and trained data, and the possible change of virtualized network platform, a method of transfer learning is proposed to improve the quality of generalization for IDTLM. Finally, online test is done for IDTLM, and the result shows that the accuracy rate of trained applications can reach 99% and the accuracy rate can reach 96% if untrained applications are included. Meanwhile, theoretical analysis has been carried out for the transfer of IDTLM. The analysis shows that IDTLM system is of high recognizing speed and low false positive rate and it could adapt to the transfer of different scenarios.

Abstract 2873: A molecular signature predictive of indolent prostate cancer

Abstract With more than 230,000 estimated new cases in 2013, prostate cancer represents the most commonly diagnosed cancer in American men. While a subgroup of patients will develop an aggressive and lethal prostate cancer, most present with a low-risk form of the disease that requires little or no treatment. It is therefore crucial to distinguish patients whose tumors are likely to progress to aggressive disease from those whose tumors will remain indolent and for whom treatment may be more detrimental than beneficial. In our recent study, we hypothesized that low Gleason score prostate tumors can be distinguished as indolent and aggressive based on expression of genes associated with aging and senescence (1). Using gene set enrichment analyses and a decision tree learning model, we identified a 3-gene panel_FGFR1, PMP22, and CDKN1A_that is enriched in indolent prostate cancers and accurately predicts outcome of low Gleason score tumors at both the RNA and protein levels. In particular, protein expression of this 3-gene panel in biopsy samples correctly distinguished Gleason 6 patients who failed active surveillance over a 10-year period. Following from our initial study, our current efforts are focused on exploiting this 3-gene panel to develop a robust clinical assay for distinguishing indolent and aggressive tumors. We are currently using qRT-PCR, immunofluorescence, and immunohistochemistry assays to assess the most effective way to quantify the simultaneous expression of the 3-gene panel in low Gleason score tumor biopsies/tissues. This quantitative assay will ultimately be applied to a large cohort study assessing the predictive power of the 3-gene signature in patients under active surveillance. We envision that such a prognostic test will significantly improve our ability to predict the outcome of patients with low Gleason score tumors and facilitate appropriate decisions of treatment for men under active surveillance. 1. Irshad S, et al. (2013) A molecular signature predictive of indolent prostate cancer. Science translational medicine 5(202):202ra122. Citation Format: Shazia Irshad, Mukesh Bansal, Clementine Le Magnen, Risham Dillon, Mireia Castillo-Martin, Tian Zheng, Alvaro Aytes, Sven Wenske, Paolo Guarnieri, Pavel Sumazin, Mitchell Benson, Michael M. Shen, Andrea Califano, Cory Abate-Shen. A molecular signature predictive of indolent prostate cancer. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 2873. doi:10.1158/1538-7445.AM2014-2873

Neuroticism and impulsivity: Their hierarchical organization in the personality characterization of drug-dependent patients from a decision tree learning perspective

ObjectiveNeuroticism and impulsivity are the personality variables most consistently associated with drug-dependent patients. To date, no data mining procedures have been applied to explore the differential role of personality variables in this population. MethodsThe personality profile of 336 drug-dependent patients was compared with that of a sample of community participants in the context of a decision tree learning approach using the Alternative Five Factor Model. The resulting discriminant model was cross-validated. ResultsNeuroticism and impulsivity were the most relevant variables in the resulting model, but their association appeared to be hierarchically organized. In the personality characterization of these patients, neuroticism became the main discriminant dimension, whereas impulsivity played a differential role, explained by means of an interaction effect. Decision tree learning models appear to be a heuristic theoretical and empirical approximation to the study of relevant variables, such as personality traits, in drug-dependency research.

Ensemble of Decision Tree Classifiers for Mining Web Data Streams

The World Wide Web (www or w3 commonly known as the web) is the largest database available with growth at the rate of millions of pages a day and presents a challenging task for mining web data streams. Currently extraction of knowledge from web data streams is getting more and more complex, because the structure of data doesn’t match the attribute-values when considering the large volume of web data. In this paper, an ensemble of decision tree classifiers is presented, which is an efficient mining method to obtain a proper set of rules for extracting knowledge from a large amount of web data streams. We built a web server using Model 2 Architecture to collect the web data streams and applied the ensemble classifier for generating decision rules using several decision tree learning models. Experimental results demonstrate that the proposed method performs well in decision making and predicting the class value of new web data streams.

Using decision tree learning to predict the responsiveness of hepatitis C patients to drug treatment

The recommended treatment for patients with chronic hepatitis C, pegylated interferon α (PEG-IFN-α) plus rebavirin (RBV), does not provide a sustained virologic response in all patients, especially those with hepatitis C virus (HCV) genotype 1. It is therefore important to predict whether or not a new patient with HCV genotype 1 will be cured by the recommended treatment. We propose a prediction method for a new patient using a decision tree learning model based on SNPs evaluated in a genome-wide association study. By the decision tree learning for 142 Japanese patients with HCV genotype 1 (78 with null virologic response and 64 with virologic response), we can predict with high probability (93%) whether or not a new patient with HCV will be helped by the recommended treatment.

LEARNING DECISION TREES WITH LOG CONDITIONAL LIKELIHOOD

In machine learning and data mining, traditional learning models aim for high classification accuracy. However, accurate class probability prediction is more desirable than classification accuracy in many practical applications, such as medical diagnosis. Although it is known that decision trees can be adapted to be class probability estimators in a variety of approaches, and the resulting models are uniformly called Probability Estimation Trees (PETs), the performances of these PETs in class probability estimation, have not yet been investigated. We begin our research by empirically studying PETs in terms of class probability estimation, measured by Log Conditional Likelihood (LCL). We also compare a PET called C4.4 with other representative models, including Naïve Bayes, Naïve Bayes Tree, Bayesian Network, KNN and SVM, in LCL. From our experiments, we draw several valuable conclusions. First, among various tree-based models, C4.4 is the best in yielding precise class probability prediction measured by LCL. We provide an explanation for this and reveal the nature of LCL. Second, compared with non tree-based models, C4.4 also performs best. Finally, LCL does not dominate another well-established relevant metric — AUC, which suggests that different decision-tree learning models should be used for different objectives. Our experiments are conducted on the basis of 36 UCI sample sets. We run all the models within a machine learning platform — Weka. We also explore an approach to improve the class probability estimation of Naïve Bayes Tree. We propose a greedy and recursive learning algorithm, where at each step, LCL is used as the scoring function to expand the decision tree. The algorithm uses Naïve Bayes created at leaves to estimate class probabilities of test samples. The whole tree encodes the posterior class probability in its structure. One benefit of improving class probability estimation is that both classification accuracy and AUC can be possibly scaled up. We call the new model LCL Tree (LCLT). Our experiments on 33 UCI sample sets show that LCLT outperforms all state-of-the-art learning models, such as Naïve Bayes Tree, significantly in accurate class probability prediction measured by LCL, as well as in classification accuracy and AUC.

Decision tree learning with fuzzy labels

Label semantics is a random set based framework for “Computing with Words” that captures the idea of computation on linguistic terms rather than numerical quantities. Within this new framework, a decision tree learning model is proposed where nodes are linguistic descriptions of variables and leaves are sets of appropriate labels. In such decision trees, the probability estimates for branches across the whole tree is used for classification, instead of the majority class of the single branch into which the examples fall. By empirical experiments on real-world datasets it is verified that our algorithm has better or equivalent classification accuracy compared to three well known machine learning algorithms. By applying a new forward branch merging algorithm, the complexity of the tree can be greatly reduced without significant loss of accuracy. Finally, a linguistic interpretation of trees and classification with linguistic constraints are introduced.