Using Model Trees and Their Ensembles for Imbalanced Data
نویسندگان
چکیده
Model trees are decision trees with linear regression functions at the leaves. Although originally proposed for regression, they have also been applied successfully in classification problems. This paper studies their performance for imbalanced problems. These trees give better results that standard decision trees (J48, based on C4.5) and decision trees specific for imbalanced data (CCPDT: Class Confidence Proportion Decision Trees). Moreover, different ensemble methods are considered using these trees as base classifiers: Bagging, Random Subspaces, AdaBoost, MultiBoost, LogitBoost and specific methods for imbalanced data: Random Undersampling and SMOTE. Ensembles of Model Trees also give better results than ensembles of the other considered trees.
منابع مشابه
Ensembles of (α)-Trees for Imbalanced Classification Problems
This paper introduces two kinds of decision tree ensembles for imbalanced classification problems, extensively utilizing properties of α-divergence. First, a novel splitting criterion based on α-divergence is shown to generalize several wellknown splitting criteria such as those used in C4.5 and CART. When the α-divergence splitting criterion is applied to imbalanced data, one can obtain decisi...
متن کاملEnhancing Learning from Imbalanced Classes via Data Preprocessing: A Data-Driven Application in Metabolomics Data Mining
This paper presents a data mining application in metabolomics. It aims at building an enhanced machine learning classifier that can be used for diagnosing cachexia syndrome and identifying its involved biomarkers. To achieve this goal, a data-driven analysis is carried out using a public dataset consisting of 1H-NMR metabolite profile. This dataset suffers from the problem of imbalanced classes...
متن کاملA simple plug-in bagging ensemble based on threshold-moving for classifying binary and multiclass imbalanced data
Class imbalance presents a major hurdle in the application of classification methods. A commonly taken approach is to learn ensembles of classifiers using rebalanced data. Examples include bootstrap averaging (bagging) combined with either undersampling or oversampling of the minority class examples. However, rebalancing methods entail asymmetric changes to the examples of different classes, wh...
متن کاملImproving Imbalanced data classification accuracy by using Fuzzy Similarity Measure and subtractive clustering
Classification is an one of the important parts of data mining and knowledge discovery. In most cases, the data that is utilized to used to training the clusters is not well distributed. This inappropriate distribution occurs when one class has a large number of samples but while the number of other class samples is naturally inherently low. In general, the methods of solving this kind of prob...
متن کاملNeighbourhood sampling in bagging for imbalanced data
Various approaches to extend bagging ensembles for class imbalanced data are considered. First, we review known extensions and compare them in a comprehensive experimental study. The results show that integrating bagging with under-sampling is more powerful than over-sampling. They also allow to distinguish Roughly Balanced Bagging as the most accurate extension. Then, we point out that complex...
متن کامل