SMOTE: Synthetic Minority Over-sampling Technique

Abstract

An approach to the construction of classifiers from imbalanced datasets is described. A dataset is imbalanced if the classification categories are not approximately equally represented. Often real-world data sets are predominately composed of "normal" examples with only a small percentage of "abnormal" or "interesting" examples. It is also the case that the cost of misclassifying an abnormal (interesting) example as a normal example is often much higher than the cost of the reverse error. Under-sampling of the majority (normal) class has been proposed as a good means of increasing the sensitivity of a classifier to the minority class. This paper shows that a combination of our method of over-sampling the minority (abnormal) class and under-sampling the majority (normal) class can achieve better classifier performance (in ROC space) than only under-sampling the majority class. This paper also shows that a combination of our method of over-sampling the minority class and under-sampling the majority class can achieve better classifier performance (in ROC space) than varying the loss ratios in Ripper or class priors in Naive Bayes. Our method of over-sampling the minority class involves creating synthetic minority class examples. Experiments are performed using C4.5, Ripper and a Naive Bayes classifier. The method is evaluated using the area under the Receiver Operating Characteristic curve (AUC) and the ROC convex hull strategy

Diversity analysis on imbalanced data sets by using ensemble models

“…At the data level, various re-sampling techniques are applied to balance class distribution, including over-sampling minority class instances and under-sampling majority class instances [5], [6], [7], [8] . Particularly, SMOTE (Synthetic Minority Over-sampling Technique) [1] is a popular approach designed for generating new minority class data, which could expand decision boundary towards majority class. At the algorithm level, solutions are proposed by adjusting algorithm itself, including adjusting the costs of various classes to counter the class imbalance, adjusting the decision threshold, and recognition-based (i.e., learning from one class) rather than discrimination-based (two class) learning.…”

“…These methods operate by taking a base learning algorithm and invoking it many times with different training sets. Therefore, some algorithms are proposed based on these two ensemble models by changing their re-sampling methods, such as BEV (Bagging Ensemble Variation) [11], SMOTEBoost [1], and DataBoost [12]. More details will be introduced in the Section 2.…”

Analyzing Medical Image Search Behavior: Semantics and Prediction of Query Results

“…To prevent deceitful results due to unbalanced classes, the Synthetic Minority Over-Sampling Technique (SMOTE) [21] was used to balance the classes. Once this was done, a set of machine learning algorithms was selected based on the state-ofthe-art tools used in the field, and experiments were conducted on these methods to determine which has the best performance in this specific case.…”

Combining experts in order to identify binding sites in yeast and mouse genomic data

“…In this paper we address the problem of our imbalanced data in two ways: firstly by using data based sampling techniques [16,17] and secondly by using different SVM error costs for the two classes [18].…”

“…Under sampling the majority class can be done by just randomly selecting a subset of the class. Over sampling the minority class is not so simple and here we use the Synthetic Minority Oversampling Technique (SMOTE) [16]. For each member of the minority class its nearest neighbours in the same class are identified and new instances are created, placed randomly between the instance and its neighbours.…”