• Journal of the Korean Statistical Society
• /
• v.36 no.1
• /
• pp.175-182
• /
• 2007

The support vector machine has been successful in many applications because of its flexibility and high accuracy. However, when a training data set is large or imbalanced, the support vector machine may suffer from significant computational problem or loss of accuracy in predicting minority classes. We propose a modified version of the support vector machine using the K-means clustering that exploits the information in class labels during the clustering process. For large data sets, our method can save the computation time by reducing the number of data points without significant loss of accuracy. Moreover, our method can deal with imbalanced data sets effectively by alleviating the influence of dominant class.

• The Korean Journal of Applied Statistics
• /
• v.35 no.4
• /
• pp.469-484
• /
• 2022

Since a large text corpus contains hundred-thousand unique words, text data is one of the typical large-dimensional data. Therefore, various feature selection methods have been proposed for dimension reduction. Feature selection methods can improve the prediction accuracy. In addition, with reduced data size, computational efficiency also can be achieved. The chi-square statistic and the information gain are two of the most popular measures for identifying interesting terms from text data. In this paper, we investigate the theoretical properties of the chi-square statistic and the information gain. We show that the two filtering metrics share theoretical properties such as non-negativity and convexity. However, they are different from each other in the sense that the information gain is prone to select more negative features than the chi-square statistic in imbalanced text data.

• Journal of Intelligence and Information Systems
• /
• v.19 no.2
• /
• pp.125-140
• /
• 2013

We call a data set in which the number of records belonging to a certain class far outnumbers the number of records belonging to the other class, 'imbalanced data set'. Most of the classification techniques perform poorly on imbalanced data sets. When we evaluate the performance of a certain classification technique, we need to measure not only 'accuracy' but also 'sensitivity' and 'specificity'. In a customer churn prediction problem, 'retention' records account for the majority class, and 'churn' records account for the minority class. Sensitivity measures the proportion of actual retentions which are correctly identified as such. Specificity measures the proportion of churns which are correctly identified as such. The poor performance of the classification techniques on imbalanced data sets is due to the low value of specificity. Many previous researches on imbalanced data sets employed 'oversampling' technique where members of the minority class are sampled more than those of the majority class in order to make a relatively balanced data set. When a classification model is constructed using this oversampled balanced data set, specificity can be improved but sensitivity will be decreased. In this research, we developed a hybrid model of support vector machine (SVM), artificial neural network (ANN) and decision tree, that improves specificity while maintaining sensitivity. We named this hybrid model 'hybrid SVM model.' The process of construction and prediction of our hybrid SVM model is as follows. By oversampling from the original imbalanced data set, a balanced data set is prepared. SVM_I model and ANN_I model are constructed using the imbalanced data set, and SVM_B model is constructed using the balanced data set. SVM_I model is superior in sensitivity and SVM_B model is superior in specificity. For a record on which both SVM_I model and SVM_B model make the same prediction, that prediction becomes the final solution. If they make different prediction, the final solution is determined by the discrimination rules obtained by ANN and decision tree. For a record on which SVM_I model and SVM_B model make different predictions, a decision tree model is constructed using ANN_I output value as input and actual retention or churn as target. We obtained the following two discrimination rules: 'IF ANN_I output value <0.285, THEN Final Solution = Retention' and 'IF ANN_I output value ${\geq}0.285$, THEN Final Solution = Churn.' The threshold 0.285 is the value optimized for the data used in this research. The result we present in this research is the structure or framework of our hybrid SVM model, not a specific threshold value such as 0.285. Therefore, the threshold value in the above discrimination rules can be changed to any value depending on the data. In order to evaluate the performance of our hybrid SVM model, we used the 'churn data set' in UCI Machine Learning Repository, that consists of 85% retention customers and 15% churn customers. Accuracy of the hybrid SVM model is 91.08% that is better than that of SVM_I model or SVM_B model. The points worth noticing here are its sensitivity, 95.02%, and specificity, 69.24%. The sensitivity of SVM_I model is 94.65%, and the specificity of SVM_B model is 67.00%. Therefore the hybrid SVM model developed in this research improves the specificity of SVM_B model while maintaining the sensitivity of SVM_I model.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.6
• /
• pp.1063-1068
• /
• 2016

Anomalous propagation echo is a kind of abnormal radar signal occurred by irregularly refracted radar beam caused by temperature or humidity. The echo frequently appears in ground-based weather radar due to its observation principle and disturb weather forecasting process. In order to improve accuracy of weather forecasting, it is important to analyze radar data precisely. Therefore, there are several ongoing researches about identifying the anomalous propagation echo with data mining techniques. This paper conducts researches about implementation of classification method which can separate the anomalous propagation echo in the raw radar data using naive Bayes classifier with various kinds of observation results. Considering that collected data has a class imbalanced problem, this paper includes SMOTE method. It is confirmed that the fine classification results are derived by the suggested classifier with balanced dataset using actual appearance cases of the echo.

• Journal of Korea Water Resources Association
• /
• v.54 no.12
• /
• pp.1305-1316
• /
• 2021

In this study, a new empirical formula for 2D transverse dispersion coefficient was developed using the results of previous tracer test studies, and the performance of the formula was evaluated. Since many tracer test studies have been conducted under the conditions where the width-to-depth ratio is less than 50, the existing empirical formulas developed using these imbalanced tracer test results have limitations in applying to rivers with a width-to-depth ratio greater than 50. Therefore, in order to develop an empirical formula for transverse dispersion coefficient using the imbalanced tracer test data, the Synthetic Minority Oversampling TEchnique (SMOTE) was used to oversample new data representing the properties of the existing tracer test data. The hydraulic data and the transverse dispersion coefficients in conditions of width-to-depth ratio greater than 50 were oversampled using the SMOTE. The reliability of the oversampled data was evaluated using the ROC (Receiver Operating Characteristic) curve. The empirical formula of transverse dispersion coefficient was developed including the oversampled data, and the performance of the results were compared with the empirical formulas suggested in previous studies using R². From the comparison results, the value of R² was 0.81 for the range of W/H < 50 and 0.92 for 50 < W/H, which were improved accuracy compared to the previous studies.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.7
• /
• pp.295-302
• /
• 2023

Software requirements written in natural language may have different meanings from the stakeholders' viewpoint. When designing an architecture based on quality attributes, it is necessary to accurately classify quality attribute requirements because the efficient design is possible only when appropriate architectural tactics for each quality attribute are selected. As a result, although many natural language processing models have been studied for the classification of requirements, which is a high-cost task, few topics improve classification performance with the imbalanced quality attribute datasets. In this study, we first show that the classification model can automatically classify the Korean requirement dataset through experiments. Based on these results, we explain that data augmentation through EDA(Easy Data Augmentation) techniques and undersampling strategies can improve the imbalance of quality attribute datasets, and show that they are effective in classifying requirements. The results improved by 5.24%p on F1-score, indicating that handling imbalanced data helps classify Korean requirements of classification models. Furthermore, detailed experiments of EDA illustrate operations that help improve classification performance.

• The Korean Journal of Applied Statistics
• /
• v.29 no.5
• /
• pp.961-975
• /
• 2016

The hierarchically penalized support vector machine (H-SVM) has been developed to perform simultaneous classification and input variable selection when input variables are naturally grouped or generated by factors. However, the H-SVM may suffer from estimation inefficiency because it applies the same amount of shrinkage to each variable without assessing its relative importance. In addition, when analyzing imbalanced data with uneven class sizes, the classification accuracy of the H-SVM may drop significantly in predicting minority class because its classifiers are undesirably biased toward the majority class. To remedy such problems, we propose the weighted adaptive H-SVM (WAH-SVM) method, which uses a adaptive tuning parameters to improve the performance of variable selection and the weights to differentiate the misclassification of data points between classes. Numerical results are presented to demonstrate the competitive performance of the proposed WAH-SVM over existing SVM methods.

• Journal of Korean Institute of Industrial Engineers
• /
• v.41 no.1
• /
• pp.25-33
• /
• 2015

From the viewpoint of applications to manufacturing, data mining is a useful method to find the meaningful knowledge or information about states of processes. But the data from manufacturing processes usually have two characteristics which are multicollinearity and imbalance distribution of data. Two characteristics are main causes which make bias to classification rules and select wrong variables as important variables. In the paper, we propose a new data mining procedure to solve the problem. First, to determine candidate variables, we propose the multiple hypothesis test. Second, to make unbiased classification rules, we propose the decision tree learning method with different weights for each category of quality variable. The experimental result with a real PDP (Plasma display panel) manufacturing data shows that the proposed procedure can make better information than other data mining procedures.

• Industrial Engineering and Management Systems
• /
• v.13 no.4
• /
• pp.454-462
• /
• 2014

Retention of possible churning customer is one of the most important issues in customer relationship management, so companies try to predict churn customers using their large-scale high-dimensional data. This study focuses on dealing with large data sets by reducing the dimensionality. By using six different dimension reduction methods-Principal Component Analysis (PCA), factor analysis (FA), locally linear embedding (LLE), local tangent space alignment (LTSA), locally preserving projections (LPP), and deep auto-encoder-our experiments apply each dimension reduction method to the training data, build a classification model using the mapped data and then measure the performance using hit rate to compare the dimension reduction methods. In the result, PCA shows good performance despite its simplicity, and the deep auto-encoder gives the best overall performance. These results can be explained by the characteristics of the churn prediction data that is highly correlated and overlapped over the classes. We also proposed a simple out-of-sample extension method for the nonlinear dimension reduction methods, LLE and LTSA, utilizing the characteristic of the data.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.25 no.1
• /
• pp.15-21
• /
• 2015

In support vector machine, the values of parameters included in kernels affect strongly generalization ability. It is often difficult to determine appropriate values of those parameters in advance. It has been observed through our studies that the burden for deciding the values of those parameters in support vector regression can be reduced by utilizing ensemble learning. However, the straightforward application of the method to large scale problems is too time consuming. In this paper, we propose a method in which the original data set is decomposed into a certain number of sub data set in order to reduce the burden for parameter tuning in support vector regression with large scale data sets and imbalanced data set, particularly.