• Communications for Statistical Applications and Methods
• /
• v.30 no.2
• /
• pp.215-226
• /
• 2023

Deep learning has made great strides in the field of unstructured data such as text, images, and audio. However, in the case of tabular data analysis, machine learning algorithms such as ensemble methods are still better than deep learning. To keep up with the performance of machine learning algorithms with good predictive power, several deep learning methods for tabular data have been proposed recently. In this paper, we review the latest deep learning models for tabular data and compare the performances of these models using several datasets. In addition, we also compare the latest boosting methods to these deep learning methods and suggest the guidelines to the users, who analyze tabular datasets. In regression, machine learning methods are better than deep learning methods. But for the classification problems, deep learning methods perform better than the machine learning methods in some cases.

• Journal of Intelligence and Information Systems
• /
• v.26 no.2
• /
• pp.105-129
• /
• 2020

This study uses corporate data from 2012 to 2018 when K-IFRS was applied in earnest to predict default risks. The data used in the analysis totaled 10,545 rows, consisting of 160 columns including 38 in the statement of financial position, 26 in the statement of comprehensive income, 11 in the statement of cash flows, and 76 in the index of financial ratios. Unlike most previous prior studies used the default event as the basis for learning about default risk, this study calculated default risk using the market capitalization and stock price volatility of each company based on the Merton model. Through this, it was able to solve the problem of data imbalance due to the scarcity of default events, which had been pointed out as the limitation of the existing methodology, and the problem of reflecting the difference in default risk that exists within ordinary companies. Because learning was conducted only by using corporate information available to unlisted companies, default risks of unlisted companies without stock price information can be appropriately derived. Through this, it can provide stable default risk assessment services to unlisted companies that are difficult to determine proper default risk with traditional credit rating models such as small and medium-sized companies and startups. Although there has been an active study of predicting corporate default risks using machine learning recently, model bias issues exist because most studies are making predictions based on a single model. Stable and reliable valuation methodology is required for the calculation of default risk, given that the entity's default risk information is very widely utilized in the market and the sensitivity to the difference in default risk is high. Also, Strict standards are also required for methods of calculation. The credit rating method stipulated by the Financial Services Commission in the Financial Investment Regulations calls for the preparation of evaluation methods, including verification of the adequacy of evaluation methods, in consideration of past statistical data and experiences on credit ratings and changes in future market conditions. This study allowed the reduction of individual models' bias by utilizing stacking ensemble techniques that synthesize various machine learning models. This allows us to capture complex nonlinear relationships between default risk and various corporate information and maximize the advantages of machine learning-based default risk prediction models that take less time to calculate. To calculate forecasts by sub model to be used as input data for the Stacking Ensemble model, training data were divided into seven pieces, and sub-models were trained in a divided set to produce forecasts. To compare the predictive power of the Stacking Ensemble model, Random Forest, MLP, and CNN models were trained with full training data, then the predictive power of each model was verified on the test set. The analysis showed that the Stacking Ensemble model exceeded the predictive power of the Random Forest model, which had the best performance on a single model. Next, to check for statistically significant differences between the Stacking Ensemble model and the forecasts for each individual model, the Pair between the Stacking Ensemble model and each individual model was constructed. Because the results of the Shapiro-wilk normality test also showed that all Pair did not follow normality, Using the nonparametric method wilcoxon rank sum test, we checked whether the two model forecasts that make up the Pair showed statistically significant differences. The analysis showed that the forecasts of the Staging Ensemble model showed statistically significant differences from those of the MLP model and CNN model. In addition, this study can provide a methodology that allows existing credit rating agencies to apply machine learning-based bankruptcy risk prediction methodologies, given that traditional credit rating models can also be reflected as sub-models to calculate the final default probability. Also, the Stacking Ensemble techniques proposed in this study can help design to meet the requirements of the Financial Investment Business Regulations through the combination of various sub-models. We hope that this research will be used as a resource to increase practical use by overcoming and improving the limitations of existing machine learning-based models.

• The Korean Journal of Nuclear Medicine
• /
• v.38 no.6
• /
• pp.486-491
• /
• 2004

Purpose: factor analysis and independent component analysis (ICA) has been used for handling dynamic image sequences. Theoretical advantages of a newly suggested ICA method, ensemble ICA, leaded us to consider applying this method to the analysis of dynamic myocardial $H_2^{15}O$ PET data. In this study, we quantified patients' blood flow using the ensemble ICA method. Materials and Methods: Twenty subjects underwent $H_2^{15}O$ PET scans using ECAT EXACT 47 scanner and myocardial perfusion SPECT using Vertex scanner. After transmission scanning, dynamic emission scans were initiated simultaneously with the injection of $555{\sim}740$ MBq $H_2^{15}O$. Hidden independent components can be extracted from the observed mixed data (PET image) by means of ICA algorithms. Ensemble learning is a variational Bayesian method that provides an analytical approximation to the parameter posterior using a tractable distribution. Variational approximation forms a lower bound on the ensemble likelihood and the maximization of the lower bound is achieved through minimizing the Kullback-Leibler divergence between the true posterior and the variational posterior. In this study, posterior pdf was approximated by a rectified Gaussian distribution to incorporate non-negativity constraint, which is suitable to dynamic images in nuclear medicine. Blood flow was measured in 9 regions - apex, four areas in mid wall, and four areas in base wall. Myocardial perfusion SPECT score and angiography results were compared with the regional blood flow. Results: Major cardiac components were separated successfully by the ensemble ICA method and blood flow could be estimated in 15 among 20 patients. Mean myocardial blood flow was $1.2{\pm}0.40$ ml/min/g in rest, $1.85{\pm}1.12$ ml/min/g in stress state. Blood flow values obtained by an operator in two different occasion were highly correlated (r=0.99). In myocardium component image, the image contrast between left ventricle and myocardium was 1:2.7 in average. Perfusion reserve was significantly different between the regions with and without stenosis detected by the coronary angiography (P<0.01). In 66 segment with stenosis confirmed by angiography, the segments with reversible perfusion decrease in perfusion SPECT showed lower perfusion reserve values in $H_2^{15}O$ PET. Conclusions: Myocardial blood flow could be estimated using an ICA method with ensemble learning. We suggest that the ensemble ICA incorporating non-negative constraint is a feasible method to handle dynamic image sequence obtained by the nuclear medicine techniques.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.8
• /
• pp.97-105
• /
• 2015

Content-adaptive steganography methods embed secret messages in hard-to-model regions of covers such as complicated texture or noisy area. Content-adaptive steganalysis methods often need high dimensional features to capture more subtle relationships of local dependencies among adjacent pixels. However, these methods require many computational complexity and depend on the location of hidden message and the exploited distortion metrics. In this paper, we propose an improved steganalysis method for content-adaptive steganography to enhance detection rate with small number features. We first show that the features form the difference between DCT coefficients are useful for analyzing the content-adaptive steganography methods, and present feature extraction mehtod using first-order Markov probability for the the difference between DCT coefficients. The extracted features are used as input of ensemble classifier. Experimental results show that the proposed method outperforms previous schemes in terms of detection rates and accuracy in spite of a small number features in various content-adaptive stego images.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.35 no.2
• /
• pp.327-340
• /
• 2015

General Circulation Models (GCMs) are the basic tool used for modelling climate. However, the spatio-temporal discrepancy between GCM and observed value, therefore, the models deliver output that are generally required calibration for applied studies. Which is generally done by Multi-Model Ensemble (MME) approach. Stochastic downscaling methods have been used extensively to generate long-term weather sequences from finite observed records. A primary objective of this study is to develop a forecasting scheme which is able to make use of a MME of different GCMs. This study employed a Nonstationary Hidden Markov Chain Model (NHMM) as a main tool for downscaling seasonal ensemble forecasts over 3 month period, providing daily forecasts. Our results showed that the proposed downscaling scheme can provide the skillful forecasts as inputs for hydrologic modeling, which in turn may improve water resources management. An application to the Nakdong watershed in South Korea illustrates how the proposed approach can lead to potentially reliable information for water resources management.

• Journal of the Korea Society for Simulation
• /
• v.20 no.2
• /
• pp.77-85
• /
• 2011

It is believed that the anomalies or diseases of human organs are identified by the analysis of the patterns. This paper proposes a new classification technique for the identification of cancer disease using the proteome patterns obtained from two-dimensional polyacrylamide gel electrophoresis(2-D PAGE). In the new classification method, three different classification methods such as support vector machine(SVM), multi-layer perceptron(MLP) and k-nearest neighbor(k-NN) are extended by multi-boosting method in an array of subclassifiers and the results of each subclassifier are merged by ensemble method. Genetic algorithm was applied to obtain optimal feature set in each subclassifier. We applied our method to empirical data set from cancer research and the method showed the better accuracy and more stable performance than single classifier.

• Journal of KIISE:Software and Applications
• /
• v.31 no.4
• /
• pp.525-536
• /
• 2004

Gene expression profile is numerical data of gene expression level from organism, measured on the microarray. Generally, each specific tissue indicates different expression levels in related genes, so that we can classify disease with gene expression profile. Because all genes are not related to disease, it is needed to select related genes that is called feature selection, and it is needed to classify selected genes properly. This paper Proposes GA based method for searching optimal ensemble of feature-classifier pairs that are composed with seven feature selection methods based on correlation, similarity, and information theory, and six representative classifiers. In experimental results with leave-one-out cross validation on two gene expression Profiles related to cancers, we can find ensembles that produce much superior to all individual feature-classifier fairs for Lymphoma dataset and Colon dataset.

• Journal of Digital Convergence
• /
• v.17 no.1
• /
• pp.123-130
• /
• 2019

One of the most commonly used methods of web recommendation techniques is collaborative filtering. Many studies on collaborative filtering have suggested ways to improve accuracy. This study proposes a method of movie recommendation using Word2Vec and an ensemble convolutional neural networks. First, in the user, movie, and rating information, construct the user sentences and movie sentences. It inputs user sentences and movie sentences into Word2Vec to obtain user vectors and movie vectors. User vectors are entered into user convolution model and movie vectors are input to movie convolution model. The user and the movie convolution models are linked to a fully connected neural network model. Finally, the output layer of the fully connected neural network outputs forecasts of user movie ratings. Experimentation results showed that the accuracy of the technique proposed in this study accuracy of conventional collaborative filtering techniques was improved compared to those of conventional collaborative filtering technique and the technique using Word2Vec and deep neural networks proposed in a similar study.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.4
• /
• pp.603-610
• /
• 2021

In this paper, we extract the user's power consumption patterns, and model the optimal consumption patterns by applying the user's environment and emotion. Based on the comparative analysis of these two patterns, we present an efficient power consumption method through changes in the user's power consumption behavior. To extract significant consumption patterns, vector standardization and binary data transformation methods are used, and learning about the ensemble's ensemble with k-means clustering is applied, and applying the support factor according to the value of k. The optimal power consumption pattern model is generated by applying forced and emotion-based control based on the learning results for ensemble aggregates with relatively low average consumption. Through experiments, we validate that it can be applied to a variety of windows through the number or size adjustment of clusters to enable forced and emotion-based control according to the user's intentions by identifying the correlation between the number of clusters and the consistency ratios.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.7
• /
• pp.972-980
• /
• 2022

The butterfly species recognition technology based on machine learning using images has the effect of reducing a lot of time and cost of those involved in the related field to understand the diversity, number, and habitat distribution of butterfly species. In order to improve the accuracy and time efficiency of butterfly species classification, various features used as the inputs of machine learning models have been studied. Among them, branch length similarity(BLS) entropy or color intensity entropy methods using the concept of entropy showed higher accuracy and shorter learning time than other features such as Fourier transform or wavelet. This paper proposes a feature extraction algorithm using RGB color intensity entropy for butterfly color images. In addition, we develop butterfly recognition systems that combines the proposed feature extraction method with representative ensemble models and evaluate their performance.