• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.05a
• /
• pp.872-875
• /
• 2009

In a bankruptcy prediction model, the accuracy is one of crucial performance measures due to its significant economic impacts. Ensemble is one of widely used methods for improving the performance of classification and prediction models. Two popular ensemble methods, Bagging and Boosting, have been applied with great success to various machine learning problems using mostly decision trees as base classifiers. In this paper, we analyze the performance of boosted neural networks for improving the performance of traditional neural networks on bankruptcy prediction tasks. Experimental results on Korean firms indicated that the boosted neural networks showed the improved performance over traditional neural networks.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.31 no.7
• /
• pp.450-454
• /
• 2018

Because of the rapidly changing environment and high uncertainties, the semiconductor industry is in need of appropriate forecasting technology. In particular, both the cost and time in the test process are increasing because the process becomes complicated and there are more factors to consider. In this paper, we propose a prediction model that predicts a final "good" or "bad" on the basis of preconditioning test data generated in the semiconductor test process. The proposed prediction model solves the classification and regression problems that are often dealt with in the semiconductor process and constructs a reliable prediction model. We also implemented a prediction model through various machine learning algorithms. We compared the performance of the prediction models constructed through each algorithm. Actual data of the semiconductor test process was used for accurate prediction model construction and effective test verification.

• Nuclear Engineering and Technology
• /
• v.54 no.8
• /
• pp.3027-3033
• /
• 2022

Background: In this study, various types of deep-learning models for predicting in vitro radiosensitivity from gene-expression profiling were compared. Methods: The clonogenic surviving fractions at 2 Gy from previous publications and microarray gene-expression data from the National Cancer Institute-60 cell lines were used to measure the radiosensitivity. Seven different prediction models including three distinct multi-layered perceptrons (MLP), four different convolutional neural networks (CNN) were compared. Folded cross-validation was applied to train and evaluate model performance. The criteria for correct prediction were absolute error < 0.02 or relative error < 10%. The models were compared in terms of prediction accuracy, training time per epoch, training fluctuations, and required calculation resources. Results: The strength of MLP-based models was their fast initial convergence and short training time per epoch. They represented significantly different prediction accuracy depending on the model configuration. The CNN-based models showed relatively high prediction accuracy, low training fluctuations, and a relatively small increase in the memory requirement as the model deepens. Conclusion: Our findings suggest that a CNN-based model with moderate depth would be appropriate when the prediction accuracy is important, and a shallow MLP-based model can be recommended when either the training resources or time are limited.

• Korean Journal of Artificial Intelligence
• /
• v.12 no.2
• /
• pp.1-7
• /
• 2024

In this paper, we compare three models (logistic regression, Random Forest, and XGBoost) for predicting stroke occurrence using data from the Korea National Health and Nutrition Examination Survey (KNHANES). We evaluated these models using various metrics, focusing mainly on recall and F1 score to assess their performance. Initially, the logistic regression model showed a satisfactory recall score among the three models; however, it was excluded from further consideration because it did not meet the F1 score threshold, which was set at a minimum of 0.5. The F1 score is crucial as it considers both precision and recall, providing a balanced measure of a model's accuracy. Among the models that met the criteria, XGBoost showed the highest recall rate and showed excellent performance in stroke prediction. In particular, XGBoost shows strong performance not only in recall, but also in F1 score and AUC, so it should be considered the optimal algorithm for predicting stroke occurrence. This study determines that the performance of XGBoost is optimal in the field of stroke prediction.

• International Journal of Contents
• /
• v.9 no.4
• /
• pp.11-15
• /
• 2013

"Protein Folding Problem" is considered to be one of the "Great Challenges of Computer Science" and prediction of disordered protein is an important part of the protein folding problem. Machine learning models can predict the disordered structure of protein based on its characteristic of "learning from examples". Among many machine learning models, we investigate the possibility of multilayer perceptron (MLP) as the predictor of protein disorder. The investigation includes a single hidden layer MLP, multi hidden layer MLP and the hierarchical structure of MLP. Also, the target node cost function which deals with imbalanced data is used as training criteria of MLPs. Based on the investigation results, we insist that MLP should have deep architectures for performance improvement of protein disorder prediction.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.7
• /
• pp.307-314
• /
• 2022

Recently, researches using deep learning-based models are being actively conducted to replace statistical-based time series forecast techniques to predict electric power demand. The result of analyzing the researches shows that the performance of the LSTM-based prediction model is acceptable, but it is not sufficient for long-term regional-wide power demand prediction. In this paper, we propose a WaveNet deep learning model to predict electric power demand 24-hour-ahead with temperature data in order to achieve the prediction accuracy better than MAPE value of 2% which statistical-based time series forecast techniques can present. First of all, we illustrate a delated causal one-dimensional convolutional neural network architecture of WaveNet and the preprocessing mechanism of the input data of electric power demand and temperature. Second, we present the training process and walk forward validation with the modified WaveNet. The performance comparison results show that the prediction model with temperature data achieves MAPE value of 1.33%, which is better than MAPE Value (2.33%) of the same model without temperature data.

• Journal of the Society of Naval Architects of Korea
• /
• v.58 no.4
• /
• pp.234-242
• /
• 2021

In this study, we introduce the prediction of brake power for low-speed full ships and container carriers using the linear regression and a machine learning approach. The residual resistance coefficient, wake fraction coefficient, and thrust deduction factor are predicted by regression models using the main dimensions of ship and propeller. The brake power of a ship can be calculated by these coefficients according to the 1978 ITTC performance prediction method. The mean absolute error of the predicted power was under 7%. As a result of several validation cases, it was confirmed that the machine learning model showed slightly better results than linear regression.

• Geomechanics and Engineering
• /
• v.38 no.5
• /
• pp.507-515
• /
• 2024

An accurate estimation of the geotechnical parameters in front of tunnel faces is crucial for the safe construction of underground infrastructure using tunnel boring machines (TBMs). This study was aimed at developing a data-driven model for predicting the rock quality designation (RQD) of the ground formation ahead of tunnel faces. The dataset used for the machine learning (ML) model comprises seven geological and mechanical features and 564 RQD values, obtained from an earth pressure balance (EPB) shield TBM tunneling project beneath the Han River in the Republic of Korea. Four ML algorithms were employed in developing the RQD prediction model: k-nearest neighbor (KNN), support vector regression (SVR), random forest (RF), and extreme gradient boosting (XGB). The grid search and five-fold cross-validation techniques were applied to optimize the prediction performance of the developed model by identifying the optimal hyperparameter combinations. The prediction results revealed that the RF algorithm-based model exhibited superior performance, achieving a root mean square error of 7.38% and coefficient of determination of 0.81. In addition, the Shapley additive explanations (SHAP) approach was adopted to determine the most relevant features, thereby enhancing the interpretability and reliability of the developed model with the RF algorithm. It was concluded that the developed model can successfully predict the RQD of the ground formation ahead of tunnel faces, contributing to safe and efficient tunnel excavation.

• Journal of Broadcast Engineering
• /
• v.27 no.3
• /
• pp.332-340
• /
• 2022

Recently, with the development of deep learning and artificial neural network technologies, research on the application of neural network has been actively conducted in the field of video coding. In particular, deep learning-based intra prediction is being studied as a way to overcome the performance limitations of the existing intra prediction techniques. This paper presents a method of context-adaptive neural network-based intra prediction model training and its coding performance analysis. In other words, in this paper, we implement and train a known intra prediction model based on convolutional neural network (CNN) that predicts a current block using contextual information from reference blocks. Then, we integrate the trained model into HM16.19 as an additional intra prediction mode and evaluate the coding performance of the trained model. Experimental results show that the trained model gives 0.28% BD-rate bit saving over HEVC in All Intra (AI) coding mode. In addition, the coding performance change of training considering block partition is also presented.

• IEIE Transactions on Smart Processing and Computing
• /
• v.6 no.3
• /
• pp.183-192
• /
• 2017

k-nearest neighbor (K-NN) is a well-known classification algorithm, being feature space-based on nearest-neighbor training examples in machine learning. However, K-NN, as we know, is a lazy learning method. Therefore, if a K-NN-based system very much depends on a huge amount of history data to achieve an accurate prediction result for a particular task, it gradually faces a processing-time performance-degradation problem. We have noticed that many researchers usually contemplate only classification accuracy. But estimation speed also plays an essential role in real-time prediction systems. To compensate for this weakness, this paper proposes correlation coefficient-based clustering (CCC) aimed at upgrading the performance of K-NN by leveraging processing-time speed and plurality rule-based density (PRD) to improve estimation accuracy. For experiments, we used real datasets (on breast cancer, breast tissue, heart, and the iris) from the University of California, Irvine (UCI) machine learning repository. Moreover, real traffic data collected from Ojana Junction, Route 58, Okinawa, Japan, was also utilized to lay bare the efficiency of this method. By using these datasets, we proved better processing-time performance with the new approach by comparing it with classical K-NN. Besides, via experiments on real-world datasets, we compared the prediction accuracy of our approach with density peaks clustering based on K-NN and principal component analysis (DPC-KNN-PCA).