Jo Oct 20, 2026
Dynamic time-warping is a method of matching by computing similarity using the elastic distance between two time series. This method was developed for speech recognition and it has been applied to many areas.
This matching method makes one time series to be compressed and stretched nonlinearly in the time axis direction based on the dynamic programming to minimize the elastic distance between two time series of different lengths.
In the literature, matching is implemented by changing the scale of time axis in the time series.
Jong Sung Il, a lecturer at the Faculty of Applied Mathematics, has proposed a time-phase scale varying DTW method of implementing matching by varying the scale of phase as well as that of time axis. Then, he applied it to a sine function with varying amplitude to verify its effectiveness.
For further details, please refer to his paper “Pattern Detection Method using Phase Varying Dynamic Time Warping in Time Series” in “Proceedings of KUTIC-2025”.
...
Jo Oct 19, 2026
In recent years, with the rapid popularization of intelligent terminals and sensors, massive data have been accumulated rapidly, and the processing technology of big data has attracted more and more attention. Among them, large information data with functional characteristics such as physiological data, growth curve data, temperature data, etc. are also attracting significant attention. In fact, these data are discrete samples of continuous functions, and these data are usually known as functional data.
The main problem in functional data analysis is dimensionality reduction, and one of the common ways to reduce dimension is to take only a few principal components by performing functional principal component analysis (PCA) on given functional data. The conventional functional principal component analysis method uses the L2-norm, which has the advantage of being computationally convenient, but it is not efficient for data with outliers because the L2-norm is highly affected by outliers. It can be expected that the PCA method using the L1-norm instead of the L2-norm will have better robustness. Without considering the computational complexity when using the L1-norm, high computational cost of functional data analysis cannot be avoided.
Paek Hyon I, a researcher at the Faculty of Applied Mathematics, proposed a method for improving robustness by using L1-norm instead of L2-norm in the functional principal component analysis for functional data with vertical outliers, and constructed an efficient computational algorithm to reduce computational complexity.
This algorithm can achieve more efficient functional principal components with high robustness from data contaminated by vertical outliers while reducing the computational complexity of functional principal component analysis.
It was demonstrated through modeling studies that this method is more robust to outliers than L2-norm functional principal component analysis.
You can find details in his paper “L1-norm Principal Component Analysis Method for Functional Data with Vertical Outliers” in “Proceedings of KUTIC-2025”.
...
Jo Oct 18, 2026
Kim Ung, a section head at the Faculty of Applied Mathematics, has studied a deterministic infinite horizon, a mixed continuous and impulse control problem with unbounded cost functional.
Major difficulties arise from the unboundedness of value functions.
He proved the comparison result of viscosity solution by converting unbounded value functions into bounded ones by suitable transformation and proved that the value function is the unique viscosity solution of the related first order Hamilton-Jacobi quasi-variational inequality (QVI). He constructed a time-discretization scheme for the QVI and proved that an approximate value function exists and it is the unique solution of the approximate QVI.
He also proved that the solution of the time discrete scheme converges to the viscosity solution of the QVI, when the discretization step goes to zero, and that the optimal control of the time discrete system determined by the corresponding dynamic programming is a minimizing sequence of the optimal feedback control of the impulse control problem.
You can find more information in his paper “Approximation Schemes for Impulse Control Systems with Unbounded Cost Functional” in “Proceedings of KUTIC-2025”.
...
Jo Oct 16, 2026
The use of ozone gas started about a century ago. Ozone has been used to purify water polluted by microorganisms. The chemistry of ozone in aqueous solutions is complex and the direct reaction with molecular ozone and the indirect reaction with radicals generated when ozone is dissolved in water are considered for the process of water treatment with ozone.
The authors calculated the first principles of the reaction between molecular ozone and water vapour, but no first principles study of the decomposition of ozone in water has been carried out.
Choe Un Hwa, a researcher at the Faculty of Online Education, has conducted a study to select the size of ozone bubbles used for water treatment using ozone microbubbles, considering the ozonolysis characteristics in water.
She quantitatively calculated the decomposition of ozone in water using the density functional method (DFT/6-31G*). Among the four possible reactions between O3 and H2O in gas and water, the possibility of obtaining H2O2+O2 is relatively high.
Considering that H2O2 and O2 are products of indirect reactions by ozone with OH- as an initiator and the rate of indirect reactions is very fast, the size of ozone bubbles used for water treatment can be as large as a micron size, which can increase the solubility of ozone gas in aqueous solution and improve the ozonation efficiency.
You can find more information in her paper “Ozone Micro Bubble Treatment: What is the Size of Ozone Bubble Suitable for Water Treatment?” in “Proceedings of KUTIC-2025”.
...
Jo Oct 15, 2026
Femoral head osteonecrosis is a disease affecting patients at the age of 30-50 that usually progresses to femoral head collapse and often leads to total hip arthroplasty. Femoral head osteonecrosis frequently progresses to the subchondral fracture of femoral head, eventually leading to complete joint destruction.
Joint-preserving surgical interventions are generally more successful at earlier stages of bone involvement than after the occurrence of subchondral fracture. It means that the optimal time to evaluate and treat osteonecrosis of femoral head is before the occurrence of subchondral fracture. Early diagnosis can facilitate the application of interventions that can avoid or delay arthroplasty to some extent. However, misdiagnoses and delayed diagnoses are common due to the lack of preliminary symptoms, typical features, and internal fixation interference on radiographs.
Kim Ju Hyang, a researcher at the Faculty of Information Science and Technology, investigated the problem of extracting significant features for detecting bone diseases such as avascular necrosis and osteoporosis of femoral head from radiographic images of femoral head, and classifiers. On this basis, he proposed a method for early diagnosis of femoral head osteonecrosis by feature extraction and classifiers.
She performed feature extraction by 2D Dual tree complex wavelet transform, and early detection by support vector machine (SVM). Using 2D Dual tree complex wavelet transform on femoral radiographs, she extracted high-frequency component features based on six-way complex wavelet coefficients of femoral head images. Then, she selected the most significant features for classification among the spatial domain features and other features significant for avascular necrosis detection to construct feature vectors and designed a support vector machine classifier.
The experimental results show that the proposed method is cost-effective due to the use of X-ray images and has excellent detection performance.
If further information is needed, you can refer to her paper “Dual Tree Complex Wavelet Transform-Based X-Ray Image Feature Extraction and Early Diagnosis of Femoral Head Osteonecrosis by Support Vector Machine” in “Proceedings of KUTIC-2025”.
...
Jo Oct 14, 2026
The annular centrifugal extractors (ACEs) are based on the principle of Taylor-vortex flow. They essentially consist of two co-axial cylinders where the outer one is stationary and the inner one is rotating. They provide very high values of mass transfer coefficient in the annular region because of high levels of power consumption per unit volume. What is more, the rotor separates two phases into the other without any carryover of one phase. With these two advantages, they are used for several applications of liquid-liquid extraction where low inventories are desired. They include metal recovery and purification in the mining industry, processing of uranium and high plutonium fuels, caustic side solvent extraction to remove cesium from nuclear waste, hydrochloric washing of polymer phases, bromination and subsequent efficient separation of brominated polymer from aqueous phase.
Ri Hui Myong, a researcher at the Faculty of Thermal Engineering, investigated the phase separation characteristics and extraction characteristics of TBP-NaNO3 system in a 50mm annular centrifugal extractor with respect to the rotational speed and the inlet phase ratio, and derived an empirical formula for the throughput of NaNO3 under a given operating condition.
The maximum throughput of TBP-NaNO3 system in the 50mm annular centrifugal extractor was 12.6kg/h at a rotational speed of N=2 500r/min and an inlet phase ratio of Ro/a=1.
You can find more information in his paper “Phase Separation and Extraction Characteristics of TBP-NaNO3 System in an Annular Centrifugal Extractor” in “Proceedings of KUTIC-2025”.
...