Eun-Jin Kim

The Korea Superconducting Tokamak Advanced Research (KSTAR) device has recently entered a new phase of operation following the installation of a tungsten mono-block divertor, providing a unique platform to investigate high-performance, long-pulse plasma scenarios for future reactors. The 2023–2024 experimental campaigns demonstrated significant progress in commissioning the tungsten divertor, developing plasma operation scenarios, and achieving real-time control. Comparative studies revealed increased core radiation losses and associated performance degradation relative to the carbon divertor environment. Alleviation strategies, such as optimized neutral beam injection timing, boron powder seeding, and impurity control through gas puffing, were shown to reduce tungsten accumulation and extend high-performance phases, including record H-mode operation exceeding 100 s. International collaboration enabled the first demonstration of high poloidal beta scenarios with an internal transport barrier on KSTAR. Advances in control included upgraded real-time resonant magnetic perturbation schemes and disruption forecasting via the DECAF system. Furthermore, divertor detachment control using a new surrogate-model-based approach with real-time radiation imaging demonstrated active impurity and heat flux management. Supported by enhanced heating, current drive, and diagnostic systems, these achievements offer valuable insights into tungsten impurity behavior, transport physics, and control strategies for future reactors including ITER. Planned upgrades toward a full-tungsten wall and ITER-aligned real-time plasma control will further establish KSTAR as a leading experimental platform for developing operational scenarios required to achieve steady-state, burning plasma conditions.

Spatiotemporal techniques are increasingly integral to football analytics for assessing team performance, yet the inherently dynamic and interactive nature of association football continues to hinder fully objective evaluation. Although existing research has examined the randomness of match events and patterns of player interactions, the contribution of ball movement trajectories has received comparatively limited attention despite its tactical significance. This study addresses this shortcoming by introducing a novel framework to quantify the spatial complexity of team ball movement as an indicator of offensive performance. Using a time-series feature extraction methodology, the fractal dimension of two-dimensional ball trajectory representations is calculated over predefined temporal windows to capture spatial complexity. The findings indicate a positive relationship between higher spatial complexity and match success, particularly in the early stages of play. In addition, a multi-scale fractal analysis reveals scale-dependent effects, with coarse- and mid-scale spatial complexity showing positive correlations with match-winning outcomes, while fine-scale complexity exhibits an inverse relationship, indicating that spatial structure at broader scales may be more influential for effective team performance than high-frequency local variations in ball movement. Finally, a Random Forest classification model trained solely on spatial complexity features achieved an AUC–ROC score of 0.8180 in predicting match outcomes, demonstrating that spatial complexity serves as an informative, interpretable, and effective time-series metric for evaluating team performance in association football.

Modern football prioritizes team play and tactical strategies over individual brilliance. However, its low-scoring nature makes evaluating team performance challenging. Unpredictable ball movement enhances offensive play while complicating defensive setups. To better capture this dynamic nature, authors’ prior work has proposed entropy-based time-series metric to assess unpredictable ball movement by quantifying Spatial Event Distribution Randomness (EDRan). However, some teams may prefer to dominate specific areas with unpredictability, while others utilize the entire field. Existing literature has not examined whether emphasizing dominant (frequently used field regions for ball movement) or considering all regions equally, including rarely used areas, is a more effective approach for computing randomness in event distribution. Moreover, existing research has not investigated the underlying patterns of event distribution randomness, particularly how these variations differ between winning and losing teams, both in terms of overall field coverage and concentration within dominant regions. This study addresses these gaps by analyzing event distribution randomness using Rényi entropy with varying alpha values ( 0 ≤ α ≤ 2 0).Correlation analysis indicated that assigning equal weight to all field regions, including rarely used areas, with Max entropy ( α = 0 alpha) was most strongly associated with match-winning performance. In men’s data, machine learning models trained with α = 0 , 0.1 , alpha and 0.5 achieved statistically significant improvements over models trained with the traditionally used Shannon entropy ( α → 1 alpha). These results suggest that unpredictability distributed across the entire field, maximizing the use of diverse regions, is more strongly associated with success than randomness restricted to dominant areas. The best-performing model, obtained with α = 0 alpha, significantly outperformed both the baseline and existing models in the literature, achieving an accuracy of 80.61% in predicting match winners.

Purpose: Unpredictable ball movements and consistent utilization of space are crucial in association football (soccer) for outsmarting opponents and maintaining control throughout a match. Given the sport's low-scoring and unpredictable nature, consistent team performance from start to end is vital, as a single goal can often determine the outcome. In the author's prior work, event distribution randomness (EDRan) was explored for its positive contribution to match-winning performances (Bandara, 2024). This study further extends that investigation by examining the relationship between in-game variations (inconsistency) in event distribution randomness across different phases of a match and a team's chances of success. To support this analysis, the study introduces a temporal performance evaluation metric, EDRanIc (EDRan Inconsistency), designed to capture variations in the randomness of ball movement over time. Design/methodology/approach: This analysis utilized a publicly available event-log dataset containing spatiotemporal data from top-tier club competitions and international tournaments. The proposed approach involved generating “region-based cumulative possession” matrices for defined time periods within a game to obtain probability distributions of event distributions. The spatial event distribution randomness of each team was compared across successive time periods using relative entropy measures, which were then used as temporal features in developing a match-winner prediction model. Findings: A machine learning model trained on the extracted temporal features achieved 79% accuracy in predicting match winners and demonstrated a negative correlation between match-winning performances and inconsistency in EDRan between time periods (EDRanIc). This underscores the significance of maintaining consistency in event distribution randomness throughout a game for achieving match-winning outcomes. The analysis further revealed that winning teams maintained greater EDRan consistency during the early stages of the game. As the game progressed, this consistency declined, reflecting a shift toward more defensive strategies commonly adopted by leading teams in the latter stages of matches. Originality/value: In summary, this study highlights the importance of maintaining consistency in event distribution randomness for match-winning performances and proposes an approach to evaluate this EDRan consistency performance using relative entropy measures. The findings provide valuable insights into the tactical aspects of football, demonstrating how maintaining consistency in event distribution randomness contributes to match-winning performances. Additionally, the proposed team performance evaluation metric (EDRanIc) can assist team management in assessing and identifying inconsistencies in their team's ball movement.

We investigate the turbulence statistics associated with low-to-high confinement (L-H) transitions and externally applied resonant magnetic perturbations (RMPs) in KSTAR. Time-series fluctuations of electron density ne (BES), electron temperature Te (ECEI), and the time derivative of the poloidal magnetic field dBθ/dt (Mirnov coils) are analysed using information-geometric measures (information rate Γ and information length L = R Γdt), together with kurtosis κ and variance σ2. In low-density upper single-null plasmas (ne ∼ 1.2 × 1019 m−3), a ∼ 80 kHz magnetic mode coupling ne, Te, and dBθ/dt emerges prior to the L-H transition and persists into the ELMy H-mode. Edge-localised RMPs (ERMPs) suppress this coherent mode but enhance intermittency, producing frequent bursts that abruptly reshape the time- dependent Probability Density Functions (PDFs) and generate large spikes in Γ (with smaller changes in κ), signalling ERMP-driven departures from quasi-stationarity. The impact of ERMPs on background fluctuation levels depends on density, radial location, and the fluctuating variable itself (n ̃, T ̃, B ̇θ), whereas L provides a robust, regime- agnostic measure of cumulative statistical reorganisation and spatial decorrelation. In particular, at low density we observe weaker coupling between n ̃ and T ̃, along with a tendency toward decreased radial correlation—most clearly for T ̃—under ERMPs. Overall, information geometry cleanly captures intermittent events, quantifies non- equilibrium PDF evolution, and offers a compact, cross-diagnostic metric for assessing RMP effects on edge transport and correlation across densities, radial locations, and confinement states.

The first experimental results of self-regulation between the plasma relative electron density fluctuations, n ~ e , and the perpendicular velocity, v ~ ⊥ , over limit cycle oscillations (LCOs) across the edge and near scrape-off layer (SOL) regions of the MAST-U spherical tokamak are presented. The dithering phase is preceded by the localisation of the emissivity at the magnetic null point. The L–H-mode-like transitions of the LCOs are characterised by a decrease in the mean n ~ e and a narrowing of the probability density functions (PDFs) due to plasma turbulence suppression. Sharp decreases in the n ~ e standard deviation and skewness routinely coincide with the start of the LCO in a narrow region corresponding to the steepest pedestal pressure and v ~ ⊥ gradients. The decrease in n ~ e skewness during the LCOs in the regions closer to the SOL, separatrix, and core plasma follows 2–5 ms later. The reduction in n ~ e is always reflected in changes in the PDF structure; the mean perpendicular velocity consistently shifts from positive to negative values following the L–H-mode-like transitions. The most significant changes in the mean v ~ ⊥ PDFs are in the region of the maximum negative v ~ ⊥ . Further evidence of the spatially localised nature of self-regulation between n ~ e and v ~ ⊥ PDFs is provided from information geometry analysis for the formation and collapse of the ETB.

Dementia, including Alzheimer’s disease and frontotemporal dementia, is a progressive brain disorder that disrupts memory, thinking, and behavior, with early diagnosis being critical for effective intervention. This study examines the alteration of brain activity caused by dementia by analyzing electroencephalogram (EEG) signals using an information geometry method known as information rate, which captures the evolving patterns of brain signals over time rather than relying on static averages. This method is applied across standard EEG frequency bands – delta, theta, alpha, beta, and gamma – in participants with dementia and healthy controls. The characteristics of the distribution of information rate are studied through the statistical moments (such as mean, variance, skewness, and kurtosis) and Shannon entropy. The statistical comparisons are accessed using the Kruskal-Wallis test with Dunn’s post-hoc analysis, and results are compared against a conventional average-base method using Jensen-Shannon distance. The results show that dynamic features of EEG signals – particularly in the theta, alpha, and beta bands – effectively distinguish Alzheimer’s patients from healthy individuals, while the Shannon entropy of signal dynamics in frontal region differentiates frontotemporal dementia patients across the theta to gamma bands. Moreover, changes in the occipital region detected by information rate, but not by traditional method, further highlight the importance of capturing temporal variability. The method also successfully distinguishes individuals with Mild Cognitive Impairment from healthy controls, which conventional analysis failed to achieve. These results suggest that analyzing the dynamics properties of the brain signals provides a more sensitive and informative approach for identifying and distinguishing various forms of dementia.

Astrophysical and fusion plasmas share significant similarities, particularly in their ubiquitous turbulence, coherent structures, and self-organization. This paper focuses on magnetic confinement fusion plasmas, emphasizing their inherently non-equilibrium nature and the use of non-perturbative statistical approaches to quantify them. The statistical properties of fusion plasmas often deviate from Gaussian distributions, rendering low-order moments—such as means and standard deviations—inadequate for fully characterizing turbulence and its impact. The low-to-high confinement (L–H) transition, a key plasma bifurcation leading to improved confinement, is examined as a stochastic bifurcation, where the transition occurs probabilistically for a given input power. Probability density function methods help reveal how hidden variables influence the power threshold. Additionally, information theory is employed to uncover nonlinear plasma interactions, including self-regulation and causality.

Assessing team performance in association football (commonly known as football or soccer) is challenging due to the sport’s low-scoring nature and inherent unpredictability. While evaluating strategies based on space control and the creation of open spaces has been explored in the literature, the temporal aspect of space availability for the ball carrier remains under-explored. This work introduces a novel time-series performance evaluation metric, Ball Carrier Open Space ( BCOS ), which focuses on the temporal dynamics of space available to the ball carrier to assess team performance. Additionally, it presents a novel approach to quantify open space for the ball carrier using player data extracted from television footage. This work discuss on BCOS in defensive third, central third and attacking third and a machine learning model is developed to evaluate their significance and temporal patterns. Trained model achieved 80.7% accuracy in classifying match-winning performances, underscoring the significance of BCOS . Correlation analysis between temporal features and match outcomes further reveals that BCOS in central third and attacking third are more important for match winning outcomes, while first-half performance plays a more critical role in determining match results than second-half performance. Based on the results of the correlation analysis, this study proposes a weighted ball carrier open space ( wBCOS ) metric to assess team performance, assigning weights to BCOS in attacking third, central third and defensive third based on their contributions to positive match outcomes. A machine learning model trained using wBCOS achieved an 82.5% accuracy in classifying match-winning performances, surpassing the performance of any previously published match-winner classification model.