Background: Real-time ergonomic risk assessment in manufacturing environments is challenged by severe class imbalance in high-risk postures and the need for deployment-efficient models. Conventional oversampling techniques may violate biomechanical constraints, limiting their suitability for human motion data. Objectives: This study aimed to compare multiple machine learning models for real-time ergonomic risk assessment while addressing data imbalance using biomechanically appropriate learning strategies and evaluating both predictive performance and deployment efficiency. Design: Comparative study. Methods: A large-scale workplace safety dataset comprising image-based skeletal keypoints was analyzed. To mitigate class imbalance without generating biomechanically implausible samples, cost-sensitive learning and focal loss were employed instead of synthetic oversampling. Subject-wise data splitting was applied to prevent data leakage. Five model families, including Random Forest, convolutional neural networks, and a lightweight graphbased network, were evaluated using accuracy, F1-score, area under the receiver operating characteristic curve (AUC), and high-risk recall. Statistical significance was assessed using bootstrap confidence intervals and McNemar and DeLong tests. Results: The lightweight graph-based model demonstrated competitive classification performance while maintaining reduced computational complexity. Although none of the models achieved the predefined high-risk recall threshold, statistically significant performance differences were observed across model families. Conclusion: The findings suggest that biomechanically informed imbalance handling improves methodological validity in ergonomic risk assessment. While deployment feasibility appears promising, further empirical validation on edge hardware is required.

The high theoretical capacity of transition metal-based compounds makes them promising candidates for lithium-ion battery (LIB) anodes. Among them, iron selenide (FeSe2) has attracted considerable interest because of its excellent electrical conductivity and superior lithium storage capacity. However, pristine FeSe2 suffers from rapid capacity fading and structural instability during repeated cycling. Thus, this study used a facile solvothermal method to synthesize a FeSe2@rGO composite with enhanced structural integrity and electrical conductivity. By incorporating reduced graphene oxide (rGO), the composite demonstrated improved charge transfer kinetics and mechanical robustness. Morphological and structural characterizations were performed using scanning electron microscopy (SEM), transmission electron microscopy (TEM), X-ray diffraction (XRD), and X-ray photoelectron spectroscopy analyses (XPS), which confirmed the successful formation of the composite and its uniform distribution. Electrochemical properties were evaluated using cyclic voltammetry (CV), galvanostatic charge-discharge, long-term cycling, and electrochemical impedance spectroscopy. The optimized FeSe2@rGO electrode delivered a high reversible capacity of 971.95 mAhg-1 at 500 mAg-1 after 350 cycles. The underlying charge storage mechanism was investigated using scan rate-dependent CV, which revealed a dominant capacitivecontrolled contribution at higher scan rates. The study findings indicate that the FeSe2@rGO composite can serve as a high-performance anode material with excellent cycling stability and rate capability, providing a viable strategy for the development of advanced LIBs.

This study identifies the crystalline defects commonly observed in conventional yttria ceramics, and proposes a processing route to produce highly densified yttria without the use of sintering aids. The primary objective is to obtain a dense yttria monolith with optimized microstructure and enhanced functional properties. The sintering behavior, mechanical performance, and plasma etching resistance of the yttria specimens were systematically analyzed as a function of the initial powder characteristics. A three-step sintering protocol was employed to suppress abnormal grain growth, yielding fully densified ceramics with uniform and controlled grain size distribution. Calcination of the yttria powder at 1,250 °C for 48 h effectively eliminated oxygen deficiencies and minimized hydration effects. The duration and temperature of each sintering stage significantly influenced grain evolution, which was reflected in the variations in Vickers hardness, Young’s modulus, and fracture toughness (KIC). The resulting yttria ceramics exhibited superior plasma resistance compared with Al2O3, ZrO2, quartz, and Si wafer, demonstrating markedly reduced weight loss during plasma etching. These findings highlight the critical role of proper initial powder treatment and precisely controlled sintering kinetics for achieving yttria monoliths with enhanced densification, mechanical integrity, and plasma erosion resistance. This work provides a practical route for high performance yttria ceramics, offering enhanced densification, structural stability, and the reliability necessary for integration into advanced systems exposed to harsh plasma environments.

Camellia japonica L. is highly valued for its ornamental and industrial applications. However, existing limitations in conventional seed and cutting propagation necessitate the development of a stable and efficient mass propagation system. This study systemically optimized each critical stage of in vitro culture—including shoot and root development, multiple shoot induction, rooting, and acclimatization —and quantitatively assessed the overall efficiency using integrated indices. Shoot growth was most vigorous on Woody Plant Medium (WPM) without the addition of indole-3-butyric acid (IBA), while root development was notably promoted by Murashige and Skoog (MS) medium supplemented with IBA. The highest number of multiple shoots was produced using basal explants cultured on MS medium containing 0.5 mg/L thidiazuron (TDZ), yielding an average of 2.67 shoots per explant. Optimal root induction was observed following a 15-min pulse treatment with 500 mg/L IBA (producing 24,33 roots), whereas the root elongation was maximized by a 5-min treatment with 1000 mg/L IBA (2.10 cm). Acclimatization successfully resulted in 100% survival in both tested substrates (A: peat moss, perlite, and cocopeat mixed in a 3:1:1 ratio; B: peat moss, perlite, and vermiculite mixed in a 1:1:1 ratio), with substrate B promoting a greater increase in plant height. Normalized growth parameters were averaged to calculate the Camellia Micropropagation Index (CMI). Integrated analysis identified the most efficient treatments as: WPM without IBA (shoot growth), MS with IBA (root growth), MS + 0.5 mg/L TDZ with basal explants (multiple shoots), 1000 mg/L IBA for 5 min (rooting), and substrate B (acclimatization). Despite these optimal conditions, considerable variation within treatments suggests that further fine-tuning or long-term evaluation is necessary to improve reliability. These findings provide a robust guideline for establishing a successful in vitro mass micropropagation system for C. japonica.

본 연구는 중국에서 지방 정부 디지털화가 중국 지방정부의 보조금 배분에 미치는 영향을 체계적 으로 분석하고, 기업 수준의 디지털 전환과의 상호작용 효과 및 발전 단계별 시간적 역학을 규명한 다. 2011년부터 2023년까지 중국 A주 상장기업의 21,742개 기업-연도 자료를 활용하여 패널 데 이터 분석을 수행하였다. 정부 디지털화는 주성분분석을 활용한 종합지수로 측정하였으며, 기업 디 지털화는 연차보고서 텍스트 마이닝을 통해 정량화하였다. 분석 결과, 지방 정부의 디지털화는 보 조금 규모를 직접적으로 확대하기보다는, 기업의 디지털 전환 수준과 상호보완적으로 작용하여 디 지털 역량이 높은 기업에 보조금이 더 효율적으로 배분되도록 하는 방식으로 작동하였다. 시기별 분석에서는 임계치 효과가 나타나, 기초 단계에서는 미약한 효과, 발전 단계에서는 유의하지 않은 효과, 심화 단계에서는 뚜렷한 긍정적 효과가 확인되었다. 또한 이질성 분석에서는 디지털 균등화 효과가 나타나, 국유기업 선호나 대기업 우위와 같은 전통적 제도적 이점은 약화된 반면, 하이테크 기업은 전통 산업보다 정부 디지털화의 혜택을 더 크게 누리는 것으로 나타났다. 본 연구는 디지털 정부와 기업 간 상호작용의 새로운 실증적 근거를 제시하며, 시너지 효과와 제도적 균등화 효과를 규명함으로써 기존의 이론적 예상을 재검토한다. 이를 통해 자원의존이론과 제도이론을 디지털 시 대의 맥락 속에서 확장하는 데 기여한다.

Flavonoids extracted from the roots of Petasites japonicus were evaluated for their cytoprotective, antioxidant, and anti-aging effects. The MTT assay confirmed that cell viability remained above 95% across concentrations up to 12 μg/mL, indicating no cytotoxicity. The extract demonstrated strong DPPH radical scavenging activity in a concentration-dependent manner, reaching 86.7% at 12 μg/mL, which is comparable to vitamin C. ROS scavenging activity also increased with dosage, showing significant suppression at concentrations of 6 μg/mL and above, thus effectively mitigating oxidative stress. Collagen synthesis assays revealed an initial decrease at low concentrations, followed by a clear recovery and significant enhancement at higher doses (10–12 μg/mL). Additionally, the extract inhibited collagenase activity, with notable suppression occurring at concentrations above 10 μg/mL, suggesting protective effects against collagen degradation. Elastase activity was reduced in a dose-dependent manner, achieving over 60% inhibition at 10–12 μg/mL. These results imply that flavonoids have dual functions: they stimulate collagen production while suppressing the enzymes that degrade collagen and elastin. The strong antioxidant and anti-inflammatory potential of these flavonoids likely contributes to their protective effects on dermal structure and function. Collectively, these findings highlight the potential of flavonoids as promising natural ingredients for anti-aging cosmetics and skin health applications.