LGBTQ+ 권리와 종교의 자유의 조화를 위한 2015년 미국 유타주의 입법 타협 안은 양측의 권리를 둘러싼 갈등이 첨예하게 대립하는 상황 속에서 실질적이고 의미 있는 입법적 진전을 이룬 사례로 평가된다. 이 타협안이 마련되기 전까지 유타주는 성적 지향과 성 정체성에 기반한 차별로부터 개인을 보호하는 포괄적인 주(州) 차원의 법적 장치를 갖추지 못한 상태였다. 그러나 동성 결혼의 합법화와 같은 사법 판결, 시민 여론의 급격한 변화, 예수 그리스도 후기 성도 교회(LDS 교회)의 결정적인 지지 등이 계기가 되어, 성소수자 옹호 단체, 종교 단체, 입법 자, 기업인 등 다양한 이해관계자들이 협상 테이블에 함께하게 되었다. 그 결과 두 개의 상호보완적 법안이 제정되었는데, SB 296은 LGBTQ+ 개인을 위한 고용 및 주거 차별 금지를 주 전체로 확대하는 한편, 종교 기관에 대한 특정 예외 조 항을 포함하였고, SB 297은 결혼과 성과 관련된 종교적 표현의 자유와 양심적 거부권을 보호하는 내용을 담고 있다. 이 입법 타협은 시민적 다원주의와 실용적 정치 협상의 모범 사례로 많은 찬사를 받았지만, 동시에 보수와 진보 양측으로부 터 비판도 제기되었다. 보수 진영은 이 법안이 성소수자의 권리를 법적으로 인정 함으로써 종교의 자유를 훼손했다고 보았고, 진보 진영은 종교적 예외 조항이 평 등권의 실현을 지나치게 제한한다고 주장했다. 그럼에도 불구하고, 유타주의 경 험은 민주주의적 거버넌스, 상호 존중에 기반한 공존, 그리고 실용적 협상이라는 측면에서 중요한 통찰을 제공한다. 본 논문은 유타 절충안의 역사적 배경, 협상 과정, 입법 내용, 사회적 반응 및 정책적 함의를 분석하고, 특히 한국 사회에서 현재 논의되고 있는 LGBTQ+ 권리와 종교 자유의 균형 문제에 이 모델이 갖는 비교적 시사점을 제공한다.

In this paper, a water rescue mission system was developed for water safety management areas by utilizing unmanned mobility( drone systems) and AI-based visual recognition technology to enable automatic detection and localization of drowning persons, allowing timely response within the golden time. First, we detected suspected human subjects in daytime and nighttime videos, then estimated human skeleton-based poses to extract human features and patterns using LSTM models. After detecting the drowning person, we proposed an algorithm to obtain accurate GPS location information of the drowning person for rescue activities. In our experimental results, the accuracy of the Drown detection rate is 80.1% as F1-Score, and the average error of position estimation is about 0.29 meters.

The expansion of online retail markets has driven the development of personalized product recommendation services leveraging platform-based product and customer data. Large retailers have implemented seller-oriented recommendation systems, where AI analyzes POS sales data to identify similar stores and recommend products not yet introduced but successful elsewhere. However, small and medium-sized retailers face challenges in adapting to rapidly evolving online market trends due to limited resources. This study proposes a recommendation algorithm tailored for small-scale retailers using sales data from an online shopping mall. We analyzed 600,000 transaction records from 13,607 sellers and 95,938 products, focusing on Beauty Supplies, Kitchenware, and Cleaning Supplies categories. Three algorithms—Attentional Factorization Machines (AFM), Deep Factorization Machines (DeepFM), and Neural Collaborative Filtering (NCF)—were applied to recommend top 10% weekly sales items, with an ensemble model integrating their strengths. To address class imbalance, the Synthetic Minority Oversampling Technique (SMOTE) was employed, and performance was evaluated using AUC, Accuracy, Precision, and Recall metrics on separate training and test datasets. The ensemble model outperformed individual models across all metrics, while DeepFM excelled in Precision. These findings demonstrate that ensemble-based recommendation algorithms enhance recommendation accuracy for suppliers in large-scale online retail environments, offering practical implications for small-scale retailers.

Ensuring operational safety and reliability in Unmanned Aerial Vehicles (UAVs) necessitates advanced onboard fault detection. This paper presents a novel, mobility-aware multi-sensor health monitoring framework, uniquely fusing visual (camera) and vibration (IMU) data for enhanced near real-time inference of rotor and structural faults. Our approach is tailored for resource-constrained flight controllers (e.g., Pixhawk) without auxiliary hardware, utilizing standard flight logs. Validated on a 40 kg-class UAV with induced rotor damage (10% blade loss) over 100+ minutes of flight, the system demonstrated strong performance: a Multi-Layer Perceptron (MLP) achieved an RMSE of 0.1414 and R² of 0.92 for rotor imbalance, while a Convolutional Neural Network (CNN) detected visual anomalies. Significantly, incorporating UAV mobility context reduced false positives by over 30%. This work demonstrates a practical pathway to deploying sophisticated, lightweight diagnostic models on standard UAV hardware, supporting real-time onboard fault inference and paving the way for more autonomous and resilient health-aware aerial systems.

This study proposes a weighted ensemble deep learning framework for accurately predicting the State of Health (SOH) of lithium-ion batteries. Three distinct model architectures—CNN-LSTM, Transformer-LSTM, and CEEMDAN-BiGRU—are combined using a normalized inverse RMSE-based weighting scheme to enhance predictive performance. Unlike conventional approaches using fixed hyperparameter settings, this study employs Bayesian Optimization via Optuna to automatically tune key hyperparameters such as time steps (range: 10-35) and hidden units (range: 32-128). To ensure robustness and reproducibility, ten independent runs were conducted with different random seeds. Experimental evaluations were performed using the NASA Ames B0047 cell discharge dataset. The ensemble model achieved an average RMSE of 0.01381 with a standard deviation of ±0.00190, outperforming the best single model (CEEMDAN-BiGRU, average RMSE: 0.01487) in both accuracy and stability. Additionally, the ensemble's average inference time of 3.83 seconds demonstrates its practical feasibility for real-time Battery Management System (BMS) integration. The proposed framework effectively leverages complementary model characteristics and automated optimization strategies to provide accurate and stable SOH predictions for lithium-ion batteries.

This integrative review examined 38 published articles on the topic of formative assessment conducted in South Korean EFL settings, from 2014 to 2023. Despite strong interests in using quality formative assessments that are aimed at improving English learners’ performance, no prior systematic analyses have been performed to date. To that end, this study draws on Wiliam and Thompson’s (2008) conception of the formative assessment five-strategy model, and Yan and Pastore’s (2022) Teacher Formative Assessment Practice Scale (TFAPS), as a guiding framework to examine the extent to which formative assessment research conducted in South Korean EFL classroom settings has enacted the five key strategies and thereby promoted learning and improved teaching. For all five key strategies, more studies showcased weak evidence of implementing each strategy, and that positive student learning was more likely to be seen in studies that fully, or nearly fully, implemented the key strategies. Recommendations for enhancing formative assessment teaching and research practice are provided at the end.

Small and medium-sized manufacturing enterprises(SMEs) have traditionally relied on skilled labor to support multi-variety, small-batch production. However, demographic changes such as low birth rates and aging populations have led to severe labor shortages, prompting increased interest in collaborative robots(cobots) as a viable alternative. Despite this necessity, many SMEs continue to face significant challenges in implementing such technologies due to technical, organizational, and environmental(TOE) constraints. While prior research has mainly focused on technology adoption from the perspective of user organizations, this study adopts a differentiated approach by analyzing adoption factors from the perspective of smart factory experts—specifically, evaluators/mentors and solution providers—who play a critical role in Korea’s policy-driven smart manufacturing environment. Using the Analytic Hierarchy Process(AHP), the study evaluates the relative importance and prioritization of adoption factors across three dimensions: technology, organization, and environment. Survey data collected from 20 smart factory experts indicate that top management support, relative advantage, and safety are key determinants in cobot adoption. Furthermore, the findings reveal that organizational readiness and technical effectiveness have greater influence on implementation decisions than external pressures such as partner pressure. This study provides new insights by incorporating expert perspectives into the adoption framework and offers practical policy and managerial implications to support cobots implementation in the SMEs.

Business model(BM) innovation is widely known as a differentiated strategy and strategic framework for companies to secure a sustainable competitive advantage in an uncertain environment. While prior research has studied new business models in accordance with changes in manufacturing trends such as digitalization and servitization, empirical understanding of the dynamic processes of BM innovation is still lacking. This study addresses this gap by proposing an analytical framework of the BM innovation matrix that classifies companies' BM innovation cases into four types according to the degree of BM change and the influential level of the industry/market outcome through a critical literature review on business models and dynamics. Drawing on this framework, we conduct longitudinal case studies of leading global 3D printing firms to examine the dynamic processes and external environmental factors that shape the evolution of BM innovation. Our findings reveal previously underexplored patterns of co-evolution between firms’ business models and their broader industrial and market environments. This study has the significance of constructing a framework for dynamically analyzing BM innovation based on longitudinal case studies of emerging 3D printing companies. We presented implications for companies seeking successful commercialization of emerging technologies, such as the strategic usefulness of the BM innovation framework and the importance of co-evolution with industrial structure and environmental factors in the process of change.

The semi-direct speaking test format has limitations, particularly due to its lack of situational authenticity and contextualized input. To address this issue, virtual reality (VR) can be integrated into speaking proficiency tests to enhance authenticity. In this study, a newly designed VR speaking test was administered, and test-takers’ performances were compared with those on a conventional computer-delivered speaking test. Additionally, test-takers’ perceptions of the VR-based speaking test were examined through a post-test questionnaire. The results revealed a statistically significant difference in mean scores between the two test formats, indicating that the VR-based test enhanced test-takers’ speaking performance. More specifically, a one-way MANOVA showed that test-takers performed better on nearly all scoring criteria in the VR mode compared to the computer-delivered mode, except for completion and fluency. Furthermore, the 32 test-takers who participated in the VR test highlighted the highly contextualized settings and immersive experience as the most distinctive and positive aspects of using VR in speaking assessments.

The casting manufacturing process of aluminum automotive wheels often involves processing various wheel models during stages such as flow forming, machining, packaging, and delivery. Traditionally, separate equipment or production lines were required for each model, which led to higher facility investment costs and increased labor costs for classification. However, the implementation of machine learning-based model classification technology has made it possible to automatically and accurately distinguish between different wheel models, resulting in significant cost savings and enhanced production efficiency. Additionally, this approach helps prevent product mix-ups during the final inspection process and allows for the quick and precise identification of wheel models during packaging and delivery, reducing shipping errors and improving customer satisfaction. Despite these benefits, the high cost of machine learning equipment presents a challenge for small and medium-sized enterprises(SMEs) to adopt such technologies. Therefore, this paper analyzes the characteristics of existing machine learning architectures applicable to the automotive wheel manufacturing process and proposes a custom CNN(Convolutional Neural Network) that can be used efficiently and cost-effectively.

본 논문은 초국가적 종교 인식론적 공동체의 관점에서 탈동성애 운동 의 세계정치적 양상을 분석한다. 일반적으로 탈동성애 운동은 동성애 정 체성을 벗어났다고 주장하는 개인들에 의해 주도되는 지역적·문화적 현 상으로 여겨지지만, 실제로는 서구권의 보수적 복음주의 세력이 주도하 는 조직화된 국제 네트워크임을 확인할 수 있다. 이러한 네트워크는 특 정 분야의 정책 형성에 영향을 미치는 지식과 권위를 공유하는 전문가 집단, 즉 인식론적 공동체로서 기능한다. 특히 탈동성애 운동의 종교 보 수 활동가들은 자신을 인간의 성(性)에 관한 전문가로 규정하고, 과학적 외형을 갖춘 증거, 헌법적·권리 기반의 논리, 개인의 간증 등 다양한 전 략을 활용하여 동성애가 변화 가능하며 바람직하지 않은 것이라고 주장 한다. 본 연구는 한국을 주요 사례로 삼아, 한국의 보수적 복음주의 엘리 트들이 서구(특히 미국)의 복음주의자들로부터 탈동성애 담론과 전략을 체계적으로 수입하고 현지화한 방식을 경험적으로 분석한다. 구체적으로, 한국의 활동가들이 초국적으로 유통되는 진정성, 피해자성, 인권과 같은 자유민주주의적 레토릭을 전략적으로 차용하여, 자신들의 LGBTQ+ 인권 반대 활동을 종교의 자유 및 주체적인 삶의 경험(lived experiences)의 정당한 표현으로 재구성하고 있음을 보여준다.

This research, grounded in the extended technology acceptance model, aimed to explore the relationships among factors influencing Korean EFL learners’ acceptance of ChatGPT for English learning in a voluntary usage context. To this end, a questionnaire was distributed to college students who had used ChatGPT for language learning, utilizing a convenience sampling method. A total of 400 responses were analyzed to test hypotheses using structural equation modeling (SEM). Findings revealed that learners’ perceived usefulness significantly predicted their intention to continue using ChatGPT, while perceived ease of use did not. Moreover, learners’ result demonstrability was found to be a predictor of perceived usefulness, whereas subjective norm was not. Both playfulness and output quality significantly influenced learners’ perceived ease of use. This study identified key factors that could enhance EFL learners’ acceptance of ChatGPT by improving perceptions of usefulness and ease of use, offering valuable insights for integrating ChatGPT into English education.

This study investigates the effectiveness of self-correction in improving lexical stress placement among Korean English learners, a critical yet challenging feature for speakers of Korean, which lacks lexical stress contrasts. Grounded in Schmidt’s (1990) Noticing Hypothesis, the research compares the benefits of self-correction— where learners reflect on and correct their own pronunciation errors —with the shadowing technique. Forty-seven college students participated, with an experimental group practicing self-correction and a control group engaged in shadowing. Pre- and post-test analyses revealed that the self-correction group demonstrated significantly greater improvement, particularly with trisyllabic and tetrasyllabic words, while the shadowing group showed minimal change. These findings highlight self-correction’s role in promoting learner engagement, error awareness, and deeper cognitive processing, offering practical implications for pronunciation instruction that emphasizes learner autonomy and focused attention.

The electronic structures of graphene nanoflakes (GNFs) were estimated for various shapes, sizes, symmetries, and edge configurations. The Hückel molecular orbital (HMO) method was employed as a convenient way of handling the variety of possible GNF structures, since its simplicity allows the rapid solution of large system problems, such as tailoring optoelectronic characteristics of molecule containing large number of carbon atoms. The HMO method yielded the electronic structures with respect to the energy state eigenvalues, with results comparable to those obtained by other approaches, such as the tightbinding method reported elsewhere. The analyses included the consideration of various types of edge configurations of 68 GNF systems grouped by their geometric shape, reflecting symmetry. It was inferred that GNFs in the small length scale regimes, below 1 nm, which are effectively small polycyclic aromatic hydrocarbon molecules, exhibit the optoelectronic characteristic of quantum dots. This is due to the widely spaced discrete energy states, together with large energy gaps between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO). With increasing size this arrangement evolves into graphene-like ones, as revealed by the narrowing HOMO-LUMO gaps and decreasing energy differences between eigenstates. However, the changes in electronic structure are affected by the symmetries, which are related to the geometric shapes and edge configurations.

This study proposes a mathematical model to optimize the fighter aircraft-weapon combinations for the ROKAF(Republic Of Korea Air Force). With the recent emergence of the population declining issue in Republic of Korea, there is an urgent need for efficient weapon system operations in light of decreasing military personnel. In order to solve these issues, we consider to reduce the workload of pilots and maintenance personnel by operating an optimal number of weapons instead of deploying all possible armaments for each aircraft type. To achieve this, various factors for optimizing the fighter-weapon combinations were identified and quantified. A model was then constructed using goal programming, with the objective functions based on the compatibility, CEP(Circular Error Probable), and fire range of the weapons, along with the planned wartime mission-specific weapon ratios for each aircraft type. The experimental result's analysis of the proposed model indicate a significant increase in mission performance efficiency compared to the existing system in both operational and maintenance aspects. We hope that our model will be reflected to help improve the operational capabilities of Republic of Korea Air Force.

The study explores the relationship among teacher identity, teacher transparency, teacher self-efficacy, and teachers’ adaptation to digital change. Eighty-four English teachers participated in the study. For comparison between English and other subject teachers, 38 survey results of different subject teachers were included in the analysis. The results showed that English teachers’ scores were lower across all the constructs in terms of both transparency and self-efficacy compared to the scores of teachers in other subjects. For further analysis, the Structural Equation Modeling was run, and the results revealed that teacher transparency influences teacher self-efficacy, facilitating digital adaptation. Instructional Transparency and Peer Transparency were significant predictors of selfefficacy, directly influencing digital adaptation. This result illustrates the dynamic interplay between evolving teacher identity and self-efficacy in relation to digital adaptation through the relationship between teacher transparency and teacher selfefficacy. The findings indicate the need for targeted programs to enhance English teachers’ transparency and self-efficacy as a pathway to their digital adaptation.

This study proposes a weight optimization technique based on Mixture Design of Experiments (MD) to overcome the limitations of traditional ensemble learning and achieve optimal predictive performance with minimal experimentation. Traditional ensemble learning combines the predictions of multiple base models through a meta-model to generate a final prediction but has limitations in systematically optimizing the combination of base model performances. In this research, MD is applied to efficiently adjust the weights of each base model, constructing an optimized ensemble model tailored to the characteristics of the data. An evaluation of this technique across various industrial datasets confirms that the optimized ensemble model proposed in this study achieves higher predictive performance than traditional models in terms of F1-Score and accuracy. This method provides a foundation for enhancing real-time analysis and prediction reliability in data-driven decision-making systems across diverse fields such as manufacturing, fraud detection, and medical diagnostics.

Korean English medium instruction (EMI) classes aim to foster active discussions and communicative interactions in English between instructors and students. However, many Korean students in these classes struggle due to their limited English proficiency. This paper examines the challenges faced by Korean EFL students in EMI environments, highlighting the necessity for support in both English and their native language to facilitate effective learning. It also identifies teaching strategies that have proven effective in helping these students navigate language barriers. The findings indicate that participants had difficulty developing their writing skills for assignments in EMI settings and encountered limited opportunities to communicate their understanding of course material with instructors. To address these challenges, it is important to assess students’ language skills and find a balance between Korean and English. Implementing flexible teaching methods can enhance the learning experience, making it more effective and supportive. By providing multiple approaches to learning, such as interactive activities or peer support, learning gaps can be bridged and overall educational outcomes enhanced.

이 논문은 W. B. 예이츠의 유산과 멜리스 베스리의 소설 􋺷흩어진 사랑􋺸에 서 나타나는 뼈의 상징적 공명을 탐구한다. 소설에서 예이츠의 유령은 그의 삶, 예술, 그 리고 자신의 유해라는 수수께끼 같은 운명을 되새기며 지속적인 주제들을 반추한다. 뼈의 모티프는 사랑, 죽음, 그리고 기억의 교차점을 포괄하는 서사적이고 은유적인 축으로 기능한다. 이러한 관점에서 이 논문은 베스리가 예이츠의 미완의 열정, 특히 모드 곤에 대한 사랑을 어떻게 재구성하며 그의 정체성의 더 넓은 문화적, 역사적 차원을 어떻게 탐구하는지를 살펴본다. 예이츠의 유해 발굴과 재매장에 얽힌 논란은 예술적 불멸성과 신체의 덧없음 사이의 긴장을 심화시키며, 이는 아일랜드의 복잡하고 종종 단절된 역사 적 서사를 반영한다. 예이츠를 현대 문학적 맥락에 배치함으로써 베스리는 그의 비전을 확장하면서도 변모시키며, 사랑, 상실, 유산에 대한 예이츠의 사색을 현대적 맥락에 녹여 낸다. 이 논문은 베스리의 소설이 뼈를 기억의 매개체로 부각시키며 예이츠의 시적 상상 력과 그녀 자신의 재해석 사이의 다리를 놓는다고 주장한다. 궁극적으로, 이 연구는 예 이츠와 베스리가 공유하는 죽음, 역사, 그리고 예술과 감정의 초월적 힘에 대한 탐구가 여전히 중요한 의미를 가진다는 점을 강조한다. 􋺷흩어진 사랑􋺸 은 예이츠의 사랑의 유산 을 재구성하며, 그 자체로 그 유산의 본질적 일부가 된다.

Defective product data is often very few because it is difficult to obtain defective product data while good product data is rich in manufacturing system. One of the frequently used methods to resolve the problems caused by data imbalance is data augmentation. Data augmentation is a method of increasing data from a minor class with a small number of data to be similar to the number of data from a major class with a large number of data. BAGAN-GP uses an autoencoder in the early stage of learning to infer the distribution of the major class and minor class and initialize the weights of the GAN. To resolve the weight clipping problem where the weights are concentrated on the boundary, the gradient penalty method is applied to appropriately distribute the weights within the range. Data augmentation techniques such as SMOTE, ADASYN, and Borderline-SMOTE are linearity-based techniques that connect observations with a line segment and generate data by selecting a random point on the line segment. On the other hand, BAGAN-GP does not exhibit linearity because it generates data based on the distribution of classes. Considering the generation of data with various characteristics and rare defective data, MO1 and MO2 techniques are proposed. The data is augmented with the proposed augmentation techniques, and the performance is compared with the cases augmented with existing techniques by classifying them with MLP, SVM, and random forest. The results of MO1 is good in most cases, which is believed to be because the data was augmented more diversely by using the existing oversampling technique based on linearity and the BAGAN-GP technique based on the distribution of class data, respectively.