교육 분야에서 생성형 인공지능 기술의 보급과 함께 중국 고등학생의 이용의도의 영향 요인에 관한 분석은 이론적 및 실무적 의의가 있다. 본 연구의 목적은 중국 고등학교 교육 환경에 적합한 모형을 구축하고 이용 의도의 영향 요인을 분석함으로써 교육 분야의 인공지능 응용을 위한 근 거 자료를 제공하는 데 있다. 정보시스템 성공 모형, UTAUT 모형과 사 회적 인지 이론을 기반으로 사회적 영향, 자기효능감, 신뢰, 의인화, 경 제적 저항과 관습적 저항 등을 활용하여 만족도와 이용의도를 포함한 모 형을 구축했다. 설문조사와 구조방정식 모델을 통해 실증분석을 진행했 다. 연구 결과에 따르면, 첫째 시스템 품질, 정보 품질과 서비스 품질은 만족도에 긍정적인 영향을 미쳤다. 둘째, 사회적 영향과 신뢰는 교사, 학 부모, 또래집단과 학교 규범을 통해 만족도를 높일 수 있다. 셋째, 자기 효능감, 의인화와 만족도는 이용의도에 긍정적인 영향을 미쳤으며 의인 화가 가장 큰 영향을 미쳤다. 넷째, 경제적 저항과 관습적 저항은 만족도 와 이용의도에 각각 부정적인 영향을 미쳤다. 본 연구는 고등학생 집단 을 대상으로 정보시스템 성공 모형의 적합성을 검증하고 감정적 및 문화 적 요소를 도입하여 교육 기술 수용 관련 연구 모형을 구축했다. 실무적 시사점으로 첫째, 시스템 성능과 감정적 인터랙티브 디자인을 최적화한 다. 둘째, 교사와 학부모의 영향을 강화하고 긍정적인 이용 분위기를 조 성한다. 셋째, 이용 장벽을 낮추고 교육 자원의 보급성을 높인다.

This study explores how to integrate the generative artificial intelligence (AI) tool Midjourney into the fashion design process, emphasizing the visualization of sporty fashion concepts. The research applied Midjourney at every stage of the fashion design process: mood board, fashion sketch, flat drawing, production package, fashion show presentation, and store display and sales. Specifically, sporty fashion was selected as the theme, and customized prompts were developed from prior research and design principles to generate visual outputs for each stage. Furthermore, three apparel design experts evaluated the AI-generated images to assess Midjourney’s practical applicability and effectiveness in each phase of the fashion design workflow. Expert evaluations revealed that Midjourney was particularly effective in the early stages, offering diverse and visually engaging imagery that supported creative ideation and mood expression. The tool allowed quick exploration of different silhouettes during the sketching stage but was imprecise in detailed forms and proportions. Limitations became more evident in the flat drawing and work instruction stages, where outputs failed to accurately reflect material textures and technical construction. Prompt refinements and referencebased prompts were tested but often resulted in inconsistent or stylized outputs. Additionally, continuity between stages was missing. Midjourney shows potential as a creative tool, but experts highlight its limitations for practical industry application. Further research is needed to improve prompt optimization and training data for enhanced accuracy and usability in AI-assisted fashion design workflows.

This study aims to improve the interpretability and transparency of forecasting results by applying an explainable AI technique to corporate default prediction models. In particular, the research addresses the challenges of data imbalance and the economic cost asymmetry of forecast errors. To tackle these issues, predictive performance was analyzed using the SMOTE-ENN imbalance sampling technique and a cost-sensitive learning approach. The main findings of the study are as follows. First, the four machine learning models used in this study (Logistic Regression, Random Forest, XGBoost, and CatBoost) produced significantly different evaluation results depending on the degree of asymmetry in forecast error costs between imbalance classes and the performance metrics applied. Second, XGBoost and CatBoost showed good predictive performance when considering variations in prediction cost asymmetry and diverse evaluation metrics. In particular, XGBoost showed the smallest gap between the actual default rate and the default judgment rate, highlighting its robustness in handling class imbalance and prediction cost asymmetry. Third, SHAP analysis revealed that total assets, net income to total assets, operating income to total assets, financial liability to total assets, and the retained earnings ratio were the most influential factors in predicting defaults. The significance of this study lies in its comprehensive evaluation of predictive performance of various ML models under class imbalance and cost asymmetry in forecast errors. Additionally, it demonstrates how explainable AI techniques can enhance the transparency and reliability of corporate default prediction models.