Earticle

Cordycepin is the principal bioactive compound produced by Cordyceps militaris and exhibits diverse pharmacological properties. However, cordycepin production is highly sensitive to cultivation conditions, leading to substantially variable production amounts and challenges in process optimization. An interpretable machine learning framework was established in this study to predict the cordycepin produced by C. militaris cultivated on Pinus densiflora sawdust. Three key cultivation parameters—input weight, growth weight, and particle size—were quantified using submerged mycelial culture. The cordycepin content was measured via high-performance liquid chromatography. Four predictive models (random forest, support vector machine, XGBoost, and artificial neural network) were optimized through a randomized hyperparameter search and evaluated using internal validation and Tropsha’s external quantitative structure-activity relationship criteria. The validation accuracy of XGBoost was the highest (root mean square error = 42.67 μg/mL), whereas the external performance of random forest was the most reliable (R² = 0.898). Shapley additive explanations revealed that input weight most strongly influenced cordycepin production, followed by growth weight and particle size, with distinct nonlinear and interaction-driven effects among the cultivation variables. Kernel density and dependence analyses confirmed the occurrence of multimodal production regimes associated with the substrate loading and particle size characteristics. Finally, the best-performing model was deployed through a streamlit-based graphical user interface, enabling the real-time prediction of cordycepin concentration with a 95% confidence interval. The results collectively demonstrate the utility of interpretable AI-driven modeling for unveiling complex biological responses, providing a practical decision-support tool for optimizing cordycepin production in fungal biotechnologies.

Data-driven learning is the most representative deep learning method. Generative adversarial networks (GANs) are designed to generate sufficient data to support such learning. The learning process of GAN models typically trains a generator and discriminator in turn. However, overfitting problems occur when the discriminator depends excessively on the training data. When this problem persists, the image created by the generator shows a similar appearance to the learning image. Images similar to learning images eventually lose the meaning of data augmentation. In this paper, we propose a limited discriminator GAN (LDGAN) model that explains the results of GAN, which is a model that can not be analyzed externally, such as a black box. The part explained in LDGAN becomes the discriminator model of GAN, and it is possible to check which area of the image is used as the basis for determining fake/real by the discriminator. In the end, a method for limiting the learning of discriminator is proposed based on the described results. Through this, it is possible to avoid the overfitting problem of the discriminator and to generate various images different from the learning image. The LDGAN method allows users to perform meaningful data augmentation with only specific objects except for complex images or backgrounds that require analysis. Compare the LDGAN method with the existing DCGAN and present the extensive simulation results. The extensive simulation result shows that the image generated by the proposed LDGAN including the estimation area is about 10% more.

생성형 인공지능 (AI)의 확산은 게임 개발의 기획, 프로그래밍, 아트 제작, 품질관리 전반을 변화시키고 있으며, 이에 따라 대학 게임교육에서 요구되는 역량도 재정의되고 있다. 본 연구는 생성형 AI 시대의 게임 대학교육을 위한 Korea Game AI Education Model (K-GAEM) 1.0 역량 프레임워크를 개발하는 것을 목적으로 한다. 이를 위해 게임기획 2명, 게임개발 2명, 게임아트 2명으로 구성된 산업계 전문가 6명, 4년제 대학교 게임학과 학과장 3명과 2년제 대학교 게임학과 학과장 3명으로 구성된 학계 전문가 6명, 교육학 박사 2명 등 총 14명의 전문가 패널을 구성하였다. 2025년 9월부터 12월까지 3회전 수정 델파이 조사를 실시하였으며, 3개 라운드 모두 응답률 100%를 확보하였다. 이후 동일 패널을 대상으로 AHP 쌍대비교 분석을 수행하였다. 그 결과 K-GAEM 1.0은 AI 리터러시, AI 융합 역량, 게임 전공 역량, 창의적 기획 역량, 산업 실무 역량의 5개 핵심 영역, 15개 하위 영역, 49개 역량 요소로 구성되었다. 본 연구는 국내 게임 관련 학과의 교육과정 개편과 후속 FGI 기반 박사학위 연구의 기초자료를 제공한다.

The rapid diffusion of generative artificial intelligence (AI) has reshaped game-development workflows and has created new competency requirements for entry-level game professionals. Existing game university education in Korea, however, remain organized mainly around traditional engine, graphics, and planning courses and have not fully incorporated AI-augmented production. This study develops the Korea Game AI Education Model (K-GAEM) 1.0, a competency framework for game university education in the generative AI era. A three-round modified Delphi survey was designed with 14 experts (six industry practitioners in game planning, development, and art; six heads of game departments from four-year and two-year colleges; and two doctors of education) between September 2025 and December 2025, achieving a 100% response rate across all three rounds. The validated competency structure was then prioritized through the Analytic Hierarchy Process (AHP). The resulting framework consists of five domains, 15 sub-domains, and 49 competency elements. AHP weights placed AI Literacy first (0.276), followed by AI Convergence Competency (0.231), Game Major Competency (0.203), Creative Planning Competency (0.166), and Industry Practical Competency (0.124). K-GAEM 1.0 provides a structured basis for curriculum redesign, credit allocation, and future FGI-based validation.

본 연구는 기업의 AI Transformation(AX)을 지원하기 위해, Vision Language Model(VLM) 기반 지능 형 문서처리 플랫폼을 설계하고, Qwen2.5VL-7B를 활용한 영수증 처리 프로토타입을 구현하였다. 제안된 플랫폼 은 3-Tier 마이크로서비스 아키텍처를 기반으로, 프롬프트 관리 체계와 기능별 모듈화를 통해 유연하고 확장 가능 한 구조를 구현하였다. 실험 결과, 평균 91.7%의 정보 추출 정확도를 달성하였으며, 사전 템플릿 없이 다양한 문서 형식에 대응 가능한 처리 유연성을 바탕으로 실무 적용 가능성을 입증하였다. 본 연구는 OCR 중심 기술의 한계를 보완하는 프롬프트 기반 VLM 아키텍처를 실증적으로 제시하고, 금융·물류·의료 등 산업 전반에서 적용 가능한 문 서 자동화 기반을 제공하였다는 점에서 학문적·실무적 의의를 갖는다.

This study supports corporate AI Transformation (AX) by designing a document processing platform based on a Vision Language Model (VLM) and implementing a prototype using Qwen2.5VL-7B. The platform employs a three-tier microservice architecture with prompt management and modular components to ensure flexibility and scalability. Experiments showed an average information extraction accuracy of 91.7%, and the system demonstrated practical applicability by handling diverse document formats without predefined templates. This research provides an empirical implementation of a prompt-based VLM architecture that overcomes limitations of OCR technologies, offering academic and practical value as a foundation for document automation across sectors such as finance, logistics, and healthcare.

MCN은 인간 크리에이터 의존에서 비롯된 구조적 취약성을 내포한다. 본 연구는 BMC와 초정상자극 이론의 통합 프레임워크를 통해 AI 인플루언서 도입이 MCN 비즈니스 모델을 어떻게 전환시키는지 탐색한다. 문헌 고찰과 다중 사례 연구 방법론으로 5개 사례에서 BMC 9개 구성 요소의 전환 패턴을 도출하고, 초정상자극 이론으로 고객 관계 요소의 소비자 수용 메커니즘을 설명한다. 교차 사례 분석에서 가치 제안 전환, 핵심 자원 재편, 수익 모델 다변화의 3단계 경향성이 도출되었다. 본 연구는 BMC와 초정상자극 이론의 통합 적용 가능성을 탐색적으로 제시한다.

Ns face structural vulnerabilities stemming from human creator dependency. This study proposes an integrated framework of BMC and supernormal stimuli theory to analyze how AI influencer adoption transforms MCN business models. Multiple case study methodology across five AI influencer cases examines transformations across all nine BMC components, with supernormal stimuli theory explaining consumer acceptance mechanisms in the customer relationship component. Cross-case analysis reveals a three-stage trajectory: value proposition shift, key resource restructuring, and revenue model diversification. This study presents an exploratory integration of BMC and supernormal stimuli theory, offering a strategic roadmap for MCN operators navigating AI-driven transformation.

Mobile applications can be easily downloaded and installed via markets. However, malware and malicious applications containing unwanted advertisements exist in these application markets. Therefore, smartphone users install applications with reference to the application review to avoid such malicious applications. An application review typically comprises contents for evaluation; however, a false review with a specific purpose can be included. Such false reviews are known as fake reviews, and they can be generated using artificial intelligence (AI)-based text-generating models. Recently, AI-based text-generating models have been developed rapidly and demonstrate high-quality generated texts. Herein, we analyze the features of fake reviews generated from Generative Pre-Training-2 (GPT-2), an AI-based text-generating model and create a model to detect those fake reviews. First, we collect a real human-written application review from Kaggle. Subsequently, we identify features of the fake review using natural language processing and statistical analysis. Next, we generate fake review detection models using five types of machine-learning models trained using identified features. In terms of the performances of the fake review detection models, we achieved average F1-scores of 0.738, 0.723, and 0.730 for the fake review, real review, and overall classifications, respectively.

본 연구는 AI 기반 생체인식 감시기술이 공항과 같은 고위험 공공 인프라 환경에서 어떠한 방식으로사회적으로 수용되는지를 정보시스템(IS) 관점에서 분석한다. 기존 기술수용 연구가 개인의 인지적평가와 기술 성능에 기반한 선형적 설명에 주로 초점을 맞추어 온 것과 달리, 본 연구는 감시기술수용이 제도적 신뢰, 감시 불안, 사회적 편향 인식이 상호작용하는 구조적 긴장 관계 속에서 형성된다는점에 주목한다. 이에 신뢰-불안-편향의 균형 구조를 설명하는 이론적 틀로서 Triple Tension Model을제안하고, Responsible AI 조정 메커니즘(설명가능성, 동의, 감독, 참여)이 이러한 균형 구조에 어떠한영향을 미치는지를 분석한다. 문헌 기반 질적 비교사례연구(Qualitative Comparative Case Study)를적용하여 유럽연합(EU), 미국, 대한민국, 중국의 주요 국제공항을 분석한 결과, 국가별 감시 거버넌스의특성과 Responsible AI 제도화 수준에 따라 신뢰-불안-편향의 균형 유형이 상이하게 형성됨이 확인되었다. EU 사례에서는 제도적 신뢰가 불안과 편향을 조정하는 조화형(trust-balanced) 구조가 나타났으며, 미국사례에서는 불안과 편향 요인이 신뢰 형성에 상대적으로 두드러지는 양상을 보이는 긴장 우세형(Tension-dominant) 구조로 해석되었다. 대한민국은 제도화 수준에 따라 균형이 이동 가능한 전환형(transitional) 구조를 보였으며, 중국은 제도적 관리 체계를 중심으로 불안과 편향 인식이 낮은 수준으로유지되는 비대칭 안정형(asymmetrical stability) 구조로 해석되었다. 본 연구는 AI 감시기술의 사회적수용성이 기술 성능 자체의 차이보다는 제도적 신뢰 형성 방식과 Responsible AI 조정 메커니즘의제도화 수준과 밀접하게 연관되어 있음을 보여준다. 이를 통해 기존 IS 기술수용 연구의 분석 범위를고위험․비자발․권력 비대칭 환경으로 확장하고, 신뢰 기반 감시 거버넌스 설계의 이론적․정책적함의를 제시한다.

This study examines how AI-based biometric surveillance technologies are socially accepted in high-risk public infrastructure environments such as airports from an information systems (IS) perspective. Moving beyond traditional technology acceptance models that emphasize individual cognition and performance, this research conceptualizes acceptance as a structural tension among institutional trust, surveillance anxiety, and perceived bias. The study proposes the Triple Tension Model and analyzes how Responsible AI mechanisms—explainability, consent, supervision, and participation—shape this balance. Using a qualitative comparative case study of major international airports in the EU, the United States, South Korea, and China, the findings reveal distinct equilibrium patterns across governance contexts. The EU exhibits a trust-balanced structure, the United States a tension-dominant structure, South Korea a transitional structure, and China an asymmetrical stability structure. The results highlight that social acceptance of AI surveillance is less determined by technological performance than by institutional trust formation and the level of Responsible AI implementation. This study extends IS research into high-risk, non-voluntary, and power- asymmetric contexts and offers theoretical and policy implications for designing trust-based surveillance governance.

[연구목적] 인공지능(이하, AI) 기술 기반 리걸테크(Legaltech)는 법(legal)과 기술(technology)의 융합된 기술 분야로서 법률과 기술을 결합하여 변호사 전환했다. 이러한 혁신적 기술의 발전에 따른 AI 프로그램의 사용시에 발생될 수 있는 문제를 도출해 AI 프로그램 활용을 위한 윤리적 규범을 구체적으로 제안하고자 한다. [연구방법] 미국 변호사협회 모범변호사업무규칙(ABA Model Rules of Professional Conduct)에서 AI 기술을 변호사가 사용할 때 문제가 될 수 있는 규정을 종합적으로 분석하고, 우리나라의 변호사윤리장전에서의 시사점과 구체적 개선 방안을 도출하였다. 이를 통해 미래 AI 프로그램을 사용할 때의 법적·윤리적 문제를 사전에 해결할 수 있는 이정표를 제시하였다. [연구결과] AI 기반 프로그램의 법률 분야 활용은 변호사 업무 효율성을 높이고, 새로운 업무 처리 방식 전환의 기회를 제공하지만, 동시에 윤리적 문제, 법적 책임, 편향 및 차별 등의 문제점을 일으킬 수 있다. 변호사법은 상대적으로 개정 절차가 복잡하고 시간이 오래 걸리며, 빠르게 변화하는 AI 기술 발전 속도에 적응하기 어렵다. 이에 변호사윤리장전은 상대적으로 개정 절차가 간편하고 유연하여, AI 기술 발전에 맞춰 윤리적 기준을 신속하게 수정하고 보완할 수 있다. AI 기반 프로그램의 잠재적 문제점을 해결하고, 변호사의 윤리적 책임을 강화하며, 신뢰성 및 명확성을 확보하기 위한 기반을 마련하기 위해 변호사윤리장전의 개정(안)을 제시했다. 첫째, AI 기반 프로그램의 명확한 정의가 필요하다. 둘째, AI 기반 프로그램의 투명성 및 책임성 강화가 필요하다. 셋째, AI 기반 프로그램의 편견 및 차별 방지가 필요하다. 넷째, AI 기반 프로그램을 사용하는 변호사는 전문지식을 갖추어야 한다. 다섯째, 변호사의 지속적인 교육 및 역량 강화가 필요하다. [연구시사점] 최근 AI 기술 발전과 함께 법률 분야에서도 AI 프로그램 활용이 증가한바 변호사 업무 효율성 향상, 새로운 서비스 제공 가능성, 변호사 전문성 강화 등 다양한 긍정적 효과를 기대할 수 있다. 다만, 프로그램 사용에 따른 윤리적 문제 해결과 법적 제도 마련 등 해결해야 한다. 이에 이 연구에서는 AI 프로그램을 효과적으로 활용하면서, 변호사의 전문성을 유지하고 발전시킬 수 있는 변호사윤리장전의 개선 방안을 제시하였다.

[Purpose] Legaltech, which is based on artificial intelligence technology, is a technical field that combines legal and technology, transforming the way lawyers work by combining law and technology. We will identify problems that may arise when using AI programs that accompany the development of such innovative technologies, and propose concrete ethical norms for the use of AI programs. [Methodology] We comprehensively analyzed the provisions of the ABA Model Rules of Professional Conduct that may cause problems when lawyers use AI technology, and derived suggestions and specific improvement proposals for the Korean Code of Ethics for Lawyers. This provided a milestone that could help resolve legal and ethical issues when using AI programs in the future. [Findings] The use of AI-based programs in the legal field can improve the work efficiency of lawyers and provide opportunities for new work style transformation, but at the same time it can also raise problems such as ethical issues, legal liability, bias and discrimination. . The revision process for the Lawyers Act is relatively complex and time-consuming, and it is difficult to adapt to the rapidly changing speed of development of AI technology. This makes the Lawyer Code of Ethics relatively easy and flexible to amend, and the ethical standards can be quickly revised and supplemented in line with the development of AI technology. Proposed amendments to the Code of Ethics for Lawyers to resolve potential problems with AI-based programs, strengthen lawyers' ethical responsibilities, and provide a foundation for ensuring credibility and clarity. First, we need a clear definition of AI-based programs. Second, we need greater transparency and accountability for AI-based programs. Third, we need to prevent bias and discrimination in AI-based programs. Fourth, lawyers using AI-based programs should have expertise. Fifth, there is a need for continued education and capacity building for lawyers. [Implications] Recently, with the development of AI technology, the use of AI programs has increased in the legal field, and various positive effects are expected, such as improved work efficiency for lawyers, the possibility of providing new services, and strengthening of lawyers' expertise. can. However, ethical issues related to the use of the program and the development of a legal system must be resolved. This study proposed ways to improve the Code of Ethics for Lawyers, which can maintain and develop the professionalism of lawyers while effectively utilizing AI programs.

The problem of discrimination due to information alienation among Korean sign language users is continuously mentioned, and sign language translation research is actively being conducted to solve this problem. However, due to technological limits in translating text into Korean Sign Language, the need for specialist equipment causes annoyance and spatial constraints. Furthermore, it isn't easy to replicate the vocabulary and grammatical structure of the Korean language in Korean Sign Language. Furthermore, the service's commercialization is complicated by the need for more nonmanual signal (NMS) identification technology.

Approximately 40% of underground water and sewage pipes in Korea are more than 20 years old. Consequently, potential accidents related to water drainage systems are to be expected. In this study, using a special machine learning method that employs various available data, we developed a system that receives and analyzes data in smart pipes called the "digital twin-based smart pipe integrated management system" (DTMS-IM). This system presents an integrated approach for the efficient operation and monitoring of water pipes, allowing the innovative operation of groundwater pipes through smart decision-making. We trained the model using these data. This well-trained model has become able to predict the aging level of pipes. Similar artificial intelligence prediction models, widely used in various industrial applications, are also discussed.

The exponential growth of digitally produced content has necessitated the advent of smart, automated systems that can generate quality, search-optimized materials. IntelliWriter.io unveils a multi-model AI architecture that harmonizes transformer-based language models for SEO-centric content creation, improvement, and dissemination. In contrast to standard text generators, IntelliWriter employs domain-specific fine-tuning, keyword clustering, and contextual weighting to deliver both relevance and readability. By its being integrated with such platforms as WordPress, Shopify, and Wix, it facilitates the seamless auto-publishing and real-time metadata optimization. Experimental assessment suggests that IntelliWriter is cutting the editing time by 62% and improving the SEO ranking performance by 38% which makes it a next-generation framework for intelligent content automation.

Accurate classification of public policy is essential for effective policy analysis, design, comparison, and formulation across countries. However, manual classification by policy experts can lead to inconsistencies and human errors, highlighting the need for a more reliable and efficient approach. This study proposes a deep learning-based model to support policy classification using artificial intelligence. Leveraging Korean policy datasets, comprising administrative data (1988–2018), legislative data (1987–2018), and media data (1988–2020), previously curated by experts, we developed an AI model for automated policy classification based on the KoBERT language model. Designed as a supplementary tool for policy experts, this model enhances classification consistency, reduces decision-making time, and improves overall productivity. Moreover, the model enables the classification, comparison, and evaluation of diverse policies at both local and national levels, offering valuable support for strategic public policy development. The proposed model achieved a Top-1 accuracy of 62.4% and a Top-3 accuracy of 71.6%, outperforming traditional baselines and demonstrating its practical potential for real-world policy analysis.

Our study proposes a socio-relational approach intended to inform the training of an artificial intelligence system for botnet detection. First, a corpus of accounts likely to be automated was assembled using individual criteria defined by the Beelzebot team (Brachotte et al.). These accounts were then analysed through their interaction dynamics in order to identify relational configurations that could serve as relevant signals for automated detection. The article presents a socio-relational analysis based on a three-step protocol: (1) identifying forms of self-interaction; (2) examining internal interactions among suspected accounts; and (3) analysing their external interactions with third-party actors. Conducted within the framework of the ANR Beelzebot project, which aims to develop the first French-language solution capable of detecting information manipulation strategies deployed by automated networks in the French-speaking X-sphere, this research constitutes an exploratory phase designed to calibrate the data-preparation methodologies required for training an AI model that integrates socio-relational indicators. In addition to producing a quantitative score, our model aims to provide a complementary qualitative output that offers insight into the characteristics of the botnet and the functional roles occupied by different bot profiles within the network. From an ethical standpoint, this approach contributes to the development of a more explainable AI model.

한국의 사회서비스 전달체계는 부처별 분절, 공급자 중심 운영, 반복적 신청 절차로 인해 복지 사각지대와 행정 비효율을 초래하고 있다. 본 연구의 목적은 이러한 한계를 극복하기 위한 사용자 중심 AI 통합 모델, 즉 초개인화 사회복지 플랫폼(Hyper-Personalized Social Welfare Platform, H-SWP)을 제안하는 데 있다. 본 연구는 공공서비스 설계, 서비스 주권, AI 기반 공공서비스 혁신 관련 선행연구를 검토하고, 한국과 해외의 디지털정부 및 통합서비스 사례를 비교 분석하여 한국형 사회서비스 전달체계 개편모형을 설계하였다. 분석 결과, H-SWP는 AI 기반 욕구 예측, 통합 서비스 매칭, 디지털 케어 조정, 피드백 기반 정책 개선의 네 모듈로 구성될 때 가장 높은 정책적 정합성을 갖는 것으로 나타났다. 특히 선행연구와 공식 사례를 종합하면, 통합서비스 모델은 반복 신청 부담을 줄이고, 취약위험의 조기 식별과 연계 추적을 가능하게 하며, 사용자 중심 서비스 경험과 정책학습을 동시에 강화할 수 있는 잠재력을 가진다. 반면, 개인정보 보호, 부처 간 데이터 연계, 알고리즘 편향, 디지털 격차는 핵심 제약요인으로 확인되었다. 따라서 한국형 H-SWP의 실현을 위해서는 단계적 파일럿, 법·제도 정비, privacy-by-design 기반 데이터 거버넌스, human-in-the-loop 심사체계, 그리고 기술복지전문가 양성이 병행되어야 한다. 본 연구는 AI 기반 복지전달체계의 방향을 제시하는 설계지향형 논문으로서, 향후 실증 검증과 지역 단위 파일럿 연구를 위한 측정 가능 지표와 검증 프레임을 함께 제시한다.

South Korea’s social service delivery system remains fragmented across ministries and agencies, generating service gaps, duplicated administrative burdens, and a provider-centered experience for users. This study proposes the Hyper-Personalized Social Welfare Platform (H-SWP) as a user-centered AI integration model for restructuring the Korean social service delivery system. This evidence-informed, design-oriented study reviews the literature on public service design, service sovereignty, AI-enabled public sector innovation, and integrated service delivery, while also drawing on comparative digital-government cases to construct an implementation framework appropriate for Korea. The analysis suggests that the H-SWP becomes most coherent when structured around four interconnected modules: AI-based need prediction, integrated service matching, digital care coordination, and feedback-based policy refinement. Existing scholarship and official public-sector cases indicate that integrated and user-centered service models can reduce administrative burden, improve continuity of care, support proactive identification of risk, and strengthen policy learning through service data. At the same time, legal constraints on data sharing, privacy risks, algorithmic bias, and the digital divide remain significant implementation barriers. The study therefore argues that the practical adoption of H-SWP requires phased piloting, institutional reform, privacy-by-design data governance, human-in-the-loop decision review, and the cultivation of Tech-Welfare Specialists. As a design-oriented conceptual study, this paper contributes not only a platform model but also a measurable validation framework for future empirical testing and pilot-based policy experimentation.

농산물은 기상, 기후 등의 변화로 인해 공급이 불규칙하고, 공급량이 10% 하락하면 가격이 50% 상승하는 가격 탄력성이 매우 높다. 이러한 농산물 가격의 변동으로 인해 소상인의 경매를 통해 생산자에게 대금의 안전성을 보장하고 있다. 그러나, 과잉생산으로 가격이 폭락할 경우, 생산자에 대한 보호 조치는 미비한 실정이다. 따라서, 본 논문에서는 농산물에 대한 가격을 인공지능 알고리즘으로 예측하여 전자거래 시스템에 활용할 수 있는 비즈니스 모델을 설계하였다. 이를 위해, 학습 패턴 쌍으로 모델을 학습시키고, ARIMA, SARIMA, RNN, CNN을 적용하여 예측 모델을 설계하였다. 최종적으로, 농산물 예측가격 데이터를 단기예측과 중기예측으로 분류하여 검증하였다. 검증 결과, 2018년 데이터를 기반으로 실제 가격과 예측 가격이 91.08%의 정확도를 나타냈다.

For agricultural products, supply is irregular due to changes in meteorological conditions, and it has high price elasticity. For example, if the supply decreases by 10%, the price increases by 50%. Due to these fluctuations in the prices of agricultural products, the Korean government guarantees the safety of prices to producers through small merchants' auctions. However, when prices plummet due to overproduction, protection measures for producers are insufficient. Therefore, in this paper, we designed a business model that can be used in the electronic transaction system by predicting the price of agricultural products with an artificial intelligence algorithm. To this end, the trained model with the training pattern pairs and a predictive model was designed by applying ARIMA, SARIMA, RNN, and CNN. Finally, the agricultural product forecast price data was classified into short-term forecast and medium-term forecast and verified. As a result of verification, based on 2018 data, the actual price and predicted price showed an accuracy of 91.08%.

As Industry 4.0 revolutionizes manufacturing, protecting worker safety, particularly in high-risk, multicultural environments, has become crucial. This study presents a conceptual model for an AI-driven safety monitoring system that combines biometric wearables, environmental sensors, behavior recognition, and multilingual alerts. It allows for real-time hazard detection and inclusive communication, all managed through a centralized dashboard. A simulation of oxygen depletion in a confined space demonstrates the system's ability to respond quickly. The model is compatible with Industry 4.0 platforms like MES, ERP, and Digital Twins, and supports migrant workers by breaking down language barriers with multilingual alerts and pictograms. The framework aligns with ISO 45001 and OSHA standards, and its modular, scalable design enables predictive risk mitigation and smart safety innovation. This framework sets the stage for future prototyping and field implementation.

The purpose of this study is to develop a narrative-timeline framework for VJ practice that integrates artistic intention with technical execution. The model combines an eight-stage emotional narrative structure with a six-phase timeline workflow, positioning the VJ as both creative director and coordination mediator in live event production. Building on this foundation, the study explores how artificial intelligence can extend the framework through functions such as automated emotional tagging, predictive workflow optimization, and real-time conflict resolution. While still conceptual, the results indicate the design feasibility of a replicable blueprint for participatory and sustainable VJ practice. Future work will involve empirical validation and the development of AI-assisted tools that preserve human creative agency while expanding narrative and operational possibilities in live media performance.

AI assistants are designed to closely interact with their users to help them complete various everyday tasks. To explain the continued use of AI assistants, we develop a research model that extends the investment model by Rusbult (1980), and empirically test it with data collected from an online survey of 172 users. The findings indicate that both satisfaction, which leads a user to become dedicated to a relationship with an AI assistant, and investment size, which constrains a user to a relationship with an incumbent AI assistant, significantly influence a user’s commitment to an AI assistant. The commitment in turn enhances a user’s continued use of an AI Assistant. We also find that perceived usefulness, rapport, and perceived social presence are key antecedents of satisfaction, while brand relationship and setup costs are key antecedents of investment size.

Heart disease is a major cause of mortality in the world that is in dire need of accurate, interpretable predictive measures that could be utilized to manage it proactively. The writer of this paper proposes an Explainable AI (XAI) Ensemble Machine Learning model to predict heart disease using an 1,025 patient record dataset. To achieve methodological rigor and generalization, 5- Fold Stratified Cross-Validation (CV) was used to evaluate all models, such as LightGBM and Random Forest. LightGBM model was stable and better in performance as it had Mean CV Accuracy of ±0.9620 ±0.0178. Integration of XAI (SHAP/LIME) is the means of creating clinical trust; analysis has confirmed maximum heart rate (thalach) and type of chest pain (cp) as medically significant characteristics. This framework supports the sustainable smart city healthcare through a highly transparent decision-support system, which manages the resources in optimizing scalable public health programs.