Earticle

소프트웨어 시스템의 복잡성 증가는 기술 부채 누적을 초래하며, 품질 저하와 유지보수 비용 증가의 원인이 된다. 본 연구는 정량적 코드 메트릭과 개발자 피드백을 통합한 지표로 기술 부채를 평가하고 예측하는 방안을 제안한다. 정적 분석 데이터와 주관적 판단을 결합한 점수화 모델을 구축하고, 머신러닝 알고리즘으로 기술 부채 발생 가능성을 예측한다. 오픈소스 Java 프로젝트 실험 결과, XGBoost가 AUC 0.91로 가장 우수한 성능을 보였 다. 제안된 프레임워크는 기술 부채의 선제적 관리, 리팩토링 우선순위 결정, 유지보수 전략 수립에 효과적이다. 본 연구는 해석 가능하고 확장성 있는 데이터 기반 기술 부채 평가 방법을 통해 소프트웨어 품질 향상에 기여한다.

The increasing complexity of software systems leads to the accumulation of technical debt, causing quality degradation and higher maintenance costs. This study proposes a method for assessing and predicting technical debt by integrating quantitative code metrics with developer feedback. A scoring model that combines static analysis data and subjective evaluations was developed, and machine learning algorithms were applied to predict the likelihood of technical debt occurrence. Experiments on open-source Java projects demonstrated that the XGBoost model achieved the best performance with an AUC of 0.91. The proposed framework effectively supports proactive technical debt management, refactoring prioritization, and maintenance strategy planning. This research contributes to improving software quality by providing an interpretable and scalable data-driven approach for technical debt assessment.

본 연구는 태권도 겨루기 경기의 판정에 관한 선수 및 대중의 기대감을 충족시키기 위해 인공지능(Artificial Intelligence, AI) 판정시스템 도입 가능성을 살펴보고 이를 기반으로 태권도 겨루기 경기의 인공지능 판정시스템 도입의 구체적인 필요성과 정당성을 고찰하는 것에 목적을 두었다. 이러한 연구의 목적을 달성하기 위해 2024년 2월부터 10월까 지 총 9개월간 문헌고찰과 내용분석 방법을 사용하였다. 구체적인 과정은 학술연구정보서비스(RISS)에서 논문 제목과 초록 그리고 본문 등의 내용에서 스포츠, 융합, 인공지능, 태권도 경기 등이 포함되어진 연구들과 관련된 자료를 검색하였 고 제 1차 검색과정을 통해 2014년부터 2024년까지 총 28편의 논문이 검토되었으나 중복된 논문과 본 연구의 목적에 부합하지 않은 논문 6편을 제외한 총 22편을 선별하여 연구를 진행하였다. 연구결과로는 태권도 겨루기 경기의 새로운 판정시스템으로 인공지능(AI) 판정시스템을 제안하였으며 인공지능(AI) 판정시스템 도입의 구체적인 필요성과 정당성 은 첫째, 겨루기 경기의 판정의 공정성 및 관중 몰입감 향상과 둘째, 인공지능 관련 플랫폼 및 산업 활성화로 제시하였다.

This study studied the possibility of introducing an Artificial Intelligence (AI) judgment system to meet the expectations of athletes and the public regarding the judgment of Taekwondo Gyeorugi competition. The purpose of this study was to examine the necessity and justification of introducing an artificial intelligence judgment system for taekwondo Gyeorugi competition through research. To achieve the purpose of this study, literature review and content analysis methods were used for a total of 9 months from February to October 2024. For the specific process, the academic research information service (RISS) searched for data related to studies that included ‘sports’, ‘convergence’ ‘artificial intelligence’ and ‘Taekwondo Gyeorugi Competitions in the title, abstract, and text of the thesis. A total of 28 papers were reviewed from 2014 to 2024, but a total of 22 papers were selected and conducted excluding duplicate papers and six papers that did not meet the purpose of this study. As a result of the study, an artificial intelligence judgment system was proposed as a new judgment system for Taekwondo Gyeorugi competition. The specific reasons for the proposal were: first, the fairness of the judgment of the competition game and the improvement of audience immersion, and second, the activation of artificial intelligence-related platforms and industries.

본 연구는 공기소총(남ㆍ여) 사격경기의 규정 변경에 따른 중점적으로 요구되는 체력적, 기술적, 심리적 측면을 파악하고, 이에 따른 기술훈련 전략이 어떠한 것이고 어떻게 실천하는지에 대한 분석을 하는 것이 목적이다. 이에 질적 연구방법의 과정에 의해 분석된 결과는 다음과 같다. 첫째, 표적크기의 규정변경에 따른 사격기술훈련의 중점적 전략은 체력적 측면으로서 전문체력과 입사자세의 중심점을 강화시켰다. 또한 유연성을 강화시키고, 사격기술 간의 시간패턴을 빠르게 연결하였다. 둘째, 점수체계의 규정변경에 따른 사격기술훈련의 중점적 전략은 기술적 측면으로서 조준 기술이 있으며, 이는 시각을 분산시키지 않고 사격동작 시 눈의 초점을 간결하게 두어 기계화하였다. 또한 정조준과 2단 스퀴즈 격발이 동시에 사격을 유도하는 격발타이밍을 강화하였다. 셋째, 경기시간의 규정변경에 따른 사격기술훈련의 중점적 전략은 심리적 측면으로서 외부관찰자의 입장에서 자신의 수행을 보는 외적심상훈련과 자신이 신체내부에서 사격장면을 상상하고 통제하는 내적심상훈련을 반복하였다. 또한 광의적 인지행동으로서 몰입을 실천하였고, 협의적 인지행동으로 서 경쟁불안을 극복하는 훈련을 반복하였다.

The purpose of this study is to identify the physical, skill and psychological aspects of the air rifle(m&w) shooting competition about change rules, following the purpose of this study was to analyze the skill training strategy and practice. The process of qualitative methodology led to the following analysis results: First, The main strategy of the shooting skill training according to the regulation of the target size strengthened the professional physical strength and the standing position balance point as the physical aspect. It also enhances flexibility and speedily connects time patterns at each shot interval. second, the focus strategy of the shooting skill training according to the change of the scoring system is the technical aspect of the aiming skill. this was mechanized by keeping the focus of the eyes short while shooting, without distracting the view. also, the main aiming and the second stage squeeze triggered the triggering timing which at the same time triggered the shooting. third, as a psychological aspect, the main strategy of the shooting skill training according to the change of the rules of the competition time was the external imagery training to see the performance of the external observer from the viewpoint of the external observer and the internal imagery training to imagine and control the scene inside the body. also, they practiced immersion as a broad cognitive behavior and overcome competition anxiety as a cooperative cognitive behavior.

Background: A simple binary scoring system (SBSS) was developed and used to assess the welfare measures ofcommercial broiler chickens in South Korea. We also correlated welfare measures with environmental parameters ofbroiler house. Our measures of welfare included lameness, hock burn (HB) and foot pad dermatitis (FPD), whilstenvironmental parameters included air temperature, relative humidity, air speed, light intensity, air quality (inparticular carbon dioxide (CO2) and ammonia (NH3) concentrations) and airborne microbes. Results: The effect of environmental parameters on welfare measures was apparent even on 10-day-old broilers. Anon-parametric correlation analysis revealed significant correlations between environmental parameters and welfaremeasures. The key environmental parameters were relative humidity and light intensity. The results indicate thatthere is a need for proper control of environmental conditions on poultry farms, which could reduce health problemsand subsequently reduce disease and mortality. Conclusions: In conclusion, the simplicity of SBSS makes it preferable over more complex scoring systems and allowsa farmer to more easily assess the welfare measures on their own farm.

This study aims to investigate the effect of the use of AWS as a way of feedback on college students’ development of syntactic ability. Based on the theory of constructivism, a quasi-experimental research was conducted, which lasted one year and involved 80 college students as participants. The students were divided into a treatment class(n=40) and a comparison class(n=40). Writing Roadmap 2.0 was used as the way of feedback for the students in the treatment class and traditional teachers’ feedback as the way of feedback for the students in the comparison class. To guarantee the validity of this research, the students in the treatment class and the students in the comparison class were taught by the same teacher, and the teaching time and teaching syllabus were the same. The syntactic ability was represented by syntactic fluency, accuracy, and complexity of compositions. After one year’s experimental teaching, the students in the treatment class outperformed those in the comparison class in terms of fluency and complexity. As for the accuracy, both classes had similar performance. And compared with traditional teachers’ feedback, the use of AWS has more advantages. Thus, the use of Writing Roadmap 2.0 in college English writing can have a positive effect on the development of college students’ syntactic ability. These findings triggered several pedagogical implications.

The Common Vulnerability Scoring System (CVSS) is a well-known framework for providing the characteristics of a vulnerability and a numerical score reflecting its severity. Since the existing CVSS has been developed for an information technology (IT) environment, it is not suitable for an intelligent building (IB) environment where OT Operational Technology (OT) and Internet of Things (IoT) are integrated. For example, compared to IT systems that prioritize confidentiality, OT systems prioritize safety and have a very long lifespan. Therefore, if a vulnerable component is exploited in an intelligent building environment, the impact may be different from that in the IT environment. This paper identifies problems that may occur when applying the traditional CVSS to an intelligent building environment. We also propose an effective method to extend the CVSS for an intelligent building environment using threat intelligence and rubric.

조직은 고객 정보 유출과 관련된 비즈니스 위험을 최소화하고, 자발적인 사전 검사를 통해 정보 보안 활동을 강화하고 부주의 방치 사고에 의한 개인 정보의 누출을 검출하는 방법을 발견해야 한다. 최근 많은 기업들이 정보유출방지솔루션을 도입하였으나, 업무산 필요에 의한 허용된 권한을 가진 내부 사용자에 의한 유출가능성이 존재한다. 이에 정보취급행위 및 활동에 대한 정보를 수집하여 분석할 수 있는 환경이 필요하다. 본 연구에서는 내부자의 활동 수준을 평가하기 위해서 RFM 모델을 응용한 SFI 분석기법을 활용, 실제 기업에 적용하여 사례 연구를 수행하였다.

The organization shall minimize business risks associated with customer information leaks. Enhance information security activities through voluntary pre-check and must find a way to detect the personal information leakage caused by carelessness and neglect accident. Recently, many companies have introduced an information leakage prevention solution. However, there is a possibility of internal data leakage by the internal user who has permission to access the data. By this thread it is necessary to have the environment to analyze the habit and activity of the internal user. In this study, we use the SFI analytical technique that applies RFM model to evaluate the insider activity levels were carried out case studies is applied to the actual business.

Videofluoroscopic swallowing study (VFS) is accepted as a gold standard tool for evaluating dysphagia. The realtime video radiographic image provides visualization of the structures, movement, and coordination of swallowing. Abnormal oropharyngeal and esophageal anatomy can be readily identified. There are several scoring systems foranalyzing VFS; A Penetration-Aspiration Scale (PAS), The Dysphagia Outcome and Severity Scale (DOSS), NationalInstitutes of Health Swallowing Safety Scale (NIH-SSS) and Videofluoroscopic Dysphagia Scale (VDS). They can providereliable quantification of selected events observed during VFS and be used as powerful outcome measures toinvestigate the efficacy of various swallowing treatments. This article reviews these scoring systems. (JKDS2011;1:55-59)

Background: Intravenous artesunate is the first-line treatment for severe malaria; however, clinical deterioration may occur during treatment. This study aimed to identify risk factors associated with clinical deterioration and to construct a risk-scoring system. Methods: We retrospectively reviewed medical records of adult patients who received at least one dose of intravenous artesunate for malaria at the National Medical Center between January 2014 and December 2023. Clinical deterioration was defined as the new onset of any severe malaria symptom after artesunate administration. Multivariable logistic regression analysis was conducted to identify independent predictors. Results: Among 98 patients included in the final analysis, 29 (29.6%) experienced clinical deterioration. Multivariable analysis revealed that female sex (adjusted odds ratio (aOR) 4.3, 95% confidence interval (CI) 1.0-19.0), age ≥60 years (aOR 3.4, 95% CI 0.8-14.3), presence of jaundice at admission (aOR 4.3, 95% CI 1.1-16.5), and elevated alanine aminotransferase (ALT) level (aOR 2.6, 95% CI 1.0-6.8) were independently associated with increased risk of clinical deterioration. Based on these factors, a risk scoring system (range: 0-5 points) was constructed and showed a stepwise increase in deterioration risk by score. Conclusions: Female sex, advanced age, jaundice, and elevated ALT were independent predictors of clinical deterioration in malaria patients treated with artesunate. The proposed risk score may aid in early identification of high-risk patients and guide clinical decision- making.

지방공기업 경영평가는 중앙정부의 지방공기업에 대한 사후적 통제 방식의 핵심 메커니즘으로 중앙정부가 지향하는 국정 기조에 맞춰 평 가지표와 배점이 탄력적으로 바뀌어 왔으며, 이를 통해 지방공기업이 중앙정부가 원하는 방향으로 운영되도록 유도하는 기능을 한다. 본 연구 는 지방공기업 경영평가 제도의 변화 과정을 살펴보면서 각 지표의 배점 부여 및 전체적인 관점에서 합리적인 배점 관리나 체계적인 접근이 부 족한 것으로 보인다는 문제의식에 기반한다. 본 연구는 이에 대한 대안으로 대표적인 다기준 의사결정 방법인 AHP의 방법론적 특성 및 장점 을 활용하여 합리적인 판단의 준거점이 될 수 있는 지표별 배점 도출 방안을 제시한다. 구체적으로, 공공성과 효율성의 관점에서 각각 평가자 개인의 정성적인 의견을 이원비교 방식을 통해 측정하고, 기하평균을 활용하여 합리적으로 집단의견을 수렴하여 최종적인 지표별 상대적인 중요도를 도출하고, 도출된 결괏값을 준거점으로 합리적인 조정을 거쳐 경영평가 지표별 배점을 산정하고 체계적으로 관리하는 것을 제안한 다. 그리고 본 연구의 개선안을 실무에 적용하기 위해 고려해야 할 사항들에 대해 다룸으로써 제도 개선을 위한 실행가능성을 높이고자 한다.

The performance evaluation of regional public corporations is a core mechanism of the central government's control over these entities. The evaluation criteria and scoring have been flexibly adjusted to align with the central government's policy directions, guiding regional public enterprises to operate in a manner consistent with the central government's objectives. This study examines the changes in the performance evaluation system for regional public corporations and is based on the awareness of issues such as the lack of rational scoring management and systematic approaches in assigning scores to each indicator from an overall perspective. As an alternative, this study proposes a method for deriving indicator-specific scores that can serve as a reference for rational judgment by utilizing the methodological characteristics and advantages of the analytic hierarchy process (AHP), a representative multi-criteria decision-making method. Specifically, the study suggests measuring the qualitative opinions of individual evaluators through pairwise comparisons from the perspectives of public interest and efficiency. This is achieved by rationally aggregating group opinions using geometric means to derive the final relative importance of each indicator, and then calculating and systematically managing the scores for each performance evaluation indicator based on the derived values as reference points. Furthermore, the study addresses considerations necessary for applying the proposed improvements in practice to enhance the feasibility of institutional reform.

우리나라는 금융환경의 변화에도 불구하고 가계대출시 담보대출 선호경향이 지속적으로 나타나고 있는데 이러한 담보위주의 금융관행에서 벗어나기 위해서는 신용위험을 예측하고 측정할 수 있는 가계대출의 신용평가능력의 발전이 시급한 시점이다. 이에 따라 본 연구에서는 영향을 미치는 요인을 분석하고 새로운 신용평가항목을 발굴하여 시스템개선에 기여하는데 그 목적을 두고자 한다. 분석결과, 첫째, 신용카드사용률이 20-29% 대인 사람들이 대출금 연체가 가장 적은 것으로 나타났고 둘째, 대출실행기준에 대한 선호도와 만족도에 있어서는 무보증 대출을 선호하거나 만족하는 사람들이 대출금 연체가 가장 적은 것으로 나타났다. 셋째, 대출금리에 대한 민감도에 있어서도 대출금리와 가산금리에 대해 민감하게 반응하는 사람들, 역시 대출금 연체가 가장 적은 것으로 나타났으나 넷째, 외국여행의 여부는 외국여행 경험이 있는 사람들이 오히려 대출금 연체경험이 있는 것으로 나타나 외국여행의 여부는 대출금 연체와 일정한 관계가 없는 것으로 나타났다. 따라서 외국여행의 여부를 제외한 3개 항목은 신용위험을 측정할 수 있는 것으로 보여 짐에 따라 CSS(개인신용평점제도) 중 BSS(행동평점제도)의 구축에 활용될 수 있을 것이다.

Even after overcoming the financial crisis following the country's crisis in 1997, the preference for mortgage loan of financial companies still remain strong in our country. In oder to deviate from financial customs that was based on the preference for mortgage loan, it was required urgently to develop ability of credit assessment for household loan to predict and measure credit risk. Thus, it is necessary to seek the new effective credit items to measure credit risk. According to the results of the questionnaire, first, people whose use rate of creditcard is about 20 ~29% have fewer overdue bank loan. Second, in relation to the preference and the degree of satisfaction for the standards of bank loan, people who prefer or satisfy with unsecured loan have fewer overdue bank loan. Third, people who respond sensitively to the interest rate of loan also have fewer overdue bank loan. Fourth, considering the experience of overseas travel, the experienced have had overdue bank loan and seems that it is not closely related to overdue bank loan. In conclusion, this study shows that three items except the fourth seems to be able to be used to measure credit risk so as to be applied to build BSS in CSS.

본 연구는 TnF.AI의 온라인 기반 아동 그림 심리분석 시스템 내 사용자 편의를 위해 개발된 SMC(Simple– Moderate–Complex) 채점체계의 신뢰도를 검증하고자 하였다. S시, K도, G시, G도 지역 5~7세 아동의 그림 153장을 대상으로, 전문가 6인이 A, B 두 그룹으로 나뉘어 AI 분석 기반 채점을 수행하였다. A그룹은 동일 그림을 2주 간격으로 두 차례 채점하여 검사-재검사 신뢰도를 검토하였고, A·B 그룹 간 채점 비교를 통해 평가자 간 신뢰 도를 분석하였다. 분석 결과, 모든 평가 영역에서 Weighted Kappa 계수가 0.70 이상으로 나타났으며, 특히 자아 개념과 정서 영역에서 높은 일치도를 보였다. 반면 관계 및 적응, 행동 특성 영역에서는 전문가 해석의 주관성에 따른 일부 차이가 확인되었다. 본 채점체계는 아동 그림의 특성을 수치화·시각화하여 부모가 이해하기 쉬운 결과 제공의 가능성을 제시하며, 향후 타당도 검증과 연령 확장, 전문가 교육 등을 통한 체계적 보완이 필요하다.

This study aimed to verify the reliability of the SMC (Simple–Moderate–Complex) scoring system developed within TnF.AI’s online child drawing analysis platform. A total of 153 drawings by children aged 5 to 7 were evaluated by six experts using AI-based scoring. Group A conducted a retest after two weeks to assess test-retest reliability, while inter-rater reliability was examined by comparing scores between Groups A and B. All evaluation domains showed high Weighted Kappa values above 0.70, though some subjectivity was observed in the domains of relationships and behavior. The scoring system demonstrates potential to provide intuitive and visualized feedback for parents, and further validation, age-range expansion, and expert training are recommended.