본 연구는 서바이벌 게임에서 맵의 디자인 요소를 분석하고 표준화하는 것을 목표로 했다. 상 용 게임 10개를 조사하여 네 가지 주요 맵 유형을 파악하였고, 이를 바탕으로 유니티 에디터를 사용해 네 가지 유형의 맵을 개발 및 구현하였다. 무작위 위치에 장애물을 스폰하고, 타일 재 배치 시스템을 통해 맵을 무한대로 재배치할 수 있도록 했다. 또한, Update() 메서드를 조정해 타일 변위 오류를 해결했다. 마지막으로, ML 에이전트를 사용해 맵을 테스트했으며, 에이전트 가 환경을 탐색하고 장애물과 적대 대상을 인식 및 공격하며 각기 다른 네 가지 맵에서 효과 적으로 대응하는 모습을 확인했다.

The alternative food market has attracted much attention due to concerns about climate change, increasing consumer awareness of value consumption, and the development of the Food-Tech industry. This study aimed to analyze the factors that drive consumers to purchase alternative food products and identify the mechanisms that can induce consumers to continue buying. We surveyed 1,200 consumers and estimated data using a logit model. The results showed that the presence of vegetarians in the household, environmental concerns, a vegetable-oriented diet, and a mixed diet of meat and vegetables were positively associated with purchasing plant-based alternatives. Of particular interest was the non-linear relationship between respondents’ age and their purchasing experience with plant-based options, with the likelihood of purchasing alternatives increasing with age from the mid-50s onward. These findings suggest that in addition to the growing number of consumers who share the environmental value, which leads to increased interest in and purchase of plant-based alternatives, alternative foods are also health-oriented and meet the needs of older consumers, who are becoming an increasingly important segment of the super-aged society, suggesting the potential for continued growth in the alternative food market.

실제세계에서 데이터 수집의 비용과 한계를 고려할 때, 시뮬레이션 생성 환경은 데이터 생성 과 다양한 시도에 있어 효율적인 대안이다. 이 연구에서는 Unity ML Agent를 로그라이크 장 르에 적합한 강화학습 모델로 구현하였다. 간단한 게임에Agent를 이식하고, 이 Agent가 적을 인식하고 대응하는 과정을 코드로 작성하였다. 초기 모델은 조준사격의 한계를 보였으나 RayPerceptionSensor-Component2D를 통해 Agent의 센서 정보를 직접 제공함으로써, Agent가 적을 감지하고 조준 사격을 하는 능력을 관찰할 수 있었다. 결과적으로, 개선된 모델 은 평균3.81배 향상된 성능을 보여주었으며, 이는 Unity ML Agent가 로그라이크 장르에서 강화학습을 통한 데이터 수집이 가능함을 입증한다.

본 연구는 주변 환경의 차이에 따른 화분매개곤충의 유입 특성을 파악하기 위하여 국립수목원 내 진화속을걷 는정원과 부추속전문전시원에 식재된 울릉산마늘의 화분매개곤충을 조사하였다. 2023년 5월 22일부터 6월 2일 까지 꽃이 70% 이상 개화하였을 때 포충망을 활용하여 8일간 곤충을 채집하였고, 각 전시원 별 식생(피도), 기후 (온도·습도·조도)를 조사하였다. 조사 결과 진화속을걷는정원에서 피도 60% 온도 26.4℃, 습도 31.5%, 조도 40953.6lx, 화분매개곤충 20과 450개체, 부추속전문전시원은 피도 90%, 온도 25.6℃, 습도 31.6%, 조도 6387lx, 화분매개곤충 15과 196개체로 나타났다. 온도와 조도가 상대적으로 높은 진화속을걷는정원이 채집된 곤충의 다양성과 방문 빈도가 높았다. 시간대별 곤충의 방문 빈도를 비교해본 결과 온도와 조도는 개체수가 증가할 때 같이 증가하는 경향을 보였으며, 습도는 반대의 경향을 보였다.

Recently, the importance of preventive maintenance has been emerging since failures in a complex system are automatically detected due to the development of artificial intelligence techniques and sensor technology. Therefore, prognostic and health management (PHM) is being actively studied, and prediction of the remaining useful life (RUL) of the system is being one of the most important tasks. A lot of researches has been conducted to predict the RUL. Deep learning models have been developed to improve prediction performance, but studies on identifying the importance of features are not carried out. It is very meaningful to extract and interpret features that affect failures while improving the predictive accuracy of RUL is important. In this paper, a total of six popular deep learning models were employed to predict the RUL, and identified important variables for each model through SHAP (Shapley Additive explanations) that one of the explainable artificial intelligence (XAI). Moreover, the fluctuations and trends of prediction performance according to the number of variables were identified. This paper can suggest the possibility of explainability of various deep learning models, and the application of XAI can be demonstrated. Also, through this proposed method, it is expected that the possibility of utilizing SHAP as a feature selection method.

To increase the selection efficiency for new cultivars in cut roses during breeding programs, phenotypes of seedlings in two cross combinations, C1 (Rosa hybrida ‘Yellow King’ x ‘Aladin’) and C2 (R. hybrida ‘Evelien’ x ‘Sunny Isle’), were investigated. Although most seedlings showed classic-type flower shapes, some (0.4% - 0.7%) had tomato-shaped flowers, especially at the young plant stage (6 months old). The variation in petal color was high, and pink and pink-mixed flowers were dominant, accounting for 72% in C1 and 76% in C2 at the mature plant stage (13-month-old plants). However, the seedlings had changes in the expression of petal color at different growth stages. The number of petals at the young plant stage had a strong positive correlation with those at the mature plant stage in C1 (r = 0.84***) and C2 (r = 0.73***). Sizes of flowers and petals increased with the growth of the seedlings, and their traits at the young plant stage had positive correlations with those at the mature plant stage in both C1 (r = 0.32***) and C2 (r = 0.39***). Prickle expression on stems was mostly in level 5 at the young plant stage and then dispersed to various classes at the mature plant stage, which increased the variation and coefficient of variation. To select seedlings for new cultivars through breeding processes, breeders should consider the growth stage for targeted traits.

Recently, a study of prognosis and health management (PHM) was conducted to diagnose failure and predict the life of air craft engine parts using sensor data. PHM is a framework that provides individualized solutions for managing system health. This study predicted the remaining useful life (RUL) of aeroengine using degradation data collected by sensors provided by the IEEE 2008 PHM Conference Challenge. There are 218 engine sensor data that has initial wear and production deviations. It was difficult to determine the characteristics of the engine parts since the system and domain-specific information was not provided. Each engine has a different cycle, making it difficult to use time series models. Therefore, this analysis was performed using machine learning algorithms rather than statistical time series models. The machine learning algorithms used were a random forest, gradient boost tree analysis and XG boost. A sliding window was applied to develop RUL predictions. We compared model performance before and after applying the sliding window, and proposed a data preprocessing method to develop RUL predictions. The model was evaluated by R-square scores and root mean squares error (RMSE). It was shown that the XG boost model of the random split method using the sliding window preprocessing approach has the best predictive performance.