비용함수와 파라미터를 이용한 효과적인 디지털 데이터 기계학습 방법론

지상민,박지은,Ji, Sangmin,Park, Jieun 한국디지털정책학회 2021 디지털융복합연구 Vol.19 No.10

Machine learning is the process of constructing a cost function using learning data used for learning and an artificial neural network to predict the data, and finding parameters that minimize the cost function. Parameters are changed by using the gradient-based method of the cost function. The more complex the digital signal and the more complex the problem to be learned, the more complex and deeper the structure of the artificial neural network. Such a complex and deep neural network structure can cause over-fitting problems. In order to avoid over-fitting, a weight decay regularization method of parameters is used. We additionally use the value of the cost function in this method. In this way, the accuracy of machine learning is improved, and the superiority is confirmed through numerical experiments. These results derive accurate values for a wide range of artificial intelligence data through machine learning. 기계학습은 학습에 이용되는 학습 데이터와 데이터를 예측할 인공신경망을 이용하여 비용함수를 만들고, 비용함수를 최소화시키는 파라미터들을 찾는 과정이다. 파라미터들은 비용함수의 그래디언트 기반 방법들을 이용하여 변화하게 된다. 디지털 신호가 복잡할수록, 학습하고자 하는 문제가 복잡할수록, 인공신경망의 구조는 더욱 복잡해지고 깊어진다. 복잡하고, 깊어지는 인공신경망 구조는 과적합(Over-fitting) 문제를 발생시킨다. 과적합 문제를 해결하기 위하여 파라미터의 가중치 감소 정규화 방법이 사용되고 있다. 우리는 이러한 방법에서 추가로 비용함수의 값을 이용한다. 이러한 방법으로 기계학습의 정확도가 향상되는 결과를 얻었으며 이는 수치 실험을 통하여 우수성이 확인된다. 이러한 결과는 기계학습을 통한 인공지능의 폭넓은 데이터에 대한 정확한 값을 도출한다.

그림자 효과를 고려한 태양전지 모듈의 발전량 예측 연구

김민수(Minsu Kim),지상민(Sangmin Ji),오수영(Soo Young Oh),정재학(Jae Hak Jung) 한국태양광발전학회 2016 Current Photovoltaic Research Vol.4 No.2

Since the last five years it has become a lot of solar power plants installed. However, by installing the large-scale solar power station it is not easy to predict the actual generation years. Because there are a variety of factors, such as changes daily solar radiation, temperature and humidity. If the power output can be measured accurately it predicts profits also we can measure efficiency for solar power plants precisely. Therefore, Prediction of power generation is forecast to be a useful research field. In this study, out discovering the factors that can improve the accuracy of the prediction of the photovoltaic power generation presents the means to apply them to the power generation amount prediction.

두유박 첨가 먹이원 급이에 따른 갈색거저리 유충의 영양성 증진 효과

김소윤(So-Yun Kim),박민지(Min Ji Park),송정훈(Jeong-Hun Song),지상민(Sangmin Ji),장규동(Gyu-Dong Chang),김선영(Sun Young Kim) 한국생명과학회 2024 생명과학회지 Vol.34 No.3

두유박 첨가 먹이원 급이 유충의 영양학적 우수성과 안전성을 검증하고자 밀기울로 사육한 갈색거저리 유충의 영양성분과 유해물질을 비교분석하였다. 건조중량 기준 조단백질 함량은 10% 두유박 첨가 급이군(SR)에서 54.0%로 밀기울 급이 대조군(WB) 43.5%보다 1.2배 더 많았다. 식이섬유 또한 SR에서 4.9%로 2.5%보다 1.9배 더 많은 경향을 나타냈다. 불포화지방산 중 리놀레산 함량은 SR (32.9%)에서 WB (29.0%)보다 1.1배 더 많은 것으로 나타났다. 다량무기질 중 가장 많은 함량을 나타낸 칼륨은 WB (1,074.5 mg/100g)에서 SR (1,014.0 mg/100 g) 보다 1.1배 더 많았다. 미량무기질 중 아연은 SR (14.5 mg/100 g)에서 WB (11.9 mg/100 g)에 비해 1.2배 더 많은 함량을 나타냈다. WB와 SR의 유해물질 분석 결과, 중금속 중 납, 카드뮴, 무기비소는 식용곤충 중금속 기준에 적합하였고, 식중독균인 대장균, 살모넬라균은 모든 군에서 검출되지 않았다. 위의 영양성분 및 유해물질 분석 결과로 볼 때, 10% 두유박 첨가 급이 갈색거저리 유충은 풍부한 영양성분을 가지고 있으며, 안전성 또한 입증되었으므로 식품 및 사료 원료로 이용하기에 좋은 소재로 판단된다. To verify the nutritional value and safety of larvae fed with a soymilk residue-added food source, we compared and analyzed the nutritional components of and harmful substances in Tenebrio molitor larvae reared on wheat bran. Crude protein content based on dry weight was 1.2 times higher in the 10% soymilk residue-fed group (SR) (54.0%) than in the wheat bran-fed group (WB) (43.5%). Dietary fiber also tended to be 1.9 times more in the SR (4.9%) than the WB (2.5%). Among unsaturated fatty acids, the linoleic acid content was found to be 1.1-fold higher in the SR (32.9%) than in the WB (29.0%). Potassium, which was the most abundant among the macro minerals, was 1.1-fold more abundant in the WB (1,074.5 mg/100 g) than in the SR (1,014.0 mg/100 g). Among the micro minerals, zinc content was 1.2-fold higher in the SR (14.5 mg/100 g) than in the WB (11.9 mg/100 g). The results of the analysis of hazardous substances in the WB and SR revealed that the amount of heavy metals met the standards for heavy metals in edible insects, and food poisoning-inducing bacteria such as Escherichia coli and Salmonella spp. were not detected in all groups. These results indicate that mealworm larvae fed with 10% soymilk residue have abundant nutrients and are safe for intake. Thus, food sources with added soymilk residue has the potential to be used as feed ingredients.

정서곤충을 활용한 치유 연구현황과 미래전략

김소윤(So-Yun Kim),송정훈(Jeong-Hun Song),지상민(Sangmin Ji),김원태(Wontae Kim) 한국식품과학회 2022 식품과학과 산업 Vol.55 No.2

The purpose of this study is to organize research trends by analyzing healing studies using emotional insects, and to suggest future research directions. As a result, only six emotional insect species were used for healing purposes, and most of the age groups of the test subjects were elementary school students and the elderly. As for the type of research, field application of the healing program was dominant at 76.5%. Effectiveness verification was mainly conducted through questionnaires before and after participating in the healing program. In some cases, stress-related salivary hormones were used as indicators or clinical tests were conducted by medical staff. Based on the results of this analysis, future strategies such as finding new emotional insect species, developing various types of healing program models, and expanding and subdividing the target subjects were suggested to promote the growth of the healing industry using emotional insects in the future.

곤충자원을 이용한 농가형 치유농업프로그램의 치유효과 분석

김소윤(So-Yun Kim),김선영(Sun Young Kim),최현명(Hyun-Myung Choi),지상민(Sangmin Ji) 한국농촌지도학회 2023 농촌지도와 개발 Vol.30 No.1

This study aims to examine the effect of healing program using insect. For the insect resources, the program uses 3 species of emotional insects. The subjects were a total of 29, the elderly with mild dementia and adults with developmental disabilities in the facility. The activities are 9 programs in total and each program takes 2 hours once a week. For data collection, we measured a blood pressure, stress level using autonomic nervous system activity, and scale of happiness, depression and insomnia by a survey. The datas were collected every week before and after the program. As a result of paired-sample t-test analysis, the participants of the agro-healing program using insects had positive effects as follows: reducing a depression, relieving a symptom of insomnia and stabilizing a blood pressure. Through the results of this study, it was possible to confirm the positive effects of the agro-healing program using insects. If farms lead agro-healing program using insects on their own, it could be expected to contribute to improve the quality of life through agriculture as well as to improve the farms' income.

곤충 사육농가의 경영성과에 영향을 미치는 요인

김소윤(So-Yun Kim),송정훈(Jeong-Hun Song),지상민(Sangmin Ji),김원태(Wontae Kim) 한국농촌지도학회 2021 농촌지도와 개발 Vol.28 No.1

It is important to understand the factors that affect the business performance of insect farm for continuous insect farm management. The purpose of this study is to investigate factors influencing the business performance of insect farm. For this study, 1,577 questionnaires were collected through a telephone survey targeting insect farm owner. As a result of analysis using linear multiple regression analysis, the factors affecting total sales were gender, age, business experience, number of workers, and national and local government support projects. The factors affecting the net profit rate were age, business experience, number of workers, national and local government support projects, and education. When the gender of the business operator is male, it only affected the increase in total sales, and it was found that both the total sales amount and the net profit margin increased with the younger the business operator’s age.