• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.33 no.6
• /
• pp.246-256
• /
• 2021

Various development projects occurring on the coast cause an imbalance of surface sediments, causing coastal disasters or irreversible coastal erosion. Coastal erosion caused by the influence of various port structures built through coastal development can be directly identified by evaluating changes in the sediment budget, longshore sediment, and cross-shore sediment. In other words, it will be possible to evaluate the causality between coastal development and coastal erosion by classifying regions due to single cause and regions due to multiple causes according to the changes in the sediment classified into the three types mentioned above. In this study, the cause of long-term and continuous erosion was analyzed based on the analysis results of the coastal development history and the Coastal Erosion Monitoring targeting the coast of Gangwon-do and Gyeongsangbuk-do on the east coast. In addition, in order to evaluate the degree of erosion caused by the construction of artificial coastal structures, the concept of erosion impact assessment was established, three methods were proposed for the impact assessment. The erosion impact of Hajeo port was assessed using the results of satellite image analysis presented in the Coastal Erosion Monitoring Report, it was assessed that the development of Hajeo port had an impact of 93.4% on erosion, and that of the coastal road construction had an impact of 6.6%.

• Journal of Digital Convergence
• /
• v.20 no.2
• /
• pp.241-250
• /
• 2022

Due to the COVID-19 pandemic, with increased 'untact' services and with unstable household economy, the bike insurance fraud is expected to surge. Moreover, the fraud methodology gets complicated. However, the fraud detection model for bike insurance is absent. we deal with the issue of skewed class distribution and reflect the criterion of fraud detection expert. We utilize a balanced random-forest algorithm to develop an efficient bike insurance fraud detection model. As a result, while the predictive performance of balanced random-forest model is superior than it of non-balanced model. There is no significant difference between the variables used by the experts and the confirmatory models. The important variables to detect frauds are turned out to be age and gender of driver, correspondence between insured and driver, the amount of self-repairing claim, and the amount of bodily injury liability.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.4
• /
• pp.173-178
• /
• 2023

As Korean literature spreads around the world, its position in the overseas publishing market has become important. As demand in the overseas publishing market continues to grow, it is essential to predict future book sales and analyze the characteristics of books that have been highly favored by overseas readers in the past. In this study, we proposed ensemble learning based prediction model and analyzed characteristics of the cumulative sales of more than 5,000 copies classified as good sellers published overseas over the past 5 years. We applied the five ensemble learning models, i.e., XGBoost, Gradient Boosting, Adaboost, LightGBM, and Random Forest, and compared them with other machine learning algorithms, i.e., Support Vector Machine, Logistic Regression, and Deep Learning. Our experimental results showed that the ensemble algorithm outperforms other approaches in troubleshooting imbalanced data. In particular, the LightGBM model obtained an AUC value of 99.86% which is the best prediction performance. Among the features used for prediction, the most important feature is the author's number of overseas publications, and the second important feature is publication in countries with the largest publication market size. The number of evaluation participants is also an important feature. In addition, text mining was performed on the four book reviews that sold the most among good-selling books. Many reviews were interested in stories, characters, and writers and it seems that support for translation is needed as many of the keywords of "translation" appear in low-rated reviews.

• Journal of Internet Computing and Services
• /
• v.25 no.4
• /
• pp.47-56
• /
• 2024

This paper proposes a method for fine-tuning BERT-based natural language processing models to perform sentiment analysis on Korean review data. By varying the input sequence length during this process and comparing the performance, we aim to explore the optimal performance according to the input sequence length. For this purpose, text review data collected from the clothing shopping platform M was utilized. Through web scraping, review data was collected. During the data preprocessing stage, positive and negative satisfaction scores were recalibrated to improve the accuracy of the analysis. Specifically, the GPT-4 API was used to reset the labels to reflect the actual sentiment of the review texts, and data imbalance issues were addressed by adjusting the data to 6:4 ratio. The reviews on the clothing shopping platform averaged about 12 tokens in length, and to provide the optimal model suitable for this, five BERT-based pre-trained models were used in the modeling stage, focusing on input sequence length and memory usage for performance comparison. The experimental results indicated that an input sequence length of 64 generally exhibited the most appropriate performance and memory usage. In particular, the KcELECTRA model showed optimal performance and memory usage at an input sequence length of 64, achieving higher than 92% accuracy and reliability in sentiment analysis of Korean review data. Furthermore, by utilizing BERTopic, we provide a Korean review sentiment analysis process that classifies new incoming review data by category and extracts sentiment scores for each category using the final constructed model.

• Journal of Sasang Constitutional Medicine
• /
• v.8 no.1
• /
• pp.395-411
• /
• 1996

1. 연구배경 사상체질의학(四象體質醫學)을 창시하여 개인(個人)의 차별성(差別性)을 강조한 동무(東武) 이제마(李濟馬)는 양생(養生)의 방법(方法)에서도 체질별(體質別) 요법(療法)을 말하고 있는데 체질별(體質別)로 과소지장(過小之臟)의 기능(機能)이 정상적(正常的)으로 이루어지는 상황을 완실무병(完實無病)의 조건으로 제시(提示)하였고 이를 위한 수단(手段)으로 성정(性情)과 함께 약물(藥物), 식품(食品) 등을 이용하였다. 특히 식이요법(食餌療法)에 있어서도 체질(體質)에 따른 구별(區別)의 필요성(必要性)을 말하고 있는데 식품(食品)이라 하더라도 그 음식(飮食)을 섭취하여 과대(過大)한 장기(臟器)의 기능(機能)은 유제(柳制)하고 과소(過小)한 기능(機能)은 보완(補完)받음으로써 불균형(不均衡)을 조정(調整)한 것이다. 당뇨병의 식단 작성은 평생동안 열량(熱量)과 영양소(營養素) 필요치(必要置)을 맞출 것을 권장하고 당뇨병학회에서 편집한 식품교환표(食品交換表)를 사용(使用)하는 것이 일반적(一般的)인데 식품교환표(食品交換表)는 많은 식품(食品)들중에 같은 영양소를 가진 식품(食品)들을 한 그룹으로 묶어 환자(患者)의 기호(嗜好)에 따라 교환(交煥)해 가면서 먹을 수 있도록 고안(考案)한 것이니 이에 지시한 수량(數量)만 섭취해도 저(低)cal식(食)으로 관양(管養)의 균형(均衡)이 잘 이루어진다. 본 연구는 체질별로 이로운 식품표에 근거하여 식이요법(食餌療法)이 특히 강조되고 하루 섭취열량이 제한되는 성인병중의 하나인 당뇨병(糖尿病)의 식단(1800kcal)을 식단작성법에 따라 구성(構成)하여 몇가지 예를 제시해 보았다. 구체적으로 태음인(太陰人), 소음인(少陰人), 소양인(少陽人)의 당뇨 환자 1800kcal에 대한 식단을 구성하여 제시했는데 즉, 태음인(太陰人)의 식단은 태음인(太陰人)에 유리(有利)한 식품(食品)들로 구성하고 해(害)로운 식품(食品)들은 제외시키는 방법(方法)을 이용하였다. 이 식단은 다분히 이론적(理論的)인 식단으로 임상(臨床)에 이용(利用)하여 본 바는 없으나 동량(同量)의 열량(熱量)을 섭취(攝取)하더라도 체질(體質)에 적합(適合)한 식품(食品)으로 구성된 식사(食事)가 각 체질의 섭생(攝生)에 더 유리(有利)하지 않올까 하는 단순(單純)한 사고(思考)에 바탕을 둔 것이다. 2. 연구방법 1) 후세가(後世家)가 주장(主張)한 체질별(體質別) 식품(食品) 분류(分類)를 종합, 정리한 체질별(體質別) 식품표(食品表)를 제시한다. 박석언의 동의사상대전, 박인상의 동의사상요결, 송일병의 알기 쉬운 사상의학, 홍순용의 사상진료보원, 홍순용, 이을호의 사상의학원론에서 체질별로 유익한 식풍을 조사하여 곡류, 과일류, 채소류, 어패류, 육류로 분류하여 살펴본다. 2) 당뇨병(糖尿病) 식이요법의 식단 작성법의 개요(槪要)를 제시한다. 3) 1)의 체질별(體質別) 식품표(食品表)로 태음인(太陰人), 소음인(少陰人), 소양인(少陽人)의 당뇨 식단 1800kcal을 작성해 제시(提示)한다. 체질별(體質別)로 유익(有益)한 식품(食品)은 1)의 식품표에 근거(根據)하고 체질별(體質別)로 해(害)로운 식품(食品)은 노정우(盧正祐), 한동석(韓東錫)의 주장에 근거(根據)한다. 3. 결과 체질별(體質別) 식품표(食品表)는 후세가의 연구를 종합하여 제시(提示)하였고, 식품(食品)을 분류(分類)한 후(後) 약명(藥名)과 성미(性味), 귀경(歸經)을 찾아 도표화 하였다. 체질별 식품들은 대부분 소음인(少陰人)의 경우 신감(辛甘) 온열(溫熱)하며 비위(脾胃)로 귀경(歸經)하고 태음인(太陰人)의 경우 감신(甘辛) 온열(溫熱)하며 폐간(肺肝)으로 귀경(歸經)하고 소양인(少陽人)의 산고(酸苦) 양한(凉寒)하고 신(腎)으로 귀경(歸經)함이 우세(優勢)함을 알 수 있다. 즉, 체질적으로 양성(陽性)인 소양인(少陽人)은 식품의 성질이 음성(陰性)인 것이 유리(有利)하고 체질적으로 음성(陰性)인 태음인(太陰人), 소음인(少陰人)은 식품의 성질이 양성(陽性)인 것이 유리(有利)하다. 다양한 식품(食品)을 섭취하고자 하는 환자의 욕구(慾求)에 맞추면서도 식품교환의 범위를 체질별로 유익한 식품들로 제한하여 동일(同一)한 열량(熱量)의 식단이라도 체질에 맞는 식품으로 차별성(差別性)을 두었는데 식단의 작성은 전문 영양사의 의견을 거쳤다. 제시된 식단은 다소 이론적(理論的)으로 작성(作成)된 단계이고 임상적(臨床的) 검증을 거친 바 없으나 활용하기에 따라 실용성을 얻을 수 있으리라 본다. <식단예> 태음인의 식단: 곡류 : 콩, 율무, 밀가루, 밀, 수수, 들깨, 고구마, 땅콩, 기장, 옥수수, 두부, 설탕등 태음인에 유리한 식품으로 교환한다 어때류 : 우렁이, 대구, 조기, 민어, 청어, 오정어, 낙지, 미역, 김, 다시마등으로 교환한다 육류 : 소고기, 우유등으로 교환한다 과일류 : 밤, 배, 호도, 은행, 잣, 살구, 매실, 자두등으로 교환한다 채소류 : 무우, 도라지, 연근, 토란, 마, 고사리, 더덕, 목이버섯, 송이버섯, 석이버섯등으로 교환한다 해로운 음식 : 닭, 돼지, 모밀, 배추, 사과, 염소고기, 조개, 계란, 곳감, 커피등은 피한다 * 아침 ; 콩나물죽, 대구포묶음, 우령이무침, 갓김치, 우유, 자두 점심 ; 기장밥, 콩나물두부찌게, 장어양념구이, 도라지나물, 열무김치, 배 저녁 ; 수수밥, 두부명란, 더덕양념구이, 깍두기 * 아침 ; 비빔국수, 토란국, 알타리김치, 두유, 살구주스 점심 ; 율무밥, 낙지전골, 김무생채, 느타리나물무침, 동치미, 귤 저녁 ; 콩밥, 감자북어국, 두부묶음, 열무김치 소음인의 식단: 곡류 : 찹쌀, 좁쌀, 차조, 감자등 소음인에 유익한 식품으로 교환한다 어패류 : 명태, 미꾸라지, 뱀장어, 뱀, 메기등 육류 : 닭, 개, 꿩, 염소, 양, 참새고기등 과일류 : 사과, 귤, 복숭아, 대추등 채소류 : 미나리, 파, 마늘, 후추, 시금치, 양배추, 생강, 고추, 당근, 양파, 감자, 쑥갓등 해로운 음식 : 메밀, 호도, 계란, 고구마, 녹두, 돼지고기, 밤, 배, 배추, 보리, 쇠고기, 수박, 오이, 참외, 팥등은 피한다. * 아침 ; 찰밥, 닭찜, 감자전, 쑥갓나물, 부추김치, 사과 점심 ; 감자밥, 메기매운탕, 명태조림, 미나리, 고들빼기김치, 사과주스 저녁 ; 좁쌀밥, 양배추감자국, 병어양념구이, 연근양념조림, 귤, 인삼차.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.42 no.4
• /
• pp.30-36
• /
• 2014

This study was conducted to determine the causes of unreasonable prices and small numbers of azalea cultivars by analyzing the price trends and the number of azalea cultivars announced over the last 25 years based on data from the Public Procurement Service(PPS), Korea Price Research Center and the Landscaping Tree Association(LTA)(hereinafter, officially announcing agencies and organizations) which are major references used when landscape planting is decided. The prices of azalea cultivars announced by the official announcing agencies and organizations have moved in similar patterns over the past 25 years because the prices of azalea cultivars announced by the LTA were referred to by other official announcing agencies and organizations when they officially announced the prices of azalea cultivars. The PPS set lower officially fixed prices of azalea cultivars compared to other official announcing agencies and organizations, and the reason for this is considered to be the intention of the PPS to suppress landscape tree price increases because of the government's policies to suppress price increases. The prices of azalea cultivars seem to change rapidly due to the imbalance between the demand and supply of azalea cultivars rather than the effects of consumer price fluctuation rates because the production periods of azalea cultivars are shorter when compared to other landscape trees. The prices of azalea cultivars from the official announcing agencies and organizations have been set higher than the prices in actual transactions. The reason for this is considered to be the intention of the official announcing agencies and organizations to allow landscaping companies to cover defect costs resulting from the practice of subcontracting planting work and secure profits of subcontractors for planting work. The official announcing agencies and organizations have simply announced prices of 5~8 main azalea cultivars that have been used in the past. The names of azalea cultivars being cultivated and criteria for classification have not been clear; thus, landscape designers have not written clear names of azalea cultivars to be cultivated on planting drawings as practice and landscapers planted those azalea cultivars which could be easily obtained. Therefore, it is assumed that there has been no demand for new azalea cultivars. Thus, the vicious circle in which the prices of only those azalea cultivars that were produced in the past have been announced is repeated.

• Journal of Nutrition and Health
• /
• v.6 no.1
• /
• pp.39-46
• /
• 1973

This study was projected to get basic data which can provide a basis for future direction in nutritional education, and also to find the way how to improve the nutritional supply by evaluating the current nutritional intake of average high school students through the survey study of their daily packed lunch. Five hundred twenty seven students from two boys high school and two girls high school including one general and one vocational school respectively were chosen as random sampling technique. Four hundred forty nine among the 527 students had brought lunch. The contents of lunch box were weighed and converted into nutritional values according to the food composition table and compared with recommended dietary allowances. The results compared and classified by sex, School and housewives' educational level were as follows: 1. The nutritional supply in the lunch box was 671 Cal of energy and 22.3 gm of protein for male students which were respectively 55.9% and 74.2% of the dietary recommendations. On the other side female student's lunch boxes were found to contain 495 Cal of energy and 21.3gm of protein which are respectively 61.8% and 80% of the dietary prescriptions. Excluding niacin, all vitamins and minerals were found to be short. 2. Calorie intake in the vocational high school was found to be higher than in the general high school but lower in protein intake especially significant difference (P<0.01) in animal protein. 3. From the nutritional point of view the educational backgrouud of the housewives was not found to have any influence in the way of preparing the lunch boxes. 4. Nutrients of lunch box were heavily inclined to grain rather than to side dishes.

• Journal of Nutrition and Health
• /
• v.52 no.4
• /
• pp.383-398
• /
• 2019

Purpose: This study investigated an association between dietary quality and use of convenience foods at convenience stores among high school students. Methods: A total of 474 high school students (225 boys and 249 girls) residing in Incheon participated in this questionnaire survey in June 2018. The subjects were divided into three groups according to the frequency of consumption of convenience foods at convenience stores; less than once a week, 1 ~ 2 times a week, and more than 3 times a week. Dietary quality was assessed using a nutrient quotient for adolescents (NA-Q). Logistic regression was used to investigate an association between dietary quality and use of convenience foods at convenience stores among high school students. Results: For boys and girls, higher monthly allowance was significantly associated with the higher frequency of consumption of convenience foods at convenience stores, whereas school grade, mother's occupational status, family size, extracurricular study, and eating speed were not significantly associated with the frequency of consumption of convenience foods. Higher intake frequency of cookies or sweet and greasy bread, processed beverage, Ramyon, night-time snack, and street food was significantly associated with the higher frequency of consumption of convenience foods for boys or girls. Boys and girls, who had a higher frequency of consumption of convenience foods at convenience stores had significantly greater odds for being in the low grade of dietary quality, especially in the moderation factor. Conclusion: The students who used convenience stores more often appeared to have more monthly allowance and to consume undesirable foods more often. Higher frequency of using convenience foods at convenience stores among high school students was associated with lower dietary quality. These study results can support efforts to provide nutrition education programs and guidelines to students who frequently use convenience foods at convenience stores.

• Journal of Korea Water Resources Association
• /
• v.57 no.3
• /
• pp.181-193
• /
• 2024

The increasing atmospheric imbalance caused by climate change leads to an elevation in precipitation, resulting in a heightened frequency of flooding. Consequently, there is a growing need for technology to detect and monitor these occurrences, especially as the frequency of flooding events rises. To minimize flood damage, continuous monitoring is essential, and flood areas can be detected by the Synthetic Aperture Radar (SAR) imagery, which is not affected by climate conditions. The observed data undergoes a preprocessing step, utilizing a median filter to reduce noise. Classification techniques were employed to classify water bodies and non-water bodies, with the aim of evaluating the effectiveness of each method in flood detection. In this study, the Otsu method and Support Vector Machine (SVM) technique were utilized for the classification of water bodies and non-water bodies. The overall performance of the models was assessed using a Confusion Matrix. The suitability of flood detection was evaluated by comparing the Otsu method, an optimal threshold-based classifier, with SVM, a machine learning technique that minimizes misclassifications through training. The Otsu method demonstrated suitability in delineating boundaries between water and non-water bodies but exhibited a higher rate of misclassifications due to the influence of mixed substances. Conversely, the use of SVM resulted in a lower false positive rate and proved less sensitive to mixed substances. Consequently, SVM exhibited higher accuracy under conditions excluding flooding. While the Otsu method showed slightly higher accuracy in flood conditions compared to SVM, the difference in accuracy was less than 5% (Otsu: 0.93, SVM: 0.90). However, in pre-flooding and post-flooding conditions, the accuracy difference was more than 15%, indicating that SVM is more suitable for water body and flood detection (Otsu: 0.77, SVM: 0.92). Based on the findings of this study, it is anticipated that more accurate detection of water bodies and floods could contribute to minimizing flood-related damages and losses.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.39 no.6
• /
• pp.826-836
• /
• 2010

This study was conducted to compare the dietary habits, nutrient intake and health status of female college students at Sahmyook University according to their majors. Specifically, women majoring in literature and science (77), food and nutrition (103) and sport (73) were evaluated. College women in the sports department were more likely to have a part-time job and had greater expenses than women in the other departments. The average height of college women in the sports department (164.3${\pm}$4.6 cm) was 2.04 cm taller than that of women with other majors (162.3${\pm}$4.7 cm). College women in the department of literature and science were more likely to have an unbalanced diet, even though they commonly ate small amounts of fruit as snacks. They were more prone to take nutrient tablets and vitamins when compared to women in the other departments. College women in the department of sport were more likely to have unbalanced meals (31.5%) and to overeat. Students in the department of food and nutrition ate more fruit, vitamin C and E but less cholesterol containing foods (p<0.05), less fast food and fried food than students in the other departments. The subjects in the department of sport ate less bread, sweet potatoes, fast foods and fried foods but more calories, fat, vitamin A, vitamin B, niacin, Ca, P and cholesterol than students in the other departments (p<0.05). They were also more likely to exercise for more than two hours a day. The most common problems among college women were going without meals, eating an unbalanced diet, overeating, intake of ill-balanced nutrients and lack of exercise. It was found that college women in the department of sport had a better intake of nutrients and maintained healthier life styles.