• 한국융합학회논문지
• /
• 제12권12호
• /
• pp.83-91
• /
• 2021

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.

• Clinical and Experimental Pediatrics
• /
• 제64권7호
• /
• pp.347-354
• /
• 2021

Background: Neutrophil gelatinase-associated lipocalin (NGAL) has emerged as a valuable biomarker of urinary tract infection (UTI) in children. Purpose: This study aimed to compare the diagnostic accuracy of urinary NGAL (uNGAL) with those of serum C-reactive protein (CRP) and white blood cell (WBC) count for predicting UTI and acute pyelonephritis (APN) in febrile children. Methods: The medical charts of children undergoing uNGAL measurements between November 2017 and August 2019 were retrospectively reviewed. Patients with a suspected or diagnosed UTIs were included. The diagnostic accuracies of uNGAL, serum CRP, and WBC count for detecting UTI and APN were investigated. Independent predictors of UTI and APN were investigated using multivariable logistic regression analyses. Results: A total of 321 children were enrolled in this study. The uNGAL levels were higher in the UTI group (n=157) than in the non-UTI group (n=164) (P<0.05). Among children with a UTI, uNGAL levels were higher in the APN group (n=70) than, the non-APN group (n=87) (P<0.05). In the multivariate analysis, uNGAL was independently associated with UTI and APN (both P<0.05). Serum CRP and WBC count were not correlated with the presence of UTI and APN. Receiver operating curve analyses showed that the uNGAL level had the highest area under the curve (AUC) for predicting UTI and APN, respectively (AUC, uNGAL vs. CRP vs. WBC count, 0.860 vs. 0.608 vs. 0.669 for UTI; 0.780 vs. 0.680 vs. 0.639 for APN, all P<0.05, respectively). The predictive values and likelihood ratios of uNGAL were superior to those of serum CRP and WBC count for detecting UTI and APN at each cutoff level. Conclusion: UNGAL may be more useful than serum CRP and WBC count for identifying and assessing UTI in febrile children.

• 대한종양외과학회지
• /
• 제14권2호
• /
• pp.135-141
• /
• 2018

Purpose: Axillary lymph node dissection (ALND) and sentinel lymph node biopsy (SLNB) are important for staging of patients with node-positive breast cancer. However, these can be avoided in select micrometastatic diseases, preventing postoperative complications. The present study evaluated the ability of axillary lymph node maximum standardized uptake value (SUVmax) on positron emission tomography-computed tomography (PET-CT) to predict axillary metastasis of breast cancer. Methods: The records of invasive breast cancer patients who underwent pretreatment (surgery and/or chemotherapy) PET-CT between January 2006 and December 2014 were reviewed. ALNs were preoperatively evaluated by PET-CT. Lymph nodes were dissected by SLNB or ALND. SUVmax was measured in both the axillary lymph node and primary tumor. Student t-test and chi-square test were used to analyze sensitivity and specificity. Receiver operating characteristic (ROC) and area under the ROC curve (AUC) analyses were performed. Results: SUV-tumor (SUV-T) and SUV-lymph node (SUV-LN) were significantly higher in the triple-negative breast cancer (TNBC) group than in other groups (SUV-T: 5.99, P<0.01; SUV-LN: 1.29, P=0.014). The sensitivity (0.881) and accuracy (0.804) for initial ALN staging were higher in fine needle aspiration+PET-CT than in other methods. For PET-CT alone, the subtype with the highest sensitivity (0.870) and negative predictive value (0.917) was TNBC. The AUC for SUV-LN was greatest in TNBC (0.797). Conclusion: The characteristics of SUV-T and SUV-LN differed according to immunohistochemistry subtype. Compared to other subtypes, the true positivity of axillary metastasis on PET-CT was highest in TNBC. These findings could help tailor management for therapeutic and diagnostic purposes.

• Asian Spine Journal
• /
• 제12권6호
• /
• pp.1069-1077
• /
• 2018

Study Design: A case study. Purpose: To assess the chronological changes of the disease-related kyphosis after chemotherapy alone, secondly to clarify the role of growth cartilage in the healed lesion on kyphosis change, and to define the accurate prediction time in assessing residual kyphosis. Overview of Literature: None of the previous papers up to now dealt with the residual kyphosis, stability and remodeling processes of the affected segments. Methods: One hundred and one spinal tuberculosis children with various stages of disease processes, age 2 to 15 years, were the subject materials, between 1971 to 2010. They were treated with two different chemotherapy formula: before 1975, 18 months of triple chemotherapy (isoniazid [INH], para-aminosalicylic acid, streptomycin); and since 1976, 12 months triple chemotherapy (INH, rifampicin, ethambutol, or pyrazinamide). The first assessment at post-chemotherapy one year and at the final discharge time from the follow-up (36 months at minimum and 20 years at maximum) were analyzed by utilizing the images effect of the remaining growth plate cartilage on chronological changes of kyphosis after initiation of chemotherapy. Results: Complete disc destruction at the initial examination were observed in two (5.0%) out of 40 cervical spine, eight (26.7%) out of 30 dorsal spine, and six (19.4%) out of 31 lumbosacral spine. In all those cases residual kyphosis developed inevitably. In the remainders the discs were partially preserved or remained intact. Among 101 children kyphosis was maintained without change in 20 (19.8%), while kyphosis decreased in 14 children (13.7%), and increased in 67 children (66.3%) with non-recoverably damaged growth plate, respectively. Conclusions: It could tentatively be possible to predict the deformity progress or non-progress and spontaneous correction at the time of initial treatment, but it predictive accuracy was low. Therefore, assessment of the trend of kyphotic change is recommended at the end of chemotherapy. In children with progressive curve change, the deformity assessment should be continued till the maturity.

• 지역사회간호학회지
• /
• 제30권2호
• /
• pp.130-140
• /
• 2019

Purpose: The aim of this study is to explore levels of suicidal ideation and identify subgroups of high suicidal risk among the depressed elderly in Korea. Methods: A descriptive cross-sectional design was adopted on secondary data from the 6th (1st year) Korean national health and nutrition examination survey (KNHANES). A total of 239 depressed elders aged 60 or over who participated in the KNHANES. The prevalence of suicidal ideation and its related factors, including sociodemographic, physical, psychological characteristics and quality of life (EQ-5D index) were examined. Descriptive statistics and a decision tree analysis were performed using the SPSS/WIN 23.0 and SPSS Modeler 14.2 programs. Results: Of the depressed elderly, 28.9% had suicidal ideation. Three groups with high suicidal ideation were identified. Predictive factors included perceived stress level, household income level, quality of life and restriction of activity. In the highest risk group were those depressed elderly with moderate and low levels of stress, less than .71 of EQ-5D index and restriction of activity, and 80.0% of these participants had suicidal ideation. The accuracy of the model was 80.8%, its sensitivity 85.9%, and its specificity 68.1%. Conclusion: Multi-dimensional intervention should be designed to decrease suicide among the depressed elderly, particularly focusing on subgroups with high risk factors. This research is expected to contribute itself to the policy design and solution building in the future as it suggests policy implications in preventing the suicide of the depressed elderly.

• 인터넷정보학회논문지
• /
• 제22권5호
• /
• pp.79-86
• /
• 2021

빅데이터 분석의 유용성이 주목을 받으면서 경영학 분야에서도 이를 활용하여 기업의 성과를 예측하고자 하는 다양한 연구들이 진행되고 있다. 이러한 선행연구들은 주로 뉴스 기사나 SNS 등 기업 외부의 자료에 의존하고 있다. 직원의 만족도나 기업에 대한 직원의 인식, 장단점 평가와 같은 기업 내부의 목소리는 기업가치에 대한 잠재적인 영향력에도 불구하고 상대적으로 확보가 어려워 관련 연구가 아직 충분치 못하다. 이에 본 연구에서는 국내 유가증권시장 상장 기업을 대상으로 임직원의 기업리뷰가 기업가치에 미치는 영향을 살펴보고, 이를 기반으로 기업가치를 예측하는 모형을 구축하고자 한다. 이를 위해 온라인 기업리뷰 사이트인 잡플래닛(Jobplanet)에 2014년부터 2019년까지 전·현직원이 남긴 97,216건의 기업리뷰를 수집하고 동 데이터에 근거하여 머신러닝 기반의 예측 모형을 제안하였다. 제안한 모형 중 LSTM 기반 모형의 정확도가 73.2%로 가장 높았고 MAE 또한 0.359로 가장 낮은 오차를 보였다. 본 연구는 국내에서 머신러닝을 활용한 기업가치 연구 분야에 유용한 사례가 될 것으로 기대한다.

• 한국융합학회논문지
• /
• 제11권12호
• /
• pp.283-290
• /
• 2020

본 연구는 한국 성인을 대상으로 고요산혈증 위험을 예측하기 위한 중성지방-혈당 지수(triglyceride and glucose index, TyG index)의 유용성을 평가하였다. 서울지역 종합병원에서 2017년부터 2019년까지 건강진단을 실시한 20세 이상 남성 14,266명, 여성 9,033명을 대상으로 하였다. TyG 지수에 따른 고요산혈증 발생 위험도는 로지스틱 회귀분석을 실시하였으며, TyG 지수의 고요산혈증 위험 예측능력을 확인하기 위해 ROC 곡선을 구하였다. 고요산혈증을 예측하기 위한 TyG 지수의 정확도는 0.68이며, 남성 0.61, 여성 0.67이었다(각각 p<0.001). TyG 지수의 고요산혈증 발생 위험은 1사분위수보다 4사분위수에서 1.69배 높았으며, 남성은 2.03배, 여성은 2.07배 높았다(각각 p<0.05). 따라서 TyG 지수는 고요산혈증의 선별검사로서 진단적 유용성은 높지 않았으나, TyG 지수와 고요산혈증간에는 관련이 있었다.

• 대한건축학회논문집:구조계
• /
• 제35권11호
• /
• pp.155-162
• /
• 2019

The construction industry is the highest safety accident causing industry as 28.55% portion of all industries' accidents in Korea. In particular, falling is the highest accidents type composed of 60.16% among the construction field accidents. Therefore, we analyzed the factors of major disaster affecting the fall accident and then derived feature importances by considering various variables. We used data collected from Korea Occupational Safety & Health Agency (KOSHA) for learning and predicting in the proposed model. We have an effort to predict the degree of occupational fall accidents by using the machine learning model, i.e., Adaboost, short for Adaptive Boosting. Adaboost is a machine learning meta-algorithm which can be used in conjunction with many other types of learning algorithms to improve performance. Decision trees were combined with AdaBoost in this model to predict and classify the degree of occupational fall accidents. HyOperpt was also used to optimize hyperparameters and to combine k-fold cross validation by hierarchy. We extracted and analyzed feature importances and affecting fall disaster by permutation technique. In this study, we verified the degree of fall accidents with predictive accuracy. The machine learning model was also confirmed to be applicable to the safety accident analysis in construction site. In the future, if the safety accident data is accumulated automatically in the network system using IoT(Internet of things) technology in real time in the construction site, it will be possible to analyze the factors and types of accidents according to the site conditions from the real time data.

• Advances in concrete construction
• /
• 제10권6호
• /
• pp.479-488
• /
• 2020

The experimental and numerical works were carried out on high performance fiber reinforced concrete (HPFRC) with w/cm ratios ranging from 0.25 to 0.40, fiber volume fraction (Vf)=0-1.5% and 10% silica fume replacement. Improvements in compressive and flexural strengths obtained for HPFRC are moderate and significant, respectively, Empirical equations developed for the compressive strength and flexural strength of HPFRC as a function of fiber volume fraction. A relation between flexural strength and compressive strength of HPFRC with R=0.78 was developed. Due to the complex mix proportions and non-linear relationship between the mix proportions and properties, models with reliable predictive capabilities are not developed and also research on HPFRC was empirical. In this paper due to the inadequacy of present method, a back propagation-neural network (BP-NN) was employed to estimate the 28-day compressive strength of HPFRC mixes. BP-NN model was built to implement the highly non-linear relationship between the mix proportions and their properties. This paper describes the data sets collected, training of ANNs and comparison of the experimental results obtained for various mixtures. On statistical analyses of collected data, a multiple linear regression (MLR) model with R2=0.78 was developed for the prediction of compressive strength of HPFRC mixes, and average absolute error (AAE) obtained is 6.5%. On validation of the data sets by NNs, the error range was within 2% of the actual values. ANN model has given the significant degree of accuracy and reliability compared to the MLR model. ANN approach can be effectively used to estimate the 28-day compressive strength of fibrous concrete mixes and is practical.

• 한국수자원학회논문집
• /
• 제54권3호
• /
• pp.157-166
• /
• 2021

효율적인 댐 운영을 위해서는 높은 신뢰도를 기반으로 하는 유입량 예측이 요구된다. 본 연구에서는 최근 다양한 분야에서 사용되고 있는 데이터 기반의 예측 방법 중 하나인 딥러닝을 댐 유입량 예측에 활용하였다. 그 중 시계열 자료 예측에 높은 성능을 보이는 Sequence-to-Sequence 구조기반의 Long Short-Term Memory 딥러닝 모형(LSTM-s2s)을 이용하여 소양강 댐의 유입량을 예측하였다. 모형의 예측 성능을 평가하기 위해 상관계수, Nash-Sutcliffe 효율계수, 평균편차비율, 그리고 첨두값 오차를 이용하였다. 그 결과, LSTM-s2s 모형은 댐 유입량 예측에 대한 높은 정확도를 보였으며, 단일 유량 수문곡선 기반의 예측 성능에서도 높은 신뢰도를 보였다. 이를 통해 홍수기와 이수기에 수자원 관리를 위한 효율적인 댐 운영에 딥러닝 모형의 적용 가능성을 확인할 수 있었다.