• 응용통계연구
• /
• 제21권4호
• /
• pp.639-647
• /
• 2008

본 연구는 국민건강보험공단의 건강검진데이터, 자격 및 보험료 그리고 진료비 데이터를 활용하여 고혈압 관리를 위한 맞춤형 고혈압 사후관리모형(고혈압 진료예측모형 및 고혈압 진료순응도세분화모형)을 개발하고자 하였다. 모형 개발에는 데이터마이닝의 로지스틱 회귀모형, 의사결정나무 그리고 앙상블 모형을 활용하였다. 고혈압 진료예측모형에서는 3가지 모형 중 로지스틱 회귀모형이 가장 우수한 모형으로 채택되었으며, 고혈압 진료순응도세분화모형은 의사결정나무모형을 통해 개발되었다. 본 연구는 전국 규모의 수년간 축적된 자료를 데이터마이닝을 활용함으로써 고혈압의 진료 및 진료순응도에 이르는 고혈압 사후관리 프로세스 전반에 걸친 결과를 도출함으로써 우리나라 고혈압 사후관리체계 구축에 기여할 것으로 사료된다.

• 천문학회지
• /
• 제52권6호
• /
• pp.217-225
• /
• 2019

We apply a modified Convolutional Neural Network (CNN) model in conjunction with transfer learning to predict whether an active region (AR) would produce a ≥C-class or ≥M-class flare within the next 24 hours. We collect line-of-sight magnetogram samples of ARs provided by the SHARP from May 2010 to September 2018, which is a new data product from the HMI onboard the SDO. Based on these AR samples, we adopt the approach of shuffle-and-split cross-validation (CV) to build a database that includes 10 separate data sets. Each of the 10 data sets is segregated by NOAA AR number into a training and a testing data set. After training, validating, and testing our model, we compare the results with previous studies using predictive performance metrics, with a focus on the true skill statistic (TSS). The main results from this study are summarized as follows. First, to the best of our knowledge, this is the first time that the CNN model with transfer learning is used in solar physics to make binary class predictions for both ≥C-class and ≥M-class flares, without manually engineered features extracted from the observational data. Second, our model achieves relatively high scores of TSS = 0.640±0.075 and TSS = 0.526±0.052 for ≥M-class prediction and ≥C-class prediction, respectively, which is comparable to that of previous models. Third, our model also obtains quite good scores in five other metrics for both ≥C-class and ≥M-class flare prediction. Our results demonstrate that our modified CNN model with transfer learning is an effective method for flare forecasting with reasonable prediction performance.

• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2006년도 춘계학술대회
• /
• pp.288-293
• /
• 2006

최근 반도체 개발 및 제조부문에서는 급속도로 성장한 Mobile 시장과 Digital Consumer Trend 에 주도적으로 대응하고자 빠른 속도의 기술적 변화를 추진하여 왔으며, 이에 따라 현장에서 발생하는 기술적 문제의 유형 또한 점점 복잡 다양해지고 있다. 기존 Commodity 제품의 경우, 세대 전환이 느리게 진행되어, 이에 수반되는 기술적 문제의 난이도와 그 발생 종류가 현재 대비 상대적으로 단순하였다. 그러나 최근 다품종 주문형 제품비중 확대로, 발생되는 문제점들이 Field Application 과 tight 한 연관성을 지니고 있어, 문제의 발생 경로, 원인 파악과 해결 대책 수립에 상당한 초기대응 시간을 필요하며, 관련 부서와의 실시간 협업 및 Cross-Checking을 통하지 않고서는 문제에 대한 최적 Solution 을 결정하기 힘든 현실이 되었다. 이러한 기술 이슈 해결 과정에 기 발생했던 유사 이슈 처리 과정 및 결과 자료가 문제 해결에 크게 기여할 수 있으나, 종래의 지식관리 시스템 체계로는 이러한 실무 형 지식을 획득하기 어려운 부분이 한계점으로 지적되고 있다. 본 논문에서는 이러한 문제점 개선이 가능한 실무형 지식 관리 (AKM: Actual Knowledge Management) 기반의 기술적 이슈 추적 관리 시스템(TITS: Technical Issue Tracking System) 구축 사례를 제시하고자 한다. TITS 는 1) 기술 이슈의 발의 및 전파 기능과 2) 이슈 해결을 위한 Action Item 부여 및 의견 교환 기능 3) 해결된 이슈의 처리 결과 등록 기능 4) 유사 이슈 사례 검색 기능 등의 크게 4 개 모듈오 구성 되어있다. 이 시스템을 통해 반도체 엔지니어들은 기술적 이슈의 발생 시점부터, 원안분석, 대책 협의 ; Action Item 수립 등 문제 해결 과정과 결과 등록 시점까지 Real time Tracking이 가능해 졌으며, 본 시스템을 통해 처리된 모든 기술적 이슈 정보는 Issue Case Database 에 분류 저장되어, 향후 유사 이슈 발생 시, 이를 활용함으로써, 빠른 초기 대응 및 문제 해결에 직접적인 도움을 받을 수 있게 되었다.

• Asian Pacific Journal of Cancer Prevention
• /
• 제16권14호
• /
• pp.6099-6104
• /
• 2015

Background: Previous studies investigating the association between miR-34b/c rs4938723 polymorphism and cancer risk showed inconclusive. Here, we performed meta-analysis to investigate the association between miR- 34b/c rs4938723 polymorphism and digestive cancer risk. Materials and Methods: Literature database including PubMed, OVID, Chinese National Knowledge Infrastructure (CNKI) were searched for publications concerning the association between the miR-34b/c rs4938723 polymorphism and digestive cancer risk. Results: A total of 6 studies consisting of 3246 cases and 3568 controls were included in this meta-analysis. The combined analysis suggested the miR-34b/c rs4938723 polymorphism significantly reduced digestive cancer risk under allelic model, homogeneous co-dominant model and recessive model (C vs T: OR=0.88, 95%CI=0.82-0.95, p-value=0.001; CC vs TT: OR =0.67, 95%CI=0.57-0.80, p-value=0.000; CC vs TT/TC: OR=0.68, 95%CI=0.58-0.80, p-value=0.000). Q-test and I2 test revealed no significant heterogeneity in all genotype comparisons. The Begger's funnel plot and Egger's test did not show significant publication bias. Conclusions: The current evidence supports the conclusion that the miR-34b/c rs4938723 polymorphism decreases an individual's susceptibility to digestive cancers.

• Asian Pacific Journal of Cancer Prevention
• /
• 제15권7호
• /
• pp.3267-3271
• /
• 2014

Background: Several studies have investigated the association between methionine synthase reductase (MTRR) A66G polymorphism and breast cancer risk, but controversial results were yielded. Therefore, we performed a meta-analysis to provide a more robust estimate of the effect of this polymorphism on susceptibility to breast cancer. Materials and Methods:Case-control studies investigating the relationship between MTRR A66G polymorphism and breast cancer risk were included by searching PubMed, EMBASE, China National Knowledge Infrastructure and Wanfang Database. Either fixed-effects or random-effects models were applied to calculate odds ratios(ORs) and 95% confidence intervals (CIs) by RevMan5.2 software. Results: A total of 9 studies bearing 7,097 cases and 7,710 controls were included in the meta-analysis. The results were that the combined ORs and 95%CIs of MTRR 66AG, GG, (AG+GG) genotypes were 0.98(0.91-1.05), 1.06(0.97-1.16) and 1.02(0.94-1.10), respectively with p=0.52, 0.19 and 0.65. We also performed subgroup analysis by specific ethnicity. The results of the combined analysis of MTRR 66AG, GG, (AG+GG) genotypes and breast cancer in Asian descent were Z=0.50, 0.53 and 0.21, with p all>0.05; for breast cancer in Caucasian descent, the results were Z=1.14, 1.65 and 0.43, with p all>0.05. Conclusions: Our findings suggested that MTRR A66G polymorphism was not associated with breast cancer susceptibility.

• Asian Pacific Journal of Cancer Prevention
• /
• 제14권8호
• /
• pp.4553-4557
• /
• 2013

Objective: The purpose of this study was to identify genes related to bladder cancer with samples from normal and disease cases by microarray chip. Methods: After downloading the gene expression profile GSE3167 from Gene Expression Omnibus database which includes 50 bladder samples, comprising 9 normal and 41 disease samples, differentially expressed genes were identified with packages in R language. The selected differentially expressed genes were further analyzed using bioinformatics methods. Firstly, molecular functions, biological processes and cell component analysis were researched by software Gestalt. Then, software String was used to search interaction relationships among differentially expressed genes, and hub genes of the network were selected. Finally, by using plugins of software Cytoscape, Mcode and Bingo, module analysis of hub-genes was performed. Results: A total of 221 genes were identified as differentially expressed by comparing normal and disease bladder samples, and a network as well as the hub gene C1QBP was obtained from the network. The C1QBP module had the closest relationship to production of molecular mediators involved in inflammatory responses. Conclusion: We obtained differentially expressed genes of bladder cancer by microarray, and both PRDX2 and YWHAZ in the module with hub gene C1QBP were most significantly related to production of molecular mediators involved in inflammatory responses. From knowledge of inflammatory responses and cancer, our results showed that, the hub gene and its module could induce inflammation in bladder cancer. These related genes are candidate bio-markers for bladder cancer diagnosis and might be helpful in designing novel therapies.

• Asian Pacific Journal of Cancer Prevention
• /
• 제13권5호
• /
• pp.2241-2247
• /
• 2012

Objective: This qualitative study retrospectively examined the experience and psychological impact of contralateral prophylactic mastectomy (CPM) among Southern Chinese females with unilateral breast cancer history who underwent BRCA1/2 genetic testing. Limited knowledge is available on this topic especially among Asians; therefore, the aim of this study was to acquire insight from Chinese females' subjective perspectives. Methods: A total of 12 semi-structured in-depth interviews, with 11 female BRCA1/BRCA 2 mutated gene carriers and 1 non-carrier with a history of one-sided breast cancer and genetic testing performed by the Hong Kong Hereditary Breast Cancer Family Registry, who subsequently underwent CPM, were assessed using thematic analysis and a Stage Conceptual Model. Breast cancer history, procedures conducted, cosmetic satisfaction, pain, body image and sexuality issues, and cancer risk perception were discussed. Retrieval of medical records using a prospective database was also performed. Results: All participants opted for prophylaxis due to their reservations concerning the efficacy of surveillance and worries of recurrent breast cancer risk. Most participants were satisfied with the overall results and their decision. One-fourth expressed different extents of regrets. Psychological relief and decreased breast cancer risk were stated as major benefits. Spouses' reactions and support were crucial for post-surgery sexual satisfaction and long-term adjustment. Conclusions: Our findings indicate that thorough education on cancer risk and realistic expectations of surgery outcomes are crucial for positive adjustment after CPM. Appropriate genetic counseling and pre-and post-surgery psychological counseling were necessary. This study adds valuable contextual insights into the experiences of living with breast cancer fear and the importance of involving spouses when counseling these patients.

• 한국컴퓨터산업학회논문지
• /
• 제2권9호
• /
• pp.1175-1182
• /
• 2001

원투원 마케팅(데이터베이스 마케팅 또는 관계 마케팅)은 컴퓨터의 발전과 더불어 기업 및 고객에게 이익을 가져올 것이며, 또한 고객의 세일 및 광고에 변화를 가져올 여러 분야 중의 하나이다. 인터넷 쇼핑몰에서 지능적인 고객 서비스의 일환으로, 본 논문에서는 데이터 마이닝 기법으로 잘 알려진 장바구니 분석을 이용한 개인화 된 광고를 제공하는 기법을 제시하고자 한다. 추천 기법의 핵심적인 이론으로 통계학, 데이터 마이닝, 인공 지능, 규칙 기반 매칭 등이 있다. 개인화 된 추천을 위한 규칙 기반 관점에서, 개인화를 위한 마케팅 규칙은 일반적으로 마케팅 전문가로부터 추출되어 고객의 데이터를 갖고 추정한다. 그러나 마케팅 전문가로부터 규칙을 추출하기란 매우 어려울 뿐만 아니라, 작성된 지식 기반 규칙을 검증하고 유지하기도 어렵다. 본 논문에서는 장바구니 분석 기법을 이용하여, 크로스 세일 마케팅 규칙을 추출한 뒤, 고객이 인터넷 쇼핑몰에 방문했을 때 개인화 된 광고를 제공하는데 초점을 두기로 한다.

• 대한한의학원전학회지
• /
• 제24권5호
• /
• pp.99-105
• /
• 2011

The medical treatment is process of collecting symptoms, making a diagnosis and treatment for the diagnosis. The remedy and the efficacy are the important hinge that links the diagnosis to the means of medical treatment. Lots of medical data are stored on database or saved as files in these days. The preferential way to find the efficacy related to the remedy is search all efficacy terms match up to the remedy term. To increase the rate of matching remedy terms with efficacy terms, we study the relation between efficacy terms and remedy terms and research the conversion way of the terms into the representative terms. The rate of matching the raw remedy terms with the raw efficacy terms is below 30%. The rate of matching the disaggregate remedy terms with the disaggregate efficacy terms is about 50%. The rate of matching the representative remedy terms with the representative efficacy terms is almost 70%.

• 한국도서관정보학회지
• /
• 제43권3호
• /
• pp.101-120
• /
• 2012

이 연구는 의류학 분야의 학문적 특성과 분류체계를 비교 분석함으로써, 의류학 분야의 분류특성과 문제점을 분석하고 이를 토대로 KDC 제5판 의류학 분야의 분류체계 수정 전개 방안을 제시하였다. 연구결과를 요약하면 아래와 같다. 첫째, 의류학 분야의 학문적 연구영역이 연구분류체계와 문헌분류체계에서는 대부분 민속학(복식), 화학공학(세탁, 염색공학), 제조업(피혁 및 모피공업, 직물 및 섬유공업, 의류제조), 생활과학(의복), 공예 및 장식미술(염직물공예) 등의 세목으로 분산 배치되어 있다. 둘째, 국립중앙도서관의 의류학 분야 KDC 유별 자료현황을 조사하여 특정 분류항목에 지나치게 자료가 집중되고 있는 KDC 문제점을 분석하고 이를 중심으로 개선방안을 제시하였다. 셋째, 분류항목의 수정 전개는 원칙적으로 의류학 분야의 학문체계에 따르되 기존의 KDC 분류체계를 가능한 한 그대로 유지하도록 하였으며, 항목간의 이동은 최소화하였다.