• 한국HCI학회:학술대회논문집
• /
• 2007.02a
• /
• pp.524-529
• /
• 2007

지능형 학습 시스템은 학습자의 학습 과정에서 수집된 데이터를 분석하여 학습자에게 맞는 전략을 세우고 적합한 서비스를 제공하는 시스템이다. 학습자에게 적합한 서비스를 위해서는 학습자 모델링 작업이 우선시 되며, 이 모델 생성을 위해서 학습자의 학습 과정에서 발생한 데이터를 수집하고 분석하게 된다. 하지만, 수집된 데이터가 학습자의 일관되지 못한 행위나 비예측 학습 성향을 포함하고 있다면, 생성된 모델을 신뢰하기 어렵다. 본 논문에서는 학습자에게서 수집된 데이터를 거리기반 이상치 선별 방법인 k-NN을 이용하여 이상치를 선별한다. 실험에서는 홈 인테리어 컨텐츠 기반에 학습자의 학습 행위에 대한 학습 성향을 진단하기 위한 DOLLS-HI를 이용하여, 수집된 학습자의 데이터에서 이상치를 분류하고 학습 성향 진단을 위한 모델을 생성하였다. 생성된 모델은 이상치 분류전과 비교하여 신뢰가 향상된 것을 확인하였다.

• The Korean Journal of Applied Statistics
• /
• v.34 no.3
• /
• pp.507-521
• /
• 2021

Detecting outliers among high-dimensional data encounters a challenging problem of screening the variables since relevant information is often contained in only a few of the variables. Otherwise, when a number of irrelevant variables are included in the data, the distances between all observations tend to become similar which leads to making the degree of outlierness of all observations alike. The subspace outlier detection method overcomes the problem by measuring the degree of outlierness of the observation based on the relevant subsets of the entire variables. In this paper, we survey recent subspace outlier detection techniques, classifying them into three major types according to the subspace selection method. And we summarize the techniques of each type based on how to select the relevant subspaces and how to measure the degree of outlierness. In addition, we introduce some computing tools for implementing the subspace outlier detection techniques and present results from the simulation study and real data analysis.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2002.02a
• /
• pp.276-281
• /
• 2002

본 연구에서는 정밀 쇄미 선별 시스템을 개발하기 위하여 원통형 홈 쇄미 선별기의 시작품을 제작하고 이것의 설계인자와 운전인자에 따른 성능 분석을 수행한 결과는 다음과 같다. 1. 원통형 홈 쇄미 선별기의 시작품은 홈의 직경이 작은 선별원통을 상단에, 홈의 직경이 큰 선별원통을 하단에 장착하는 2단형으로 설계 제작하였다. 각 단의 trough에는 스크류 컨베이어를 설치하였으며, 원통의 회전 속도와 trough의 각도 및 원통의 수평각을 변경시킬 수 있도록 하였다. 홈의 크기가 작은 상단에서 쇄미를, 홈의 크기가 큰 하단에서 준완전립을 선별하도록 하였다. 2. 원통형 홈 쇄미 선별기의 원통 회전속도가 증가하면 처리 용량이 증가하며, 이 속도에 따른 최적의 trough 각이 존재하는데, 본 실험에서는 회전 속도 35rpm에서 trough 각 37$^{\circ}$, 45rpm에서 55$^{\circ}$, 55rpm에서 73$^{\circ}$로 분석되었다. 3. 원통형 홈 쇄미 선별기의 공급율이 증가할수록 선별효율과 수거율(준완전립+쇄미)은 급격히 감소하지만, 순도(준완전립+쇄미)는 완만히 증가하였고, 완전립의 수거율과 순도는 95%이상을 유지하며 일정한 경향을 보였다. 본 실험 범위에서는 원통형 홈 쇄미선별기의 선별효율은 각 공급율에서 공히 원통의 분당 회전수 35rpm, trough 각 37$^{\circ}$에서 최대치를 나타내었다. 이 최적 조건에서 공급을 400-800kg/h 범위의 선별효율 평균치는 70% 정도로 분석되었다.

• Clinical and Experimental Pediatrics
• /
• v.50 no.11
• /
• pp.1061-1066
• /
• 2007

Purpose : Early identification of developmental disabilities allows intervention at the earliest possible point to improve the developmental potential. The Ages and Stages Questionnaire (ASQ), a parent- completed questionnaire, can be used as a substitute for formal screening tests. The purpose of this study was to evaluate the validity of the Korean version of the ASQ (K-ASQ) as a screening tool for detecting developmental delay of young Korean children in the setting of a busy pediatric outpatient clinic. Methods : Parents completed the K-ASQ in the waiting room of the pediatric outpatient clinic of St. Mary's Hospital, Catholic University Medical College. Out of 150 completed the ASQ, 67 who were born term and had no previous diagnosis of developmental delay, congenital anomalies, or neurological abnormalities were enrolled. The cut-off values of less than 2 standard deviations (SD) below the mean for the ASQ were used to define a "fail", and children who failed in one or more domains tested were classified as "screen-positive". Diagnosis of developmental delay was made when the developmental indices fell below -1 SD of the Bayley Scales of Infant Development-II. Results : (1) The mean age of children was $16.4{\pm}7.4$ months. Ten children (14.9%) were small-for- gestational age infants. The mean birth weight and gestational age were $3.1{\pm}0.6kg$ and $38.8{\pm}1.4$ weeks. Nine children (13.4%) were twins and 33 (49.0%) were male. The mean maternal education in years was $13.6{\pm}2.4$, and 31.3% had full-time jobs. The time for completing the ASQ was $10.2{\pm}3.0$ minutes. (2) Seventeen children (25.4%) were classified as screen-positive, four of them were delayed in development. Among eight children diagnosed with developmental delay, four were screen-positive and the other four were screen-negative by the ASQ. (3) The test characteristics of the ASQ were as follows: sensitivity (50.0%); specificity (78.0%); positive predictive value (23.5%); negative predictive value (92.0%). Conclusion : The high negative predictive value of the K-ASQ supports its use as a screening tool for developmental delay in the setting of a pediatric outpatient clinic.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.292-292
• /
• 2023

본 연구에서는 인공신경망을 이용한 통계적 모형을 구성하여 금강권역의 봄철(3~5월) 강수량 예측을 수행하였다. 통계적 모형의 예측인자로서는 NOAA 등에서 제공하는 AAO, AMM, AO 등 36종의 기후지수와 대상권역인 금강권역의 강수량, 기온 등의 기상인자 8종 등 총 44종의 기후지수를 활용하였다. 예측대상기간을 기준으로 선행기간(1~18개월)에 따른 상관성을 분석하여 상관도가 높은 10개의 기후지수를 예측인자로 선정하였다. 예측모형 형태는 10개의 입력층과 1개의 은닉층으로 되어 있는 인공신경망모형을 구성하였다. 모형 구성과정에서의 불확실성을 최소화하고 예측모형의 적합도를 높이기 위해 예측대상기간을 기준으로 과거 40년간의 자료에 대해 임의로 20년간 자료를 선별하여 모형을 구성하고, 너머지 기간에 대해 검증하는 무작위 교차검증을 반복하여, 예측대상기간 및 예측시점에 따라 각각 적합도가 높은 1000개의 예측모형을 선별하였다. 과거기간(1991~2022년)을 대상으로 예측시점에 따라 각 연도별 1000개의 예측결과를 도출하여, 실제 해당년도의 관측값과의 비교를 통해 예측성을 분석하였다. 예측성은 크게 예측치의 최대값과 최소값 범위 및 예측치의 25%~75% 범위 안에 관측치가 포함될 확률, 그리고 과거 관측값의 3분위 구간을 기준으로 한 예측확률 등을 평가하였다. 관측치가 예측치의 범위 안에 포함될 확률은 평균 87.5%, 예측치의 25~75% 범위 안에 포함될 확률은 30.2%로 나타났으며, 3분위 예측확률은 35.6%로 분석되었다. 관측값과의 일대일 비교는 정확도가 떨어지지만 3분위 예측확률이 33.3% 이상인 점으로 볼 때 예측성은 확보된다고 볼 수 있다. 다만, 우리나라 강수량의 불규칙성과 통계적 모형 특성상 과거 관측되지 않은 패턴에 대해서는 예측이 어려운 문제가 있어, 특정년도의 예측결과가 관측치를 크게 벗어나는 경우도 종종 나타나고 있다.

• Journal of Aquaculture
• /
• v.18 no.3
• /
• pp.154-159
• /
• 2005

This study was conducted to evaluate the effects of size grading on growth, feed efficiency and survival of juvenile olive flounder. Juvenile flounder were divided into four groups by initial average size; Small group $(1.3{\pm}0.23g)$, medium group $(3.1{\pm}0.45g)$, large group $(4.9{\pm}0.57g)$ and ungraded group $(3.3{\pm}1.66g)$. Triplicate groups of 100 fish were reared over 8 weeks. In final body weight distribution, frequency of the small size flounder (10 g) was markedly higher in the ungraded group than in the small group. Specific growth rate, feed efficiency and survival in the ungraded group were significantly lower (P<0.05) than those in the pooled data of the othor three graded groups, although feed intake in the ungraded group was significantly higher (P<0.05) than that of the pooled data of the other three graded groups. These results show that the small flounder gained significantly faster growth and higher survival in the absence of the large flounder. Therefore, size grading seems to be an important and necessary operation to improve the growth and survival of juvenile olive flounder (1-5 g).

• The Korean Journal of Applied Statistics
• /
• v.33 no.5
• /
• pp.527-540
• /
• 2020

While analyzing data, researching outliers, which are out of the main tendency, is as important as researching data that follow the general tendency. In this study we discuss the influence function for outlier discrimination. We derive sample influence functions of sample mean, sample variance, and sample standard deviation, which were not directly derived in previous research. The results enable us to mathematically examine the relationship between the empirical influence function and sample influence function. We can also consider a method to approximate the sample influence function by the empirical influence function. Also, the validity of the relationship between the approximated sample influence function and the empirical influence function is also verified by the simulation of random sampled data in normal distribution. As the result of a simulation, both the relationship between the two influence functions, sample and empirical, and the method of approximating the sample influence function through the emperical influence function were verified. This research has significance in proposing a method that reduces errors in the approximation of the empirical influence function and in proposing an effective and practical method that proceeds from previous research that approximates the sample influence function directly through empirical influence function by constant revision.

• Journal of the Korean Geotechnical Society
• /
• v.40 no.4
• /
• pp.91-103
• /
• 2024

In this study, we propose an outlier detection method that considers the spatial distribution of intelligent compaction measurement values (ICMVs) to address the high variability of ICMVs measured continuously across an entire construction area. The proposed method initially identified cases where the CMV at a specific location decreased despite an increase in the number of compaction passes. Among these, values that significantly differed from those measured within a 1.5-m radius were classified as outliers. Applying this method to CMV data obtained from field tests, we found that it effectively excluded the influence of changes in roller operating conditions unrelated to compaction quality while considering the inherent heterogeneity of the soil. However, after removing the outliers, the coefficient of variation of CMV (21.4%-26.3%) remained higher than the 20% suggested by relevant standards. Further field tests are needed to modify the proposed outlier detection method and to establish reasonable criteria for the variability of ICMV.

• Journal of Korean Society of Archives and Records Management
• /
• v.22 no.1
• /
• pp.43-59
• /
• 2022

Electronic public records are classified simultaneously as production, a preservation period is granted, and after a certain period, they are transferred to an archive and preserved. This study intends to find a way to improve the efficiency in classifying transferred records and maintain consistent standards. To this end, the current record classification work process carried out by the National Archives of Korea was analyzed, and problems were identified. As a way to minimize the manual work of record classification by converging the required improvement, the process of identifying outlier candidates based on a list consisting of classified information of the transferred records was proposed and systemized. Furthermore, the proposed outlier discrimination process was applied to the actual records transferred to the National Archives of Korea. The results were standardized and constructed as a training data format that can be used for machine learning in the future.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.45 no.1
• /
• pp.41-56
• /
• 2018

The purpose of this study was to provide clinical recommendations for restoration with selection of the most similar zirconia crown by 3-dimensional analysis of the shape of the maxillary primary central and lateral incisors in Korean individuals and prefabricated zirconia crowns. The average shape of the sound maxillary primary central and lateral incisors in 300 children was reproduced by 3-dimensional scanning. Zirconia crowns of 4 manufacturers (NuSmile $ZR^{(R)}$ Crown, Cheng $Crowns^{(R)}$, Kinder $Krowns^{(R)}$, and EZ $Pedo^{(R)}$ Crown) were scanned 3-dimensionally, and coordinates for comparison of the shape were measured to evaluate the similarity between the teeth and crowns. The most similar crowns were selected by comparing the mesiodistal length, crown height, crown shape ratio, distance between the same coordinates of a tooth and crown, the radius of curvature of the labial surface, and the volume. As a result of analysis, Cheng $Crowns^{(R)}$ size 3 and NuSmile $ZR^{(R)}$ Crown size 2 were the most similar crowns in the maxillary primary central and lateral incisors, respectively. Scanning the inner surface of the crowns and evaluating the amount of tooth reduction required suggested that an overall lesser amount of tooth reduction compared to that presented by the manufacturer's guidelines should be performed.