• 한국지리정보학회지
• /
• 제22권4호
• /
• pp.72-85
• /
• 2019

본 연구는 일상생활 가운데 누구나 겪을 수 있는 성폭력 범죄 위험의 예측을 목적으로 한다. 2011-2015년 5년 간 청주시 일부 지역에서 발생한 성폭력 범죄를 대상으로 베이지안 통계 기반의 Weight of Evidence를 적용·분석하였다. Weight of Evidence를 활용하여 분석한 결과 첫째, 투입된 관련요인(Evidence) 총 26개 중 주거용도면적, 건축물 사용승인일, 개별주택가격, 용적률, 지하층 수, 대지면적, 보안등, 오락시설 8개만이 신뢰도를 만족하여 각각의 가중치(Weight)가 산출되었다. 둘째, 가중치가 산출된 8개의 관련요인을 결합하여 최종적으로 예측 지도를 도출하였다. 성폭력 범죄가 발생할 위험이 75.5%인 지역은 대상지역 전체면적의 20.7%(2.0㎢)를 차지하였으며 16.5% 지역은 3.3%(0.3㎢), 34.5% 지역은 19.0%(1.8㎢)로 나타났다. 본 연구는 성폭력 범죄 위험의 발생 확률과 이를 감소시킬 수 있는 환경적 요인 또는 조건을 도출하였다. 이와 같은 결과는 경찰의 범죄예방 활동 등 성폭력 범죄 피해 최소화를 위한 선제적 대응방안 마련의 기초자료로서 활용될 수 있을 것이다.

• 방송공학회논문지
• /
• 제25권3호
• /
• pp.346-352
• /
• 2020

표준화 마무리 단계인 차세대 비디오 부호화 표준 VVC(Versatile Video Coding)는 HEVC(High Efficiency Video Coding) 보다 두배 이상의 압축 성능을 달성하기 위해 다양한 기술들을 채택하고 있다. VVC는 GPB(Generalized P and B) 슬라이스에서의 양예측(bi-prediction) 기법의 성능 향상을 위하여 두 예측신호에 다양한 가중치를 적용하여 최종 예측신호를 생성하는 BCW(Bi-prediction with CU-level Weight)를 채택하였다. BCW 가중치는 가용한 참조픽처 중 재생 순서상 미래 픽처의 존재 유무를 지시하는 신택스(syntax) 요소인 NoBackwardPredFlag에 따라 적응적으로 시그널링 된다. 이러한 신택스 구조는 비디오 코덱의 유연성을 저해하고 비트스트림 파싱(parsing) 단계에서의 의존성 문제를 야기한다. 본 논문에서는 이러한 문제를 해결하기 위하여 가중치 전송에서 기존의 조건 판단 없이 가능한 모든 가중치를 허용하는 기법을 제안하였으며, 부호화기에서의 다양한 가중치 탐색을 통해 제안방법의 성능을 확인하였다. 제안된 기법은 제기된 문제를 해결하면서도 실험결과 3개의 가중치 탐색을 수행하였을 때 무시할 정도의 BD-rate 손실과 5개의 가중치 탐색을 수행하였을 때 다소의 부호화 성능 향상을 확인하였다.

• 한국데이터정보과학회:학술대회논문집
• /
• 한국데이터정보과학회 2006년도 추계 학술발표회 논문집
• /
• pp.119-128
• /
• 2006

The similarity weight, the pearson's correlation coefficient, which is used in the recommender system has a weak point that it cannot predict all of the prediction value. The similarity weight, the vector similarity, has a weak point of the high MAE although the prediction coverage using the vector similarity is higher than that using the pearson's correlation coefficient. The purpose of this study is to suggest how to raise the prediction coverage. Also, the MAE using the suggested method in this study was compared both with the MAE using the pearson's correlation coefficient and with the MAE using the vector similarity, so was the prediction coverage. As a result, it was found that the low of the MAE in the case of using the suggested method was higher than that using the pearson's correlation coefficient. However, it was also shown that it was lower than that using the vector similarity In terms of the prediction coverage, when the suggested method was compared with two similarity weights as I mentioned above, it was found that its prediction coverage was higher than that pearson's correlation coefficient as well as vector similarity.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2007년도 춘계학술대회논문집
• /
• pp.51-55
• /
• 2007

In this paper Statistical Energy Analysis has been considered to predict high frequency air borne interior noise. Dash panel Insulation is major part to reduce engine excitation noise. Transmission loss and absorption coefficient are considered to predict dash insulation performance. Transmission lose is derived from coupling loss factor and absorption coefficient is derived from internal damping loss factor. Material Biot properties were used to calculate each loss factors. Insulation geometry thickness distribution was hard to measure, so FeGate software was used to calculate thickness map from CAD drawing. Each predicted transmission losses between conventional insulation and light weight insulation were compared with SEA. Transmission loss measurement was performed to validate each prediction result, and it showed good correlation between prediction and measurement. Finally interior noise prediction was performed and result showed light weight insulation system can reduce 40% weight to keep similar performance with conventional insulation system, even though light weigh insulation system has lower sound transmission loss and higher absorption coefficient than conventional system.

• 한방비만학회지
• /
• 제17권2호
• /
• pp.68-76
• /
• 2017

Objectives: The purpose of this study was to analyze the factors affecting weight loss effect of Gamitaeumjowee-tang and to see if weight loss could be predicted using influence factors and weight loss progress. Methods: From September 2016 to March 2017, we retrospectively reviewed the medical records of 139 patients who were visited to the Korean Oriental Clinic for 3 months. We conducted a regression analysis to determine whether age, gender, initial weight, patient health questionnaire-9, heart rate variability (HRV), sleep quality, drinking habit and the medication history of weight loss affect weight loss. We found weight loss prediction equations using multiple regression analysis applying significant factors and weight loss progress. Results: Gender and initial weight had a significant effect on weight loss in all periods (P<0.001). HRV had a significant effect on primary weight loss (P<0.01). Other factors did not have any significant effect on weight loss. Using the significant factors, weight loss of each period could be predicted from 23.9% to 44.6%, and tertiary weight loss could be predicted with 76.6% using factors, primary weight loss and secondary weight loss (P<0.001). Conclusions: This study suggests that weight loss effect of Gamitaeumjowee-tang maybe be affected by influence factors and that weight loss prediction equations using them can be used for obesity treatment.

• 경영정보학연구
• /
• 제9권1호
• /
• pp.85-103
• /
• 2007

본 연구의 목적은 좀 더 정확한 고객 선호도 예측을 위한 협업 필터링 알고리즘의 예측 성능을 평가하기 위한 것이다. 고객 선호도 예측의 정확도를 비교하기 위하여 이웃 기반의 협업 필터링 알고리즘과 대응평균 알고리즘에 의한 고객 선호도 예측의 MAE를 비교하였다. 예측 알고리즘의 정확성을 분석하기 위하여 MovieLens 1 Million dataset을 이용하여 실험을 하였다. 각 예측 알고리즘에 사용된 유사도 가중치는 일반적으로 이용되는 피어슨 상관계수와 벡터 유사도를 이용하였으며 분석결과 대응평균 알고리즘의 예측 정확도가 이웃 기반의 협업 필터링 알고리즘의 예측 정확도 보다 우수한 것으로 나타났다. 두 알고리즘에 사용된 유사도 가중치인 피어슨 상관계수와 벡터 유사도는 두 고객이 특정 상품에 대하여 공통으로 평가한 선호도 평가치를 이용하여 계산된다. 이때 공통으로 평가한 선호도 평가치의 개수가 적으면 계산된 유사도 가중치가 과대 평가된다. 과대 평가된 유사도 가중치를 보정하여 고객 선호도 예측의 정확도를 높이기 위하여 기존의 연구에서 고려한 공통 평가 영화의 개수 보다 확대된 범위를 적용하였으며 각 예측 방법에 따라 서로 다른 개선 경향을 파악할 수 있었다.

• 콘크리트학회논문집
• /
• 제27권2호
• /
• pp.169-176
• /
• 2015

현재까지 발표된 크기효과법칙은 보통중량 콘크리트에 기반하고 있어 파괴특성이 다른 경량골재 콘크리트에서는 그 활용성이 의문시되고 있다. 따라서 이 연구에서는 콘크리트의 기건단위질량이 압축강도의 크기효과에 미치는 영향을 예측할 수 있는 모델을 개발하고 기존 연구결과들을 모아 데이터베이스화하였다. 그리고 비선형 파괴역학에 근거한 Ba${\check{z}}$ant와 Kim and Eo의 예측모델 및 이 연구에서 제안한 식에 대한 실험상수들을 결정한 후, 상호 비교 분석하였다. 그 결과, 콘크리트의 기건단위질량을 고려한 본 연구의 예측모델이 Ba${\check{z}}$ant와 Kim and Eo의 예측모델보다 경량골재 콘크리트에 대한 실험결과를 더 잘 예측하고 있음을 알 수 있었다.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2005년도 추계학술대회논문집
• /
• pp.381-384
• /
• 2005

The purpose of this study is to develop a prediction model for evaluating heavy-weight floor impact sounds in a test building. Three rooms in the test building (slab thickness In and 240mm), which consist of frame concrete structures were tested and modeled. First, the SPL distribution in the receiving room was analyzed by measuring SPL at 90 positions using a bang machine. Then, a vibration model using finite element method is proposed considering the material properties and boundary conditions. In addition, the result of transient analysis was compared with field measurements using a standard heavy-weight impact source. Through a vibro-acoustic simulation program, an acoustic model evaluating the building elements (reflected wall, nor, window and door) was proposed. Finally, validation of the prediction model was conducted by vibro-acoustic analysis with field measurements of noise radiation characteristics in receiving rooms.

• Asian-Australasian Journal of Animal Sciences
• /
• 제29권9호
• /
• pp.1215-1221
• /
• 2016

The prediction of carcass composition in Hanwoo steers is very important for value-based marketing, and the improvement of prediction accuracy and precision can be achieved through the analyses of independent variables using a prediction equation with a sufficient dataset. The present study was conducted to develop a prediction equation for Hanwoo carcass composition for which data was collected from 7,907 Hanwoo steers raised at a private farm in Gangwon Province, South Korea, and slaughtered in the period between January 2009 and September 2014. Carcass traits such as carcass weight (CWT), back fat thickness (BFT), eye-muscle area (EMA), and marbling score (MAR) were used as independent variables for the development of a prediction equation for carcass composition, such as retail cut weight and percentage (RC, and %RC, respectively), trimmed fat weight and percentage (FAT, and %FAT, respectively), and separated bone weight and percentage (BONE, and %BONE), and its feasibility for practical use was evaluated using the estimated retail yield percentage (ELP) currently used in Korea. The equations were functions of all the variables, and the significance was estimated via stepwise regression analyses. Further, the model equations were verified by means of the residual standard deviation and the coefficient of determination ($R^2$) between the predicted and observed values. As the results of stepwise analyses, CWT was the most important single variable in the equation for RC and FAT, and BFT was the most important variable for the equation of %RC and %FAT. The precision and accuracy of three variable equation consisting CWT, BFT, and EMA were very similar to those of four variable equation that included all for independent variables (CWT, BFT, EMA, and MAR) in RC and FAT, while the three variable equations provided a more accurate prediction for %RC. Consequently, the three-variable equation might be more appropriate for practical use than the four-variable equation based on its easy and cost-effective measurement. However, a relatively high average difference for the ELP in absolute value implies a revision of the official equation may be required, although the current official equation for predicting RC with three variables is still valid.

• Asian-Australasian Journal of Animal Sciences
• /
• 제31권10호
• /
• pp.1670-1676
• /
• 2018

Objective: This study was conducted to enable on-line prediction of primal and commercial cut weights in Korean slaughter pigs by AutoFom III, which non-invasively scans pig carcasses early after slaughter using ultrasonic sensors. Methods: A total of 162 Landrace, Yorkshire, and Duroc (LYD) pigs and 154 LYD pigs representing the yearly Korean slaughter distribution were included in the calibration and validation dataset, respectively. Partial least squares (PLS) models were developed for prediction of the weight of deboned shoulder blade, shoulder picnic, belly, loin, and ham. In addition, AutoFom III's ability to predict the weight of the commercial cuts of spare rib, jowl, false lean, back rib, diaphragm, and tenderloin was investigated. Each cut was manually prepared by local butchers and then recorded. Results: The cross-validated prediction accuracy ($R^2cv$) of the calibration models for deboned shoulder blade, shoulder picnic, loin, belly, and ham ranged from 0.77 to 0.86. The $R^2cv$ for tenderloin, spare rib, diaphragm, false lean, jowl, and back rib ranged from 0.34 to 0.62. Because the $R^2cv$ of the latter commercial cuts were less than 0.65, AutoFom III was less accurate for the prediction of those cuts. The root mean squares error of cross validation calibration (RMSECV) model was comparable to the root mean squares error of prediction (RMSEP), although the RMSECV was numerically higher than RMSEP for the deboned shoulder blade and belly. Conclusion: AutoFom III predicts the weight of deboned shoulder blade, shoulder picnic, loin, belly, and ham with high accuracy, and is a suitable process analytical tool for sorting pork primals in Korea. However, AutoFom III's prediction of smaller commercial Korean cuts is less accurate, which may be attributed to the lack of anatomical reference points and the lack of a good correlation between the scanned area of the carcass and those traits.