• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.07a
• /
• pp.393-394
• /
• 2020

세 군 매칭을 수행하여 관찰 데이터를 구축하고 통계분석에 기반한 연구를 수행하는 경우가 종종 발생한다. 매칭작업은 각 군에 속한 개체의 성향점수를 서로 비교해 거리가 가까운 짝을 찾아야 하므로 카테시안 곱 만큼의 경우의 수를 따져야 하는 문제이고, 메모리 소요가 크다. 특히 세 군 매칭은 세 쌍의 거리가 가까운 triplet을 찾는 문제로, 세 개체 사이에 존재하는 세 개의 거리를 따져야 하기 때문에 메모리 소요가 두 군 매칭에 비해 훨씬 크다. 각 군에 속한 개체가 늘어나면 메모리소요가 기하 급수적으로 늘어나게 된다. R패키지에 포함된 TriMatch함수는 세 군 매칭 수행을 위해 가장 널리 사용되는 프로그램이다. 이 프로그램은 세 개체 사이의 세 개 거리가 가장 짧은 triplet을 찾는 방식으로 구현 되었다. 이 프로그램은 메모리 소요가 매우 커 각 군에 속한 개체의 수가 많아지면 메모리 부족 에러가 발생하는 경우가 많다. 본 연구에서는 세 군 매칭에 소요되는 메모리 소요를 줄일 수 있는 알고리즘을 제안하고자 한다. 이 알고리즘의 구현을 통해 각 군에 속한 개체가 늘어나도 안정적인 세 군 매칭 결과를 얻을 수 있을 것으로 기대한다.

• Journal of Labour Economics
• /
• v.32 no.3
• /
• pp.61-89
• /
• 2009

This paper undertakes an international study on impacts of private education spending on student academic performance, using the OECD's PISA 2006 data. To circumvent endogeneity of the education spending, it relies on a propensity-score matching method. The empirical results suggest thai the effect of private education spending remains only modest for all the countries examined in the PISA 2006 data. A 10 percent increase in spending leads to no more than a 0.6 percent average improvement in test scores.

• Journal of Computing Science and Engineering
• /
• v.11 no.1
• /
• pp.24-31
• /
• 2017

Face-based video retrieval has become an active and important branch of intelligent video analysis. Face profiling and matching is a fundamental step and is crucial to the effectiveness of video retrieval. Although many algorithms have been developed for processing static face images, their effectiveness in face-based video retrieval is still unknown, simply because videos have different resolutions, faces vary in scale, and different lighting conditions and angles are used. In this paper, we combined content-based and semantic-based image analysis techniques, and systematically evaluated four mainstream local features to represent face images in the video retrieval task: Harris operators, SIFT and SURF descriptors, and eigenfaces. Results of ten independent runs of 10-fold cross-validation on datasets consisting of TED (Technology Entertainment Design) talk videos showed the effectiveness of our approach, where the SIFT descriptors achieved an average F-score of 0.725 in video retrieval and thus were the most effective, while the SURF descriptors were computed in 0.3 seconds per image on average and were the most efficient in most cases.

• Progress in Medical Physics
• /
• v.23 no.1
• /
• pp.8-14
• /
• 2012

In order to develop a Patient respiratory management system includinga biofeedback function for4-dimentional radiation therapy, this study investigated anoptimal tracking algorithmfor moving target using IR (Infra-red) camera as well as commercial camera. A tracking system was developed by LabVIEW 2010. Motion phantom images were acquired using a camera (IR or commercial). After image process were conducted to convert acquired image to binary image by applying a threshold values, several edge enhance methods such as Sobel, Prewitt, Differentiation, Sigma, Gradient, Roberts, were applied. The targetpattern was defined in the images, and acquired image from a moving targetwas tracked by matching pre-defined tracking pattern. During the matching of imagee, thecoordinateof tracking point was recorded. In order to assess the performance of tracking algorithm, the value of score which represents theaccuracy of pattern matching was defined. To compare the algorithm objectively, we repeat experiments 3 times for 5 minuts for each algorithm. Average valueand standard deviations (SD) of score were automatically calculatedsaved as ASCII format. Score of threshold only was 706, and standard deviation was 84. The value of average and SD for other algorithms which combined edge detection method and thresholdwere 794, 64 in Sobel, 770, 101 in Differentiation, 754, 85 in Gradient, 763, 75 in Prewitt, 777, 93 in Roberts, and 822, 62 in Sigma, respectively. According to score analysis, the most efficient tracking algorithm is the Sigma method. Therefore, 4-dimentional radiation threapy is expected tobemore efficient if threshold and Sigma edge detection method are used together in target tracking.

• Korean Journal of Social Welfare
• /
• v.63 no.4
• /
• pp.301-326
• /
• 2011

The major purpose of this study is to evaluate the impact of the long-term care insurance program. In order to estimate the impact of policy accurately, certain bias which might hamper the validity of this study has been removed by Propensity Score Matching(PSM) and Double Difference(DD) method from the semi-experimental design. To study the effects of long-term care insurance on the elderly and their family members as social outcome variable sand the quality of life of their family and satisfaction in family relationships, the third and fourth waves of Korea Welfare Panel are used to match experimental and comparative groups by the propensity score matching. Then, DD method, using the panel fixed effects model, is applied to estimate the differences of those groups'treatment effects before and after the policy implementation. As a result, it was found that the Quality of life on the elderly and their family members is statistically meaningless, while the satisfaction in family relationships has much increased after the policy implementation. The result has a limitation in that this evaluation is performed at the point when the long-term insurance program has not been ripened enough. However, there is an important implication on the significance of realizing the main goal of the long-term care insurance to improve the quality of life of the elderly and their family members and as for the potentiality of further system improvements.

• The Journal of Information Systems
• /
• v.25 no.1
• /
• pp.49-73
• /
• 2016

Purpose Though many information systems researchers have made various attempts to investigate the relationship between information technology capability and firm performance from diverse perspectives, we have not come to a conclusion yet with some mixed results. In this research, focusing on the adoption of Enterprise Resource Planning (ERP) systems by firms as a proxy measure for information technology capability, we reexamine whether the association is significantly positive. Design/methodology/approach Previous research on this topic had some limitations to the samples and analysis method. Some research focused only on the 1990s or early 2000s, and other studies failed to adequately compare the impact of ERP adoption on firm performance between the treatment group and the control group. In this research, extending previous analysis approaches with the matched sample comparison of IT leaders and the control group, we attempt to apply propensity score matching in combination with difference-in-difference analysis with a sample of Korean firms that adopted ERP systems in the late 2000s. We match ERP adopters and non-adopters with propensity score matching and compare their financial performance with difference-in-difference estimation between the pre- and post-adoption periods. Findings According to our analysis, we find no positive and significant relationship between ERP adoption and firm performance in profit ratios. This research shows that, contrary to the era of proprietary information systems, standardized information systems today have no additional competitive advantages over competitors.

• 한국사회정책
• /
• v.19 no.3
• /
• pp.85-117
• /
• 2012

This study attempts to examine the relationship between giving & volunteering behavior and charitable bequest intention. For the examination, this study used '2011 Korean National Social Survey' that was randomly sampled from the population of Korean in 2011. In examining the relationship, this study employed the method of Propensity Score Matching that permits the comparisons between experimental group and control group. In this study, the experimental groups consist of six different combinations of philanthropic behaviors as follows: donating only; volunteering only; participating both; regular donating only; regular volunteering only; and doing both regularly. The results show that all the types of philanthropic behaviors have statistically significant positive effect on charitable bequest intention. First, more active philanthropic behavior such as regular behavior causes higher charitable bequest intention. Second, those who participate in both philanthropic behaviors (combined behavior of donating and volunteering) have stronger effect on charitable bequest intention in comparison to participating only one philanthropic behavior (either donating or volunteering). Third, giving have relatively stronger relationship with charitable bequest intention than volunteering. Throughout more careful examination of the causal relationship from philanthropic behavior to charitable bequest intention, this study provides new understanding on the effect of daily philanthropic behavior on charitable bequest and practical implication to nurture charitable bequest. Indeed, this study contributes to building a knowledge foundation for future research on charitable bequest.

• Journal of the Korean Regional Science Association
• /
• v.38 no.2
• /
• pp.21-41
• /
• 2022

The purpose of this study is to compare the differences in the subjective quality of life (QoL) perceived by residents in different settlement spaces such as urban vs. rural and Seoul Capital Area (SCA) vs. Non-Capital Area (NCA). Using the data of the 'Quality of Life Satisfaction Survey for Balanced Development' conducted nationwide by the Presidential Committee for Balanced National Development in 2018, a propensity score matching method was introduced to compare the differences between the types of settlement spaces. The results show that no significant difference is found between all the types in satisfaction with local life, and in the case of happiness, rural residents show a higher level than those living in cities. Looking at the quality of life by QoL domain, the QoL of urban residents in SCA is generally the highest in the basic living environment, medical care, and culture/leisure domains, while the QoL of rural residents in the job/income, environment, community, and welfare domains is higher than that of residents of SCA and NCA cities. Interestingly, in no domains NCA cities show their relative strength. These results show the value of rural areas in the face of increasingly prevailing urbanization trends, and suggests that the value of rural areas must be strengthened for rural development. It is necessary to pursue policy efforts such as rural regeneration based on spatial planning in order to respond to the disorganized development that damages the value of the rural environment and landscape and to harmoniously maintain and develop the rural settlement.

• The Korean Journal of Applied Statistics
• /
• v.36 no.5
• /
• pp.381-397
• /
• 2023

The most representative design used in clinical trials is randomization, which is used to accurately estimate the treatment effect. However, comparison between the treatment group and the control group in an observational study without randomization is biased due to various unadjusted differences, such as characteristics between patients. Propensity score weighting is a widely used method to address these problems and to minimize bias by adjusting those confounding and assess treatment effects. Inverse probability weighting, the most popular method, assigns weights that are proportional to the inverse of the conditional probability of receiving a specific treatment assignment, given observed covariates. However, this method is often suffered by extreme propensity scores, resulting in biased estimates and excessive variance. Several alternative methods including trimming, overlap weights, and matching weights have been proposed to mitigate these issues. In this paper, we conduct a simulation study to compare performance of various propensity score weighting methods under diverse situation, such as limited overlap, misspecified propensity score, and treatment contrary to prediction. From the simulation results overlap weights and matching weights consistently outperform inverse probability weighting and trimming in terms of bias, root mean squared error and coverage probability.

• Korean Journal of Radiology
• /
• v.23 no.6
• /
• pp.615-624
• /
• 2022

Objective: To compare the therapeutic outcomes of laparoscopic hepatic resection (LHR) and laparoscopic radiofrequency ablation (LRFA) for single subcapsular hepatocellular carcinoma (HCC). Materials and Methods: We screened 244 consecutive patients who had received either LHR or LRFA between January 2014 and December 2016. The feasibility of LRFA in patients who underwent LHR was retrospectively assessed by two interventional radiologists. Finally, 60 LRFA-feasible patients who had received LHR and 29 patients who had received LRFA as the first treatment for a solitary subcapsular HCC between 1 cm and 3 cm were finally included. We compared the therapeutic outcomes, including local tumor progression (LTP), recurrence-free survival (RFS), and overall survival (OS) between the two groups before and after propensity score (PS) matching. Multivariable Cox proportional hazard regression was also used to evaluate the difference in OS and RFS between the two groups for all 89 patients. Results: PS matching yielded 23 patients in each group. The cumulative LTP and OS rates were not significantly different between the LHR and LRFA groups after PS matching (p = 0.900 and 0.003, respectively). The 5-year LTP rates were 4.6% and 4.4%, respectively, and OS rates were 100% and 90.7%, respectively. The RFS rate was higher in LHR group without statistical significance (p = 0.070), with 5-year rates of 78.3% and 45.3%, respectively. OS was not significantly different between the LHR (reference) and LRFA groups in multivariable analyses, with a hazard ratio (HR) of 1.33 (95% confidence interval, 0.12-1.54) (p = 0.818). RFS was higher in LHR (reference) than in LRFA without statistical significance in multivariable analysis, with an HR of 2.01 (0.87-4.66) (p = 0.102). Conclusion: There was no significant difference in therapeutic outcomes between LHR and LRFA for single subcapsular HCCs measuring 1-3 cm. The difference in RFS should be further evaluated in a larger study.