• The Journal of the Convergence on Culture Technology
• /
• v.9 no.5
• /
• pp.415-419
• /
• 2023

The New Zealand Defence Force (NZDF) began using female manpower from World War II. After making various efforts to secure excellent manpower, the proportion of female manpower has risen to 24%, higher than that of Britain, the United States, Canada and Australia, which have a longer history of female military personnel than New Zealand. This is the result of NZDF efforts to open combat roles to women and allow female personnel to advance to high-ranking military positions such as generals and consular officers. In addition, policy alternatives to address women's realistic concerns such as pregnancy and childbirth, childcare, and vertical organizational culture were presented. In particular, Operation "Respect" was implemented to overcome the problem of not leaving or joining the army due to inappropriate sexual behavior and bullying. The operation respect established the role of the leader, emphasized the support of the victim, and accumulated data of the accident to prevent similar accidents. In addition, through the "Wāhine Toa" program, excellent female manpower could be introduced into the military through customized support considering the military life cycle (attract-recruit-retain-advance) of female personnel. South Korea is also considering expanding the ratio and role of female manpower as one of the ways to overcome the shortage of troops and leap into an advanced science and technology group. Implications were derived from the use of female manpower in the NZDF and the direction in which the Korean military should proceed was considered.

• 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.

• Ecology and Resilient Infrastructure
• /
• v.10 no.4
• /
• pp.107-115
• /
• 2023

High turbidity in source water can have adverse effects on water treatment plant operations and aquatic ecosystems, necessitating turbidity management. Consequently, research aimed at predicting river turbidity continues. This study developed a multi-class classification model for prediction of turbidity using LightGBM (Light Gradient Boosting Machine), a representative ensemble machine learning algorithm. The model utilized data that was classified into four classes ranging from 1 to 4 based on turbidity, from low to high. The number of input data points used for analysis varied among classes, with 945, 763, 95, and 25 data points for classes 1 to 4, respectively. The developed model exhibited precisions of 0.85, 0.71, 0.26, and 0.30, as well as recalls of 0.82, 0.76, 0.19, and 0.60 for classes 1 to 4, respectively. The model tended to perform less effectively in the minority classes due to the limited data available for these classes. To address data imbalance, the SMOTE (Synthetic Minority Over-sampling Technique) algorithm was applied, resulting in improved model performance. For classes 1 to 4, the Precision and Recall of the improved model were 0.88, 0.71, 0.26, 0.25 and 0.79, 0.76, 0.38, 0.60, respectively. This demonstrated that alleviating data imbalance led to a significant enhancement in Recall of the model. Furthermore, to analyze the impact of differences in input data composition addressing the input data imbalance, input data was constructed with various ratios for each class, and the model performances were compared. The results indicate that an appropriate composition ratio for model input data improves the performance of the machine learning model.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.29 no.7
• /
• pp.740-749
• /
• 2023

The ocean plays a vital role in the international carbon cycle, absorbing human-induced atmospheric carbon and preventing further atmospheric carbon accumulation. However, while the ocean had been considered a victim of climate change, it did not receive much attention as a solution for climate change in the major agenda of UNFCCC. Recently, a growing awareness that the ocean can provide numerous potentials to handle untapped issues to address the climate crisis has arisen, which has prompted discussions to strengthen ocean-based climate action. Since 2020, UNFCCC "Ocean and climate change dialogue" has been a forum to integrate and strengthen the ocean-climate nexus. This calls for integrating ocean action into climate action and the relevant sectors. In this regard, this study examined the background and international trends of ocean-based climate action and presented the author's perspective on the scope of content that such action should pursue and the direction to achieve it. In addition, this study identified tasks of the integrated approach and advancement of co-benefit as ways to strengthen ocean-based climate action, and it suggested domestic countermeasures for the Korean marine policy on climate change based on this.

• Informatization Policy
• /
• v.30 no.4
• /
• pp.40-61
• /
• 2023

The adoption of smart factories and smart manufacturing as strategies to enhance competitiveness and stimulate growth in the manufacturing sector is vital for a country's future competitiveness and industrial transformation. The government has consistently pursued smart manufacturing innovation policies starting with the Manufacturing Innovation 3.0 strategy in the Ministry of Industry. This study aims to identify policy areas for smart factories and smart manufacturing based on technical standards. Analyzing policy areas at the current stage where the establishment and support of domestic standards aligning with international technical standards are required is crucial. By prioritizing smart manufacturing process areas within the industry, policymakers can make well-informed decisions to advance smart manufacturing without blindly following international standardization in already well-established areas. To achieve this, the study utilizes a hierarchical analysis method including expert interviews and importance-performance analysis for the five major process areas. The findings underscore the importance of proactive participation in standardization for emerging technologies, such as data and security, instead of solely focusing on areas with extensive international standardization. Additionally, policymakers need to consider carbon emissions, energy costs, and global environmental challenges to address international trends in export and digital trade effectively.

• Journal of Environmental Impact Assessment
• /
• v.32 no.6
• /
• pp.503-515
• /
• 2023

The disconnection and fragmentation of ecological spaces that occur during the development process pose a significant threat to biodiversity. Urban center areas with high development pressure are particularly susceptible to low connectivity due to a scarcity of ecological space. This issue tends to be more pronounced in larger cities.To address this challenge, continuous efforts are needed to assess and improve the current state of ecological space connectivity at the level of individual projects and urban management. However, there is a lack of discussion regarding the analysis and improvement of ecological connectivity in metropolitan cities In line with this objective, this study evaluated the connectivity of ecological spaces in the city centers of Seoul, Busan, Daegu, Incheon, Gwangju, Daejeon, and Ulsan. The evaluation revealed that city centers exhibited lower connectivity of ecological spaces compared to their peripheries or the overall city. In addition, in the ecological network analysis that reflected regional characteristics, such as the species distribution model conducted on Daejeon, 510 optimal paths connecting forests of more than 1ha were derived. This study is significant as an example of deriving an ecological network based on regional characteristics, including quantitative figures necessary for establishing goals to improve urban ecological connectivity and biodiversity. It is anticipated that the results can be utilized to propose directions for enhancing ecological connectivity in environmental impact assessments or urban management and to establish an evaluation framework.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.5
• /
• pp.331-337
• /
• 2023

In this paper, we propose a deep learning model to detect lumbar 3 (L3) CT images to determine the occurrence and degree of sarcopenia. In addition, we would like to propose an optimization technique that uses oversampling ratio and class weight as design parameters to address the problem of performance degradation due to data imbalance between L3 level and non-L3 level portions of CT data. In order to train and test the model, a total of 150 whole-body CT images of 104 prostate cancer patients and 46 bladder cancer patients who visited Gangneung Asan Medical Center were used. The deep learning model used ResNet50, and the design parameters of the optimization technique were selected as six types of model hyperparameters, data augmentation ratio, and class weight. It was confirmed that the proposed optimization-based L3 level extraction model reduced the median L3 error by about 1.0 slices compared to the control model (a model that optimized only 5 types of hyperparameters). Through the results of this study, accurate L3 slice detection was possible, and additionally, we were able to present the possibility of effectively solving the data imbalance problem through oversampling through data augmentation and class weight adjustment.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_1
• /
• pp.577-587
• /
• 2023

The utilization of crowdsourced spatial data has been actively researched; however, issues stemming from the uncertainty of data quality have been raised. In particular, when low-quality data is mixed into drone imagery datasets, it can degrade the quality of spatial information output. In order to address these problems, the study presents a methodology for automatically validating the geometric quality of crowdsourced imagery. Key quality factors such as spatial resolution, resolution variation, matching point reprojection error, and bundle adjustment results are utilized. To classify imagery suitable for spatial information generation, training and validation datasets are constructed, and machine learning is conducted using a radial basis function (RBF)-based support vector machine (SVM) model. The trained SVM model achieved a classification accuracy of 99.1%. To evaluate the effectiveness of the quality validation model, imagery sets before and after applying the model to drone imagery not used in training and validation are compared by generating orthoimages. The results confirm that the application of the quality validation model reduces various distortions that can be included in orthoimages and enhances object identifiability. The proposed quality validation methodology is expected to increase the utility of crowdsourced data in spatial information generation by automatically selecting high-quality data from the multitude of crowdsourced data with varying qualities.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.11
• /
• pp.29-42
• /
• 2023

In this paper, we propose a method to enhance the prediction accuracy of solar irradiance for three major South Korean cities: Seoul, Busan, and Incheon. Our method entails the development of five generative models-vanilla GAN, CTGAN, Copula GAN, WGANGP, and TVAE-to generate independent variables that mimic the patterns of existing training data. To mitigate the bias in model training, we derive values for the dependent variables using random forests and deep neural networks, enriching the training datasets. These datasets are integrated with existing data to form comprehensive solar irradiance prediction models. The experimentation revealed that the augmented datasets led to significantly improved model performance compared to those trained solely on the original data. Specifically, CTGAN showed outstanding results due to its sophisticated mechanism for handling the intricacies of multivariate data relationships, ensuring that the generated data are diverse and closely aligned with the real-world variability of solar irradiance. The proposed method is expected to address the issue of data scarcity by augmenting the training data with high-quality synthetic data, thereby contributing to the operation of solar power systems for sustainable development.

• Journal of Animal Science and Technology
• /
• v.65 no.6
• /
• pp.1180-1193
• /
• 2023

Most carcass and meat quality traits are moderate to highly heritable, indicating that they can be improved through selection. Genetic evaluation for these types of traits is performed using performance data obtained from commercial and progeny testing evaluation. The performance data from commercial farms are available in large volume, however, some drawbacks have been observed. The drawback of the commercial data is mainly due to sorting of animals based on live weight prior to slaughter, and this could lead to bias in the genetic evaluation of later measured traits such as carcass traits. The current study has two components to address the drawback of the commercial data. The first component of the study aimed to estimate genetic parameters for carcass and meat quality traits in Korean Hanwoo cattle using a large sample size of industry-based carcass performance records (n = 469,002). The second component of the study aimed to describe the impact of sorting animals into different contemporary groups based on an early measured trait and then examine the effect on the genetic evaluation of subsequently measured traits. To demonstrate our objectives, we used real performance data to estimate genetic parameters and simulated data was used to assess the bias in genetic evaluation. The results of our first study showed that commercial data obtained from slaughterhouses is a potential source of carcass performance data and useful for genetic evaluation of carcass traits to improve beef cattle performance. However, we observed some harvesting effect which leads to bias in genetic evaluation of carcass traits. This is mainly due to the selection of animal based on their body weight before arrival to slaughterhouse. Overall, the non-random allocation of animals into a contemporary group leads to a biased estimated breeding value in genetic evaluation, the severity of which increases when the evaluation traits are highly correlated.