• Journal of Korean Society of Forest Science
• /
• v.78 no.2
• /
• pp.209-217
• /
• 1989

Seedlings of 6 tree species were treated with artificial acid rain and acid mist (pH 5.0, 4.0, 3.0) and ground water (pH 6.5), to select $SO_2$-resistant tree species. The growth variable, leaf injury rate and chlorophyll content were measured and compared among the various pH levels. Seedling height of Rosy multiflora decreased with deceasing pH levels of artificial acid rain and was tallest at control plot, but that of Ailanthus altissima was tallest at pH 5.0 plot. For the seedlings of Robinia pseudoacacia, Magnolia obovata and Wistaria floribunda, top and root dry weights per seedling at pH 5.0 plot were higher than those at control plot. Leaf injury rate(injured leaf area and injured leaf rate) increased with decreasing pH levels of artificial acid rain, the changes of leaf chlorophyll content was slightly different among tree species. Leaf chlorophyll content of Rosa multiflora, measured during the period July to September, decreased with decreasing pH levels of artificial acid rain. Leaf chlorophyll content of Magnolia obovata increased on July, but decreased severely on September, with decreasing pH levels, of artificial acid rain.

• Ecology and Resilient Infrastructure
• /
• v.6 no.2
• /
• pp.79-88
• /
• 2019

In order to study the water quality status and its causal environmental factors, the water quality variables of chemical oxygen demand (COD), chlorophyll a (Chl a), Total phosphorus (TP), and total nitrogen (TN), the hydrogeomorphologic variables of water level fluctuation, total water storage, dam elevation, watershed area, and shoreline development index, and the land cover variables of forest, agricultural area, and urbanized area in the watershed were investigated in total 73 reservoirs with various operational purposes, water level fluctuation and geographical distribution in South Korea. The water quality was more eutrophic in the reservoirs of the more urbanized and agricultural area in the watershed, the low altitude, the narrow water level fluctuation, the narrowed watershed area, and the more circular shape. In terms of the purposes of reservoir operation, the reservoirs for agricultural irrigation were more eutrophic than the reservoirs for flood control. The results of the variable selection and path analysis showed that COD determined by Chl a and TP was directly affected by water level fluctuation and the shoreline development of the reservoirs. TP was directly affected by the urbanized area of the watershed which was related to the elevation of the reservoir. TP was also influenced by the water level fluctuation and the shoreline development. In conclusion, the eutrophication of the reservoirs in Korea would be influenced by the land use of the watershed, hydrological and geographical characteristics of the reservoir, water level fluctuation by the anthropogenic management according to the reservoir operation purpose, and the location of the reservoirs.

• The Journal of Bigdata
• /
• v.3 no.2
• /
• pp.71-78
• /
• 2018

The study is based on machine learning techniques to increase the accuracy of the forest fire predictive model. It used 14 years of data from 2003 to 2016 in Gang-won-do where forest fire were the most frequent. To reduce weather data errors, Gang-won-do was divided into nine areas and weather data from each region was used. However, dividing the forest fire forecast model into nine zones would make a large difference between the date of occurrence and the date of not occurring. Imbalance issues can degrade model performance. To address this, several sampling methods were applied. To increase the accuracy of the model, five indices in the Canadian Frost Fire Weather Index (FWI) were used as derived variable. The modeling method used statistical methods for logistic regression and machine learning methods for random forest and xgboost. The selection criteria for each zone's final model were set in consideration of accuracy, sensitivity and specificity, and the prediction of the nine zones resulted in 80 of the 104 fires that occurred, and 7426 of the 9758 non-fires. Overall accuracy was 76.1%.

• The Journal of the Korea Contents Association
• /
• v.21 no.2
• /
• pp.667-675
• /
• 2021

This study conducted a systematic review of evidence-based interventions to confirm the importance of gait in dementia patients and to inform the necessity of various interventions necessary for gait. Based on PRISMA's guidelines and evidence-based intervention, a systematic review were conducted, and papers published in domestic journals for the past 10years were collected, and the dependent variables measured along with the intervention type and gait were analyzed. For data search, research papers from January 2011 to June 2020 were collected through RISS, KISS, the National Library of Korea, and the Library of Congress. The main search terms were 'dementia patient', 'walking', and 'walking ability'. Searched 57 papers on dementia patients and gait that meet the literature selection criteria. Among them, papers overlapping with papers before 2010, papers whose dependent variable is not related to gait ability were excluded, and finally, other than dementia diseases. As for the type of gait intervention, there were many programs related to exercise such as fall prevention and physical activity, and the dependent variables measured along with gait were physically and psychologically diverse. Although domestic studies for dementia patients are conducted in a variety of directions and methods, there are few studies on the improvement of function and gait of the lower extremity part of the body. Therefore, it is necessary to study the multifaceted and various intervention methods for walking in dementia patients.

• The Korean Journal of Applied Statistics
• /
• v.34 no.3
• /
• pp.309-327
• /
• 2021

Dynamic treatment regimes (DTRs) are decision-making rules designed to provide personalized treatment to individuals in multi-stage randomized trials. Unlike classical methods, in which all individuals are prescribed the same type of treatment, DTRs prescribe patient-tailored treatments which take into account individual characteristics that may change over time. The Q-learning method, one of regression-based algorithms to figure out optimal treatment rules, becomes more popular as it can be easily implemented. However, the performance of the Q-learning algorithm heavily relies on the correct specification of the Q-function for response, especially in observational studies. In this article, we examine a number of double-robust weighted least-squares estimating methods for Q-learning in high-dimensional settings, where treatment models for propensity score and penalization for sparse estimation are also investigated. We further consider flexible ensemble machine learning methods for the treatment model to achieve double-robustness, so that optimal decision rule can be correctly estimated as long as at least one of the outcome model or treatment model is correct. Extensive simulation studies show that the proposed methods work well with practical sample sizes. The practical utility of the proposed methods is proven with real data example.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.24 no.2
• /
• pp.153-170
• /
• 2022

Waterjet has been comprehensively used in urban areas owing to a suitable technique for cutting concrete and rock, and low noise and vibration. Recently, the abrasive waterjet technique has been adopted and applied by the Korea Atomic Energy Research Institute to demolish concrete plugging without disturbing and damaging In-situ Demonstration of Engineered Barrier System in the disposal research tunnel. In this study, the use of abrasive waterjet in the tunnel was evaluated for practical applicability and the existing cutting model was compared with the experimental results. As a variable for waterjet cutting, multi-cutting, water flow rate, abrasive flow rate, and standoff distance were selected for the diversity of analysis. As regarding the practical application, the waterjet facilitated path selection for cutting the concrete plugging and prevented additional disturbances in the periphery. The pump's noise at idling was 64.9 dB which is satisfied with the noise regulatory standard, but it exceeded the standard at ejection to air and target concrete because the experiment was performed in the tunnel space. The experimental result showed that the error between the predicted and measured cutting volume was 12~13% for the first cut and 16% for second cut. The standoff distance had a significant influence on the cutting depth and width, and the error tended to decrease with decrement of standoff distance.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.27 no.2
• /
• pp.71-86
• /
• 2022

The water quality index (WQI) has been widely used to evaluate marine water quality. The WQI in Korea is categorized into five classes by marine environmental standards. But, the WQI calculation on huge datasets is a very complex and time-consuming process. In this regard, the current study proposed machine learning (ML) based models to predict WQI class by using water quality datasets. Sihwa Lake, one of specially-managed coastal zone, was selected as a modeling site. In this study, adaptive boosting (AdaBoost) and tree-based pipeline optimization (TPOT) algorithms were used to train models and each model performance was evaluated by metrics (accuracy, precision, F1, and Log loss) on classification. Before training, the feature importance and sensitivity analysis were conducted to find out the best input combination for each algorithm. The results proved that the bottom dissolved oxygen (DO_Bot) was the most important variable affecting model performance. Conversely, surface dissolved inorganic nitrogen (DIN_Sur) and dissolved inorganic phosphorus (DIP_Sur) had weaker effects on the prediction of WQI class. In addition, the performance varied over features including stations, seasons, and WQI classes by comparing spatio-temporal and class sensitivities of each best model. In conclusion, the modeling results showed that the TPOT algorithm has better performance rather than the AdaBoost algorithm without considering feature selection. Moreover, the WQI class for unknown water quality datasets could be surely predicted using the TPOT model trained with satisfactory training datasets.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.24 no.4
• /
• pp.234-243
• /
• 2022

In Korea, hail damage occurs every year, and in the case of agriculture, it causes severe field crop and cultivation facility losses. Therefore, it is necessary to develop a hail information service system customized for Korea's primary production and crop-growing areas to minimize hail damage. However, the observation of hail is relatively more difficult than that of other meteorological variables, and the available data are also spatially and temporally variable. A hail information service system was developed to understand the temporal and spatial distribution of hail occurrence. As part of this, a hail observation database was established that integrated the observation data from Korea Meteorological Administration with the information from newspaper reports. Furthermore, a hail risk map was produced based on this database. The risk map presented the nationwide distribution and characteristics of hail showers from 1970 to 2018, and the northeastern region of South Korea was found to be relatively dangerous. Overall, hail occurred nationwide, especially in the northeast and some inland areas (Gangwon, Gyeongbuk, and Chungbuk province) and in winter, mainly on the north coast and some inland areas as graupel (small and soft hail). Analyzing the time of day, frequency, and hailstone size of hail shower occurrences by region revealed that the incidence of large hail stones (e.g., 10 cm at Damyang-gun) has increased in recent years and that showers occurred mainly in the afternoon when the updraft was well formed. By integrating multidisciplinary data, the temporal and spatial gap in hail data could be supplemented. The hail risk map produced in this study will be helpful for the selection of suitable crops and growth management strategies under the changing climate conditions.

• Economic and Environmental Geology
• /
• v.55 no.4
• /
• pp.353-366
• /
• 2022

Spatial estimation of geoscience data (geo-data) is challenging due to spatial heterogeneity, data scarcity, and high dimensionality. A novel spatial estimation method is needed to consider the characteristics of geo-data. In this study, we proposed the application of Gaussian Mixture Model (GMM) among machine learning algorithms with multivariate data for robust spatial predictions. The performance of the proposed approach was tested through soil chemical concentration data from a former smelting area. The concentrations of As and Pb determined by ex-situ ICP-AES were the primary variables to be interpolated, while the other metal concentrations by ICP-AES and all data determined by in-situ portable X-ray fluorescence (PXRF) were used as auxiliary variables in GMM and ordinary cokriging (OCK). Among the multidimensional auxiliary variables, important variables were selected using a variable selection method based on the random forest. The results of GMM with important multivariate auxiliary data decreased the root mean-squared error (RMSE) down to 0.11 for As and 0.33 for Pb and increased the correlations (r) up to 0.31 for As and 0.46 for Pb compared to those from ordinary kriging and OCK using univariate or bivariate data. The use of GMM improved the performance of spatial interpretation of anthropogenic metals in soil. The multivariate spatial approach can be applied to understand complex and heterogeneous geological and geochemical features.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.30 no.4
• /
• pp.449-454
• /
• 1985

In order to search for the semi-dwarf japonica varieties allelic to the semi-dwarf rice cultivar which is controlled by d-t gene, seven dwarf japonica varieties. Reimei, Hoyoku. Shiranui, Kokumasari, M 7. S.224 and S.295 were crossed to the semi-dwarf cultivar, wx 817. wx 817 is known to have semi-dwarf gene d-t. Their F$_1$, F$_2$ and F$_3$ were grown in 1984 and 1985 and culm lengths were measured at harvest. The results are summarized as follows. 1. The F$_2$s of all 7 cross combinations showed normal distribution and no segregation. 2. The range of culm length variation in the F$_3$ was variable depending on the cross combination, but the general pattern was similar in the all 7 crosses. 3. The mean of F$_3$ and parental F$_2$ mean which were selected into short, medium and tall groups were similar and showed no segregation, implying the selection efficiency in F$_2$. 4. From the results of F$_2$ and F$_3$ segregations, it is concluded that the culm length of the 7 semi-dwarf japonicas tested here are controlled by the same major gene d-t although they are modified by different minor genes.