• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.482-482
• /
• 2022

하천의 합류부에서는 수질이 다른 유체가 혼합하여 합류 전과 다른 특성을 보인다. 하천의 합류부에서 수질을 효율적으로 관리하기 위해서는 수질의 공간적인 혼합 특성을 규명하는 것이 중요하다. 합류부에서 수질의 공간적인 혼합 특성을 분석하기 위해 본 연구에서는 토폴로지 데이터 분석(topological data analysis, TDA), 자기 조직화 지도(Self-Organizing Map, SOM), k-평균 알고리즘(K-means clustering algorithm) 세 가지 기법을 이용하였다. 세 가지 기법을 비교하여 어떤 알고리즘이 합류부의 수질 변화 특성을 더 뚜렷하게 나타내는지 분석하였다. 수질 변화 비교 인자들은 pH, chlorophyll, DO, Turbidity 등이 있고, 수질 인자들은 YSI를 활용해 측정하였다. 자료의 측정 지역은 낙동강과 황강이 합류하는 지역이며, 보트에 YSI 장비를 부착하고 횡단하여 측정하였다. 측정한 데이터를 R 프로그램을 통해 세 가지 기법을 적용시켜 수질 변화 비교를 분석한다. 토폴로지 데이터 분석(topological data analysis, TDA)은 거대하고 복잡한 데이터로부터 유의미한 정보를 추출하는 데 사용하고, 자기조직화지도(Self-Organizing Map, SOM) 기법은 차원 축소와 군집화를 동시에 수행한다. k-평균 알고리즘(K-means clustering algorithm) 기법은 주어진 데이터를 k개의 클러스터로 묶는 머신러닝 비지도학습에 속하는 알고리즘이다. 세 가지 방법들의 주목적은 클러스터링이다. 클러스터 분석(Cluster analysis)이란 주어진 데이터들의 특성을 고려해 동일한 성격을 가진 여러 개의 그룹으로 대상을 분류하는 데이터 마이닝의 한 방법이다. 군집화 방법들인 TDA, SOM, K-means를 이용해 합류 지역의 수질 특성들을 클러스터링하여 수질 패턴들을 분석해 하천 수질 오염을 방지할 수 있을 것이다. 본 연구에서는 토폴로지 데이터 분석(topological data analysis, TDA), 자기조직화지도(Self-Organizing Map, SOM), k-평균 알고리즘(K-means clustering algorithm) 세 가지 기법을 이용하여 합류부에서의 수질 특성을 비교하며 어떤 기법이 합류의 특성을 더욱 뚜렷하게 나타내는지 규명했다. 합류의 특성을 군집화 방법을 이용해 알게 된다면, 합류부의 수질 변화 패턴을 다른 합류 지역에서도 적용할 수 있을 것으로 기대된다.

• Korean Journal of Environmental Agriculture
• /
• v.26 no.1
• /
• pp.17-24
• /
• 2007

This study was carried out to classify the streams at rural watersheds by characteristics of water quality. The water quality data of 319 steams at rural watersheds in Korea were selected. Multivariate analysis was used for this purpose. The cases were divided into 5 types, and then factor analysis and cluster analysis were done. The characteristics of water quality of rural watersheds can be showed more than 40% of total water quality by first factor(organic matters and nutrients). The cluster analysis of extracted factors using factor analysis was carried out. The results showed that the Case 1 and Case 2 were classified 4 communities, Case 3 was classified 5 communities and Case 4 and 5 were classified 3 communities. Among 5 types cases, to classified the steams of rural watersheds, Case 4 - 7 water quality items - was selected as a desirable case. Many kinds of statistical analysis can be used to classify the streams of rural watersheds. Our results showed a good example to evaluate the water quality properties in Korean rural watershed.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.162-162
• /
• 2017

최근 우리나라는 전 지구적인 기후변화로 인하여 집중호우 및 돌발 홍수와 같은 극치 사상들이 증가하고 있는 추세이며, 이에 대한 분석을 위해 극치 분포를 이용한 수문통계적 특성에 대한 접근이 주로 이루어지고 있다. 이를 위해서는 충분한 수의 자료가 필요하나 우리나라 강우자료는 지점별로 자료 보유 년 수가 비교적 많지 않기 때문에, 이러한 문제를 극복하기 위하여 하나의 지역, 즉 주어진 지점을 포함하여 수문학적으로 동일한 조건을 만족하는 주변 지점의 자료를 모두 포함하여 빈도해석을 실시하는 지역빈도해석이 필요하다. 따라서 본 연구에서는 지역빈도해석과 두 개의 형상매개변수를 포함하여 다양한 극치 수문통계특성을 나타낼 수 있다고 알려진 Burr XII 분포를 이용하여 우리나라 강우자료에 대한 그 적용성을 살펴보았다. 이를 위해 군집분석을 통한 강우지점의 지역화 과정을 거치고 분류된 지역을 L-moment ratio diagram에 도시하여, Burr XII 분포 영역 내 포함여부를 통해 Burr XII 분포의 적합도를 도시적으로 살펴보고, Hosking and Wallis (1997)이 제안한 적합성 척도($^{IST}$)를 통한 적합성 여부를 판별하였다. 또한 우리나라 강우자료에 비교적 적합하다고 알려진 분포인 generalized extreme value, generalized logistic, Gumbel 분포와의 비교를 위해, 전체 지역에 대하여 재현기간에 따른 상대편의 (relative bias)와 상대평균제곱근오차 (relative root mean square error)를 산정하여 Burr XII 분포형의 적용 가능성을 살펴보았다.

• Tunnel and Underground Space
• /
• v.25 no.4
• /
• pp.303-319
• /
• 2015

Text mining is one of the branches of data mining and is used to find any meaningful information from the large amount of text. In this study, we analyzed titles and keywords of two SCI journals on rock engineering by using text mining to find major research area, trend and associations of research fields. Visualization of the results was also included for the intuitive understanding of the results. Two journals showed similar research fields but different patterns in the associations among research fields. IJRMMS showed simple network, that is one big group based on the keyword 'rock' with a few small groups. On the other hand, RMRE showed a complex network among various medium groups. Trend analysis by clustering and linear regression of keyword - year frequency matrix provided that most of the keywords increased in number as time goes by except a few descending keywords.

• Journal of the Korean GEO-environmental Society
• /
• v.9 no.7
• /
• pp.73-85
• /
• 2008

320 national groundwater monitoring stations have been constructed since 1995 and groundwater levels are measured automatically 4 times a day at each well. It has a difficulty to estimate an average recharge rate of watershed using the recharge rate of the monitoring site because of the lack of its representative on converting a point recharge rate into a spatial one. In this study, the relations between site characteristics (topography, hydraulics, geology, facilities, etc.) and recharge rates of 223 monitoring sites, which were selected using cluster analysis, were analyzed using statistical methods, and finally, regression models were constructed for a recharge rate estimation of non-measured areas. The independent variables for these simple regression models, 1) width of adjacent stream, 2) distance to the nearest stream, 3) topographic slope, and 4) rock type, are proposed using analysis of variance. These models have lots of advantages such as an easy data collection from topographic and geologic maps, a few input variables, and also simplicity in use. Suitability analysis from the comparison between estimation values and original ones at monitoring sites shows that these models are useful for a groundwater recharge estimation.

• The Korean Journal of Applied Statistics
• /
• v.33 no.6
• /
• pp.833-842
• /
• 2020

Global platforms such as Netflix, Amazon, and YouTube have developed a precise recommendation system based on various information from large set of customers and many of the items recommended here are leading to actual purchases. In this paper, a cluster analysis was conducted according to the attribute of the content, expecting that there would be a difference in user preferences according to the attribute of the recommended content. Gower distance was used for use regardless of the type of variables. In this paper, using the data of movie rating site 'The Movie Dataset', the users were grouped hierarchically and recommended movies based on genre, director and actor variables. To evaluate the recommended systems proposed, user group was divided into train set and test set to examine the precision. The results showed that proposed algorithms have far higher precision than UBCF.

• Journal of Korean Society of Transportation
• /
• v.30 no.1
• /
• pp.19-30
• /
• 2012

The majority of daily travel demands concentrate at particular time-periods, which causes the difficulties in the travel demand analysis and the corresponding benefit estimation. Thus, it is necessary to consider time-specific traffic characteristics to yield more reliable results. Traditionally, na$\ddot{i}$ve, heuristic, and statistical approaches have been applied to address the peak-hour ratio. In this study, a hybrid clustering model which is one of the statistical methods is applied to calculate the peak-hour ratio and its duration. The 2009 national 24-hour traffic data provided by the Korea institute of Construction Technology are used. The analysis is conducted dividing vehicle types into passenger cars and trucks. For the verification for the usefulness of the methodology, the toll collection system data by the Korea Express Corporation are collected. The result of the research shows lower errors during the off-peak hours and night times and increasing error ratios as the travel distance increases. Since the method proposed can reduce the arbitrariness of analysts and can accommodate the statistical significance test, the model could be considered as a more robust and stable methodology. It is hoped that the result of this paper could contribute to the enhancement of the reliability for the travel demand analysis.

• Journal of the Korean earth science society
• /
• v.41 no.5
• /
• pp.447-458
• /
• 2020

This study presents the K-means clustering analysis-based classification of the meteorological patterns affecting the occurrence of high PM₁₀ concentration in the southeastern region of the Korean peninsula for the last five years (2014-2018). Regional differences in Busan, Ulsan, and Gyeongnam related to high PM₁₀ episodes, were clarified through the statistical comparison study using synoptic scale meteorological elements using NCEP (National Centers for Environmental Prediction/FNL (Final Operational Global Analysis) re-analysis meteorological data. Meteorological patterns were classified into a total of five categories (C1-C5). The incidence of each cluster was 24.8% (C1), 21.3% (C2), 20.4% (C3), 17.3% (C4), and 16.2% (C5), respectively. The high PM₁₀ concentration in the southeastern region resulted from long and short range transports (C1, C3, C5) from outside of the region, and the emissions (C2, C4) inside the region. In the high PM₁₀ episodes in Busan, Ulsan, and Gyeongnam regions, meteorological characteristics such as different geopotential height and wind speed at 500 hPa in each cluster and the change in the location of high pressure over Korean Peninsula is strongly associated with the dispersion of PM₁₀ around inventories in the region and the tendency of long-range transportation of PM₁₀ emitted from outside of region.

• Korean journal of applied entomology
• /
• v.48 no.4
• /
• pp.439-445
• /
• 2009

Species richness (number of species) of the butterfly community in the Gwangneung forest, Korea, was estimated using data of the long-term butterfly monitoring, which had been carried out 291 times in the Korea National Arboretum or forest area of Gwangneung from 1998 to 2008. Abundance of each butterfly species was monitored using the line-transact method. In the present study, 13,333 butterflies belonging to 112 species were observed. Species accumulation curve and species richness was obtained using a software, EstimateS. The species accumulation curve shows an increase tendency even at 291 survey times, implying the possibility of the presence of a few unfound species. However, values of species richness estimated by the seven estimators were stabilized around 240-250 survey times. Species richness estimated by the estimators ranged from 120 species to 141 species with 128 species in average. However, the figure estimated by the previous studies since 1958 was 148 species. We estimated the reasonable scale of species richness on the base of recent analysis on the change of butterfly species. Species richness of the Korea National Arboretum was higher than that of natural forest and of plantation. However, species richness of butterfly was not different between natural forest and plantation. It is likely that increase of grasslands and habitat diversity in arboretum led to the increase of species richness of butterfly community.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2008.05a
• /
• pp.309-313
• /
• 2008

본 논문의 목적은 ENSO의 영향에 의한 우리나라 강우의 특성을 분석하는 것이다. 따라서 우리나라 기상관측소의 강우량 자료를 Warm(El Nino), Cold(La Nina), Normal 에피소드에 따라 기간별로 분류하였다. 또한 이렇게 분류한 자료는 Markov Chain 모형을 이용하여 100년의 자료로 모의 발생하였고 에피소드별로 빈도분석을 실시하였다. 빈도분석 결과 에피소드에 따라 각 기상관측소별로 강우의 크기에 영향을 미치고 있음을 알 수 있었다. 또한 군집분석을 실시하여 각 에피소드의 공간적인 영향에 대해서 분석하였다. 결과적으로 Warm(El Nino), Cold(La Nina) and Normal 에피소드로 대표되는 ENSO는 우리나라의 강우특성에 크게 영향을 미치는 것으로 파악되었다.