• Journal of the Korean Geosynthetics Society
• /
• v.14 no.2
• /
• pp.89-94
• /
• 2015

In this study, the mixed soil with an optimum mixed ratio was suggested in order to recycle the powdered sludge of basalt in Jeju Island as the impermeable liner materials. As the results of soil laboratory tests, the grain size of the powdered sludge of basalt is less than 0.1mm and the powdered sludge was classified into ML or CL category in accordance with the Unified Soil Classification System (USCS). Also, the grain size of natural soils is ranged from 0.1 mm to 10 mm and the soils were classified into SW category in USCS. To select the optimum mixed ratio of powdered sludge, the variable permeability test was performed to various mixed soils with different powdered sludge amount under both optimum compaction and field conditions. As the results of permeability tests, the coefficient of permeability of mixed soils was decreased with increasing the mixed ratio of powdered sludge, and the mixed soil with mixed ratio of 60% has the minimum coefficient of permeability. Therefore, the optimum mixed ratio of powdered sludge is 60% for recycling the powdered sludge of basalt as the impermeable liner materials.

• The Korean Journal of Applied Statistics
• /
• v.21 no.5
• /
• pp.825-833
• /
• 2008

Discriminant analysis based on Gaussian mixture models, an useful tool for multi-class classifications, can be extended to semi-supervised learning. We consider a model selection problem for a Gaussian mixture model in semi-supervised learning. More specifically, we adopt Bayesian information criterion to determine the number of subclasses in the mixture model. Through simulations, we illustrate the usefulness of the criterion.

• Journal of Internet Computing and Services
• /
• v.14 no.5
• /
• pp.49-57
• /
• 2013

A novel approach for the feature selection is proposed, which is the important preprocessing task of on-line document classification. In previous researches, the features based on information from their single population for feature selection task have been selected. In this paper, a mixed feature set is constructed by selecting features from multi-population as well as single population based on various information. The mixed feature set consists of two feature sets: the original feature set that is made up of words on documents and the transformed feature set that is made up of features generated by LSA. The hybrid feature selection method using both filter and wrapper method is used to obtain optimal features set from the mixed feature set. We performed classification experiments using the obtained optimal feature sets. As a result of the experiments, our expectation that our approach makes better performance of classification is verified, which is over 90% accuracy. In particular, it is confirmed that our approach has over 90% recall and precision that have a low deviation between categories.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2012.05a
• /
• pp.744-744
• /
• 2012

불투수층은 자연적인 침투를 허용하지 않는 인위적인 토지피복상태로 도시화율 추정 및 유역의 환경변화 정도를 분석하기 위한 척도로 사용되어 왔다. 특히, 수문학적 관점에서 불투수층은 단기 유출현상에 큰 영향을 끼치는 요소로 불투수율이 증가할수록 침투량이 감소하여 첨두유출량은 증가하고 도달시간은 짧아진다. 최근에는 급속한 도시화로 인해 불투수층의 영향이 더욱 커짐에 따라 불투수율의 추정에 대한 필요성이 증가하고 있다. 현재까지 위성영상을 이용한 불투수층의 추정은 고해상도 영상을 이용하여 피복분류를 수행하였다. 즉, 분류된 토지피복에 근거하여 불투수율을 산술적으로 계산하거나 분광혼합기법 및 회귀 트리기법 등 다양한 방법에 적용되어 왔다. 본 연구에서는 Sub-pixel 분류기법을 위성영상에 적용하여 섬강유역의 불투수율을 추정하고자 한다. Sub-pixel 분류는 기존 분류기법들이 다양한 토지피복이 혼합된 화소에 대해서도 가장 비중이 높은 토지피복 하나로 분류하던 것을 개선한 방법으로 fuzzy 이론을 적용하여 최소 20% 이상의 비율을 점유하는 항목 모두를 구분하여 분류하는 기법이다. 이를 위해 섬강유역의 Landsat TM 영상을 수집하고 환경부의 토지피복도와 지질도를 참조하여 트레이닝 자료를 수집하였다. 또한 결과에 영향을 미칠 수 있는 구름은 전처리를 통하여 제거하고 수집된 트레이닝 자료에 Sub-pixel 분류기법을 적용하여 섬강유역의 불투수율을 공간분포도로 작성하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10c
• /
• pp.865-867
• /
• 2001

데이터 종속성을 제거하기 위해서 명령어의 결과값을 예상하는 여러 결과값 예측기의 장점을 이용하여 놓은 성능을 얻을 수 있는 새로운 혼합형 예측 메커니즘을 제안한다. 제안된 혼합형 결과값 예측기는 예상 테이블을 모험적으로 갱신할 수 있기 때문에 부적절한(Stale) 데이터로 인해 잘못 예상되는 명령어의 수를 효과적으로 감소시킨다. 또한 정적 분류 정보를 사용하여 명령의 반입시 적절한 예측기에 할당함으로써 예상 정확도를 더욱 향상시키며, 하드웨어 비용을 효율적으로 감소시키도록 하였다. 5개의 SPECint 95 벤치마크 프로그램에 대해 SimpleScalar/PISA 3.0 툴셋을 사용하여 실험하였다. 16-이슈 폭에서 모험적 갱신을 사용한 평균 예상 정확도는 73%의 실험 결과가 나왔으며, 정적 분류 정보를 사용하였을 경우 예상 정확도가 88%로 증가된 결과를 얻었다.

• Journal of Korean Society for Geospatial Information Science
• /
• v.3 no.2 s.6
• /
• pp.91-100
• /
• 1995

The aim of this study is to classify mixed information in a pixel of a remote sensing image data (in the case of SPOT HRV's band $1{\sim}3,\;20m{\times}20m$). First, the loss of information and the uncertainty of mixed pixel are examined. To solve the problems, methods by fuzzy sigmoid function and back-propagation neural network are suggested. Then. the study simulates and comparatively analyzes the two methods.

• Journal of KIISE:Software and Applications
• /
• v.32 no.8
• /
• pp.789-794
• /
• 2005

LDA (Linear Discriminant Analysis) provides the projection that discriminates the data well, and shows a very good performance for face recognition. However, since LDA provides only one transformation matrix over whole data, it is not sufficient to discriminate the complex data consisting of many classes like honan faces. To overcome this weakness, we propose a new face recognition method, called LDA mixture model, that the set of alf classes are partitioned into several clusters and we get a transformation matrix for each cluster. This detailed representation will improve the classification performance greatly. In the simulation of face recognition, LDA mixture model outperforms PCA, LDA, and PCA mixture model in terms of classification performance.

• Proceedings of the Korean Environmental Sciences Society Conference
• /
• 2003.11b
• /
• pp.241-244
• /
• 2003

2000년에 발생한 12호 태풍(Prapiroon)에 의한 성층 호수의 식물플랑크톤 군집 변화양상을 주암호의 복교지점에서 2000년 8월 30일부터 9월 29일 까지 조사하였다. 프라피룬은 복교지점에서 초속 40m의 강풍을 동반하였는데, 이 물리력은 성층으로 인해 불연속적 층을 형성하고 있던 수괴 전체를 혼합시켰다. 수체의 혼합은 격리로 인해 이질적 환경과 생물조성을 이루고 있던 층간의 혼합을 유도하여 식물플랑크톤의 우점 분류군, 종조성, 우점종을 교체시키는 요인으로 작용한 것으로 추측되었다. 식물플랑크톤의 우점 분류군은 태풍 이전의 규조류에서 녹조류로 교체되었고, 후에 남조류로 교체되었다. 우점종은 태풍 이전에는 Melosira granulata였으나 태풍 이후에는 Scenedesmus spp.와 Staurastrum sp.가 우점하는 것으로 나타났다. 개체수의 밀도는 태풍에 의한 변화가 크지 않았다. 태풍에 의해 파괴되었던 성층은 태풍 이후 재성층화 되기 시작하여 6일째엔 안정된 성층이 관찰되었다. Secchi depth는 태풍에 의해 많이 감소하였으나 수체가 안정화되면서 다시 증가하는 양상을 나타냈다. 결론적으로 태풍은 여름의 성층으로 분리된 수괴를 혼합시킬 수 있는 물리력을 가져 수서 생태계 내 생물의 동태에 영향을 미칠 수 있음을 알 수 있다.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.16 no.11
• /
• pp.2150-2158
• /
• 1992

This study was carried out to investigate the flow structure and mixing process of a cross mixing flow formed by two round jets with respect to the ratio of mass flow rate. This flow configuration is of great practical relevance in a variety of combustion systems, and the flow behaviour of a cross jet defends critically on the ratio of mass flow rate and the cross angle. The mass flow rate ratios of two different jets were controlled as 1.0, 0.8, 0.6, and 0.4, and the crossing angle of two round jets was fixed at 45 degree. The velocities issuing from jet nozzle with an exit diameter of 20mm were adjusted to 40m/s, 32m/s, 24m/s, and 16m/s, and the measurements have been conducted in the streamwise range of $1.1X_0$to $2.5X_0$ by an on-line measurement system consisted of a constant temperature type two channel hot-wire anemometry connected to a computer analyzing system. The original air flow was generated by a subsonic wind tunnel with reliable stabilities and uniform flows in the test section. For the analysis of the cross mixing flow structure in the downstream region after the cross point, the mean velocity profiles, the resultant velocity contours, and the three-dimensional profiles depending upon the mass flow rate ratio have been concentrately studied.

• The KIPS Transactions:PartB
• /
• v.11B no.5
• /
• pp.545-554
• /
• 2004

Combining multiple classifiers to obtain improved performance over the individual classifier has been a widely used technique. The task of constructing a multiple classifier system(MCS) contains two different Issues how to generate a diverse set of base-level classifiers and how to combine their predictions. In this paper, we review the characteristics of existing multiple classifier systems : Bagging, Boosting, and Slaking. For document classification, we propose new MCSs such as Stacked Bagging, Stacked Boosting, Bagged Stacking, Boosted Stacking. These MCSs are a sort of hybrid MCSs that combine advantages of existing MCSs such as Bugging, Boosting, and Stacking. We conducted some experiments of document classification to evaluate the performances of the proposed schemes on MEDLINE, Usenet news, and Web document collections. The result of experiments demonstrate the superiority of our hybrid MCSs over the existing ones.