• Journal of the Korean Data and Information Science Society
• /
• v.20 no.2
• /
• pp.311-319
• /
• 2009

A robust L1 data depth was used in clustering and classification, so called DDclus and DDclass by Jornsten (2004). SVM-based classification works well in most of the situation but show some weakness in the presence of outliers. Proper gene selection is important in classification since there are so many redundant genes. Either by selecting appropriate genes or by gene clustering combined with classification method enhance the overall performance of classification. The performance of depth based method are evaluated among several SVM-based classification methods.

• The KIPS Transactions:PartB
• /
• v.16B no.2
• /
• pp.131-140
• /
• 2009

This paper presents a new algorithm to the segmentation of the FISH images. First, for segmentation of the cell nuclei from background, a threshold is estimated by using the gaussian mixture model and maximizing the likelihood function of gray value of cell images. After nuclei segmentation, overlapped nuclei and isolated nuclei need to be classified for exact nuclei analysis. For nuclei classification, this paper extracted the morphological features of the nuclei such as compactness, smoothness and moments from training data. Three probability density functions are generated from these features and they are applied to the proposed Bayesian networks as evidences. After nuclei classification, segmenting of overlapped nuclei into isolated nuclei is necessary. This paper first performs intensity gradient transform and watershed algorithm to segment overlapped nuclei. Then proposed stepwise merging strategy is applied to merge several fragments in major nucleus. The experimental results using FISH images show that our system can indeed improve segmentation performance compared to previous researches, since we performed nuclei classification before separating overlapped nuclei.

• 한국공간정보시스템학회:학술대회논문집
• /
• 2005.11a
• /
• pp.41-45
• /
• 2005

분광각(Spectral Angle)을 이용한 분류는 같은 종류의 지표 대상물의 분광 특성이 대기 및 지형적인 영향으로 인해 원점을 기준으로 선형적인 분포 모양을 가진다는 가정에 기초한 새로운 접근의 분류 방식이다. 최근 분광각을 이용한 무감독 분류에 대한 연구가 활발히 이루어지고 있으나, 원격탐사 데이터의 특성을 반영한 효과적인 무감독 분류에 대한 연구는 미진한 상태이다. 본 연구는 하이퍼스펙트럴 영상 분류에 있어서 기존 무감독 분광각 분류(USAC, Unsupervised Spectral Angle Classification) 연구에서 해결하지 못한 문제점들을 보완한 반복최적화 무감독 분광각 분류(ISOUSAC, Iterative Self-Organizing USAC) 기법을 제안하고 있다. 이를 위해, 무감독 분광각 분류에 적합한 각 분할(Angle Range Division) 기법을 적용하여 군집 초기 중심을 설정하였으며, 병합(Merge)과 분할(Split)를 통한 유동적인 군집 분석을 수행하였다. 결과를 통해, 제안된 알고리즘이 기존의 기법보다 수행 시간뿐 아니라 시각적인 면에서도 우수한 결과를 도출함을 확인할 수 있었다.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.6
• /
• pp.603-616
• /
• 2023

In this study, machine learning-based clustering and classification of residential noise in apartment buildings was conducted using frequency and temporal characteristics. First, a residential noise source dataset was constructed . The residential noise source dataset was consisted of floor impact, airborne, plumbing and equipment noise, environmental, and construction noise. The clustering of residential noise was performed by K-Means clustering method. For frequency characteristics, L_eq and L_max values were derived for 1/1 and 1/3 octave band for each sound source. For temporal characteristics, Leq values were derived at every 6 ms through sound pressure level analysis for 5 s. The number of k in K-Means clustering method was determined through the silhouette coefficient and elbow method. The clustering of residential noise source by frequency characteristic resulted in three clusters for both L_eq and L_max analysis. Temporal characteristic clustered residential noise source into 9 clusters for L_eq and 11 clusters for L_max. Clustering by frequency characteristic clustered according to the proportion of low frequency band. Then, to utilize the clustering results, the residential noise source was classified using three kinds of machine learning. The results of the residential noise classification showed the highest accuracy and f1-score for data labeled with L_eq values in 1/3 octave bands, and the highest accuracy and f1-score for classifying residential noise sources with an Artificial Neural Network (ANN) model using both frequency and temporal features, with 93 % accuracy and 92 % f1-score.

• Fire Science and Engineering
• /
• v.23 no.1
• /
• pp.15-20
• /
• 2009

For the effective management of forest fire, understanding of regional forest fire patterns is needed. In this paper, forest fire ignition and spread characteristics were analyzed based on forest fire statistics. Fire occurrences, burned area, rate of spread, and burned area per fire between 1991 and 2007 were parameterized for the cluster analysis, which results were displayed using GIS to detect spatial patterns of forest fire. Administrative districts such as cities and counties were classified into 5 clusters by fire susceptibility. Metropolitan areas had fire characteristics that were infrequent, slow rate of spread, and small burned area. However, 4 cities and counties showing fast rate of spread, and large burned area, in the eastern regions of Taeback Mountain range, were the most susceptible areas to forest fire. The next vulnerable cities and counties were located in the West and South Coast area.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.10a
• /
• pp.22-26
• /
• 2018

In this paper, we propose a model that classifies different transaction information by clustering and learning through similarity and transaction pattern of cryptocurrency transaction information. By using characteristics of genetic algorithms, we can get better clustering performance by eliminating unnecessary elements in clustering process. The transaction information including the clustering value is set as the training data, and the transaction information can be predicted through the classification algorithm. This can be used to automatically detect abnormal transactions from various transaction information of the cryptocurrency.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2005.05a
• /
• pp.240-249
• /
• 2005

사용자 맞춤 서비스를 위하여 온라인상에서 사용자의 관심 분야를 파악하고자 하는 경우에는 적은 수의 훈련 예제로 효율적인 학습이 가능한 능동적 학습이 적절하다. 능동적 학습을 효과적으로 적용하기 위하여 사용자에게 문의할 가치가 높은 예제를 선정하는 것도 중요하지만, 사용자 편의를 위해서는 문의 횟수를 가능한 최소화하여야 한다. 문의 횟수를 줄이면서도 많은 수의 훈련 예제를 획득하기 위해서는 복수의 문의 예제들을 사용자에게 한꺼번에 제시하고 그 관심 여부를 표한하게 하는 것이 효과적이다. 본 논문에서는 능동적 학습 적용 시 사용자에게 문의할 가치가 높은 복수 문의 예제들을 효과적으로 선정하기 위하여 가중치 반영 군집화를 적용하는 방안을 제안한다. 본 제안 방안은 먼저 각 예제의 문의 예제로서의 가치를 파악하고 이를 가중치로 삼아 군집화를 수행하여 상대적으로 유사한 예제들의 집합을 구성한다. 이어서 생성된 각각의 군집에서 가장 보편적인 예제를 문의 예제로 선정하면 선정된 각각의 문의 예지는 문의할 가치가 높으면서 함께 문의하게 될 예제들은 서로 충분히 달라 학습에 보다 유용하게 사용할 수 있는 훈련 예제들을 얻을 수 있다. 문서 분류 문제를 대상으로 본 제안 방안을 실험한 결과, 단순히 문의 가치가 높은 복수의 예제들을 함께 문의할 예제들로 선정하는 방안에 비해 학습 성능이 뛰어났으며, 한 번에 문의하는 예제 수를 증가시키더라도 분류기의 성능 저하가 적음을 확인하였다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.5
• /
• pp.755-762
• /
• 2022

This paper propose a clustered algorithm that possible more efficient COVID-19 disease learning prediction within clustering using context-aware attribute information. In typically, clustering of COVID-19 diseases provides to classify interrelationships within disease cluster information in the clustering process. The clustering data will be as a degrade factor if new or newly processing information during treated as contaminated factors in comparative interrelationships information. In this paper, we have shown the solving the problems and developed a clustering algorithm that can extracting disease correlation information in using K-means algorithm. According to their attributes from disease clusters using accumulated information and interrelationships clustering, the proposed algorithm analyzes the disease correlation clustering possible and centering points. The proposed algorithm showed improved adaptability to prediction accuracy of the classification management system in terms of learning as a group of multiple disease attribute information of COVID-19 through the applied simulation results.

• The Korean Journal of Applied Statistics
• /
• v.14 no.1
• /
• pp.177-190
• /
• 2001

전통적으로 많이 사용하는 군집분석의 방법들은 개체간의 거리를 고려하여 이들을 분류해 내는 것이며, 따라서 거리 측정 방법에 따라 여러 형태의 군집분석 방법이 나타나게 된다. 어떤 방법을 적용하던 간에 그 결과는 고정된 수치로써 나타난다. 다차원 자료의 구조파악이 몇 개의 수치로 나타나게 되면 어쩔 수 없이 정보의 손실이 발생하게 된다. 이를 보완하기 위해 시각적 매체를 동원하여 다차원 자료의 구조를 파악하는 연구가 있었으며, 이를 시각적 군집분석이라고 명명하고 있다. 본 연구에서는 시각적 군집분석에 대한 기본적 개념과 이를 위한 통계 도형의 활용, 구현방법 등에 대해 살펴보기로 한다.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.5
• /
• pp.999-1009
• /
• 2014

Hierarchical clustering analysis helps easily exploring massive microarray data and understanding biological phenomena with dendrogram. But, because hierarchical clustering algorithms only consider the absolute similarity, it is difficult to illustrate a relative dissimilarity, which consider not only the distance between a pair of clusters, but also how distant are they from the rest of the clusters. In this study, we introduced the relative hierarchical clustering method proposed by Mollineda and Vidal (2000) and compared hierarchical clustering method and relative hierarchical method using the simulated data and the real data in the various situations. The evaluation of the quality of two hierarchical methods was performed using percentage of incorrectly grouped points (PIGP), homogeneity and separation.