• Communications for Statistical Applications and Methods
• /
• 제22권6호
• /
• pp.575-587
• /
• 2015

The development in data collection techniques results in high dimensional data sets, where discrimination is an important and commonly encountered problem that are crucial to resolve when high dimensional data is heterogeneous (non-common variance covariance structure for classes). An example of this is to classify microbial habitat preferences based on codon/bi-codon usage. Habitat preference is important to study for evolutionary genetic relationships and may help industry produce specific enzymes. Most classification procedures assume homogeneity (common variance covariance structure for all classes), which is not guaranteed in most high dimensional data sets. We have introduced regularized elimination in partial least square coupled with QDA (rePLS-QDA) for the parsimonious variable selection and classification of high dimensional heterogeneous data sets based on recently introduced regularized elimination for variable selection in partial least square (rePLS) and heterogeneous classification procedure quadratic discriminant analysis (QDA). A comparison of proposed and existing methods is conducted over the simulated data set; in addition, the proposed procedure is implemented to classify microbial habitat preferences by their codon/bi-codon usage. Five bacterial habitats (Aquatic, Host Associated, Multiple, Specialized and Terrestrial) are modeled. The classification accuracy of each habitat is satisfactory and ranges from 89.1% to 100% on test data. Interesting codon/bi-codons usage, their mutual interactions influential for respective habitat preference are identified. The proposed method also produced results that concurred with known biological characteristics that will help researchers better understand divergence of species.

• Industrial Engineering and Management Systems
• /
• 제9권2호
• /
• pp.131-140
• /
• 2010

Biclustering is a method of finding meaningful subsets of objects and attributes simultaneously, which may not be detected by traditional clustering methods. It is popularly used for the analysis of microarray data representing the expression levels of genes by conditions. Usually, biclustering algorithms do not consider a sequential relation between attributes. For time series data, however, bicluster solutions should keep the time sequence. This paper proposes a new biclustering algorithm for time series data by modifying the plaid model. The proposed algorithm introduces a parameter controlling an interval between two selected time points. Also, the pruning step preventing an over-fitting problem is modified so as to eliminate only starting or ending points. Results from artificial data sets show that the proposed method is more suitable for the extraction of biclusters from time series data sets. Moreover, by using the proposed method, we find some interesting observations from real-world time-course microarray data sets and apartment price data sets in metropolitan areas.

• 정보처리학회논문지B
• /
• 제10B권1호
• /
• pp.33-40
• /
• 2003

데이터 마이닝은 지식발견 과정에서 중요한 역할을 수행하며, 여러 데이터 마이닝의 알고리즘들은 특정의 목적을 위하여 선택될 수 있다. 대부분의 전통적인 계층적 클러스터링 방법은 적은 양의 데이터 집합을 처리하는데 적합하여 제한된 리소스와 부족한 효율성으로 인하여 대용량의 데이터 집합을 다루기가 곤란하다. 본 연구에서는 대용량의 데이터에 적용되어 알려지지 않은 패턴을 발견할 수 있는 하이브리드형 신경망 클러스터링 기법의 PPC(Pre-Post Clustrering) 기법을 제안한다. PPC 기법은 인공지능적 방법인 자기조직화지도(SOM)와 통계적 방법인 계층적 클러스터링을 결합하여 두 과정에서는 군집의 내부적 특징을 나타내는 응집거리와 군집간의 외부적 거리를 나타내는 인접거리에 따라 유사도를 측정한다. 최종적으로 PPC 기법은 측정된 유사도를 이용하여 대용량 데이터 집합을 군집화한다. PPC 기법은 UCI Repository 데이터를 이용하여 실험해 본 결과, 다른 클러스터링 기법들 보다 우수한 응집도를 보였다.

• 대한수학교육학회지:학교수학
• /
• 제19권4호
• /
• pp.691-712
• /
• 2017

교사의 통계적 추론과 사고를 개발하는 것은 예비교사 대상의 통계교육에서 이루어져야 하는 중요한 역할이다. 본 연구에서는 특히, 예비초등교사들이 통계의 핵심 아이디어에 대한 추론을 발달시키기 위한 방안으로서 두 자료집합 비교하기 과제의 활용에 주목하였다. 24명의 예비초등교사들이 4명씩 6개 모둠으로 과제를 수행하고 이를 발표하게 함으로써 자료를 수집하였고, 두 자료집합 비교 활동에서 확인된 Pfannkuch(2006)의 추론 모델을 바탕으로 이를 분석하였다. 분석 결과, 연구 참여자들은 두 자료집합 비교하기 과제를 통해 통계적 문제해결을 위해 분포와 변이성에 주목하였고, 의사결정을 위해 맥락적 지식을 고려하는 모습을 보였다. 또한, 통계적 의사소통을 위한 주된 참조물로서 통계량과 그래프를 활용하였는데, 이는 절차적 지식에 고착화된 전통적 통계교육을 개선하기 위한 주요한 시사점을 제공할 수 있을 것으로 기대된다. 이를 통해, 두 자료집합 비교하기 과제가 예비초등교사교육에서 지니는 가능성을 확인함과 동시에 활용 방안에 대한 제언을 도출하였다.

• Journal of Periodontal and Implant Science
• /
• 제51권1호
• /
• pp.18-29
• /
• 2021

Purpose: The aim of this study was to compare the characteristic expression patterns of advanced periodontitis in 2 cohort data sets analyzed using different microarray platforms, and to identify differentially expressed genes (DEGs) through a meta-analysis of both data sets. Methods: Twenty-two patients for cohort 1 and 40 patients for cohort 2 were recruited with the same inclusion criteria. The 2 cohort groups were analyzed using different platforms: Illumina and Agilent. A meta-analysis was performed to increase reliability by removing statistical differences between platforms. An integrative meta-analysis based on an empirical Bayesian methodology (ComBat) was conducted. DEGs for the integrated data sets were identified using the limma package to adjust for age, sex, and platform and compared with the results for cohorts 1 and 2. Clustering and pathway analyses were also performed. Results: This study detected 557 and 246 DEGs in cohorts 1 and 2, respectively, with 146 and 42 significantly enriched gene ontology (GO) terms. Overlapping between cohorts 1 and 2 was present in 59 DEGs and 18 GO terms. However, only 6 genes from the top 30 enriched DEGs overlapped, and there were no overlapping GO terms in the top 30 enriched pathways. The integrative meta-analysis detected 34 DEGs, of which 10 overlapped in all the integrated data sets of cohorts 1 and 2. Conclusions: The characteristic expression pattern differed between periodontitis and the healthy periodontium, but the consistency between the data sets from different cohorts and metadata was too low to suggest specific biomarkers for identifying periodontitis.

• Journal of the Korean Data and Information Science Society
• /
• 제18권2호
• /
• pp.481-488
• /
• 2007

The k-means algorithm is well known for its efficiency in clustering large data sets. However, working only on numeric values prohibits it from being used to cluster real world data containing categorical values. The k-modes algorithm is to extend the k-means paradigm to categorical domains. The algorithm requires a pre-setting or random selection of initial points (modes) of the clusters. This paper improved the problem of k-modes algorithm, using the Max-Min method that is a kind of methods to decide initial values in k-means algorithm. we introduce new similarity measures to deal with using the categorical data for clustering. We show that the mushroom data sets and soybean data sets tested with the proposed algorithm has shown a good performance for the two aspects(accuracy, run time).

• Asian-Australasian Journal of Animal Sciences
• /
• 제12권6호
• /
• pp.850-853
• /
• 1999

Angus postweaning daily gain (PWDG) was analyzed to investigate effects of the heterogeneous variance and the genotypes by sex interaction on prediction of EBVs with data sets of various environmental levels. A whole data (16,239 records) was divided into six data sets according to averages of the best linear unbiased estimator (BLUE) of herd environment. The results comparing prediction models showed that single-trait model is adequate for most of the data sets except for the data set of poor environment for both of the bulls and the heifers where the heterogeneity of variance and the genotypes by sex interaction exists. In the prediction with the data set of the low environment level, the bull's EBVs by single-trait models had high product moment correlations with male EBVs of the bulls by the multitrait model. Whereas the heifer's EBVs had moderate correlations with female EBVs by the multitrait model. This moderate correlation seems to be resulted by the heterogeneity of variance and low heritability of the heifer's PWDG. The prediction models with heterogeneity of variance had little effect on the prediction of EBVs for the data sets with moderate to high genetic correlations.

• 한국지능시스템학회논문지
• /
• 제19권4호
• /
• pp.545-549
• /
• 2009

마이크로어레이 데이터는 수천가지 유전자의 발현정보를 포함할 수 있으며, 여기에서 의미있는 패턴을 추출하여 추가적인 분석을 위한 목적으로 활용되고 있다. 다수의 샘플 또는 실험에 대해서 마이크로어레이 데이터가 수집된 경우에 분석자가 관심을 갖는 유전자들이나 샘플들을 효과적으로 검색하는 것이 필요한 경우가 있다. 이 논문에서는 단순한 조건뿐만 아니라 복잡한 조건을 정의하여 원하는 특성을 만족하는 유전자나 샘플을 추출하는 방법으로 퍼지 시그너쳐 집합을 활용하는 방법을 제안한다. 퍼지 시그너쳐는 벡터값을 값을 갖는 퍼지 집합을 확장한 것으로, 벡터의 각 요소가 다시 벡터가 되는 것을 허용하는 재귀적인 구조이다. 퍼지 시그너쳐 집합은 단말 원소가 구간 [0,1] 사이에서 정의된 퍼지집합이라는 것을 제외하면 퍼지 시그너쳐와 같은 구조를 가진다. 이 논문에서는 각 내부 노드에 대해서 명시적으로 결합 연산자를 지정하도록 하고, 결합 연산을 위해 비교연산자를 사용할 수 있도록 확장한 퍼지 시그너쳐 집합을 소개한다. 또한 확장된 퍼지 시그너쳐 집합을 마이크로어레이 데이터 검색을 위해 사용하는 방법과 이를 사용한 예를 보인다.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1998년도 The Third Asian Fuzzy Systems Symposium
• /
• pp.314-319
• /
• 1998

In this paper, for efficient data classification in multispectral bands environment, a band feature extraction method using the Rough sets theroy is proposed. First, we make a look up table from training data, and analyze the properties of experimental multispectral image data, then select the efficient band usin indiscernibility relation of Rough sets theory from analysis results. Proposed method is applied to LAMDSAT TM data on 2, June, 1992. Among them, normal distributive data were experimented, mainly. From this, we show clustering trends that similar to traditional band selection results by wavelength properties, from this, we verify that can use the proposed method that centered on data properties to select the efficient bands, though data sensing environment change to hyperspectral band environments.

• Journal of Advanced Marine Engineering and Technology
• /
• 제34권3호
• /
• pp.407-412
• /
• 2010

According to the remarkable advances in sonar technology, positioning capabilities and computer processing power we can accurately image and explore the seafloor in hydrography. Especially, Multibeam Echo Sounder can provide nearly perfect coverage of the seafloor with high resolution. Since the mid-1990's, Multibeam Echo Sounders have been used for hydrographic surveying in Korea. In this study, new marine data set as an effective decision-making tool in various fields was proposed by visualizing and combining with Multibeam sonar data and marine spatial data sets such as satellite image and digital nautical chart. The proposed method was tested around the port of PyeongTaek-DangJin in the west coast of Korea. The Visualization and fusion methods are described with various marine data sets with processing. We demonstrated that new data set in marine GIS is useful in safe navigation and port management as an efficient decision-making tool.