• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제8권2호
• /
• pp.116-120
• /
• 2008

Clustering is a method for unsupervised learning. General clustering tools have been depended on statistical methods and machine learning algorithms. One of the popular clustering algorithms based on machine learning is the self organizing map(SOM). SOM is a neural networks model for clustering. SOM and extended SOM have been used in diverse classification and clustering fields such as data mining. But, SOM has had a problem determining optimal number of clusters. In this paper, we propose an improvement of SOM using gap statistic and probability distribution. The gap statistic was introduced to estimate the number of clusters in a dataset. We use gap statistic for settling the problem of SOM. Also, in our research, weights of feature nodes are updated by probability distribution. After complete updating according to prior and posterior distributions, the weights of SOM have probability distributions for optima clustering. To verify improved performance of our work, we make experiments compared with other learning algorithms using simulation data sets.

• 한국정보전자통신기술학회논문지
• /
• 제1권2호
• /
• pp.39-50
• /
• 2008

Developing computational methods for identifying cell cycle-regulated genes has been one of important topics in systems biology. Most of previous methods consider the periodic characteristics of expression signals to identify the cell cycle-regulated genes. However, we assume that cell cycle-regulated genes are relatively active having relatively many interactions with each other based on the underlying cellular network. Thus, we are motivated to apply the theory of multivariate phase synchronization to the cell cycle expression analysis. In this study, we apply the method known as "Self-Organizing Maps with statistical Phase Synchronization (SOMPS)", which is the combination of self-organizing map and multivariate phase synchronization, producing several subsets of genes that are expected to have interactions with each other in their subset (Kim, 2008). Our evaluation experiments show that the SOMPS algorithm is able to detect cell cycle-regulated genes as much as one of recently reported method that performs better than most existing methods.

• 한국건축시공학회지
• /
• 제20권4호
• /
• pp.347-355
• /
• 2020

베트남의 경제 성장률과 도시인구는 꾸준히 증가함에 따라 베트남의 인프라 개발 건설 시장 규모는 더욱 커질 전망이다. 그러나 인프라 개발을 위한 베트남 정부는 재정 및 행정 능력의 한계로 이러한 문제를 해결하기 위한 방안으로 PPP를 채택하고 있다. PPP는 10년 이상 지속되는 장기 사업으로 사업 실패 시 장기적 큰 피해가 될 수 있어 위험관리가 매우 중요하다. 본 연구에서는 리스크 관리에 대한 리스크 요인의 영향과 관리정도 분석에 SOM분석을 제안한다. SOM분석 방법은 시각화 분석으로, 각 리스크 요인의 색상 패턴을 통해 상관 분석이 가능하다. 본 연구는 SOM을 활용하여 베트남 PPP 사업의 위험요인의 우선순위를 결정하고자 한다.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1993년도 학술논문발표회 논문집 제12권 1호
• /
• pp.233-237
• /
• 1993

본 논문에서는 패턴 매칭 방법에 근거하여 인식 단위가 음소인 음소 기반 인식 시스템을 구성하였다. 선택한 신경망 구조는 생물학적 신경망인 코호넨(T. Kohonen)의 SOFM(Self-Organizing Feature Map)으로 패턴 매칭 과정 중 cluster로 사용하였다. SOFM 신경망은 신호 공간에 대해서 최적의 국소(局所) 해부적 사사에 의한 자기 조직화 과정을 수행하며, 그 결과 인식 문제에 있어서 상당히 높은 정확도를 나타낸다. 따라서 SOFM 신경망은 음소 인식에도 효과적으로 응용될 수 있다. 또한 음소 인식 시스템의 성능 향상을 위해 K-means 클러스터링 알고리즘이 결합된 학습 알고리즘을 제안하였다. 제안된 음소 인식 시스템의 성능을 평가하기 위해, 먼저, 우리말 음소들을 모음, 파열음, 마찰음, 파찰음, 유음 및 비음, 종성의 6개 음소군으로 분류하고 각 음소군에 대한 특징 지도를 구성하여 labeler의 기능을 수행하게 하였다. 화자 종속 인식실험 결과 87.2%의 인식률을 보였으며 제안한 학습법의 빠른 수렴성과 인식률 향상을 확인하였다.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2009년도 학술발표회 초록집
• /
• pp.1999-2003
• /
• 2009

우리나라는 기상학적으로 연중 총강수량의 약 2/3가 $6^{\sim}9$월에 편중해서 내리고 있고, 지형적으로 국토의 70% 이상이 산지로 구성되어 경사가 급해 수해를 입을 가능성이 매우 크다. 또한 산업화 및 도시화로 인해 불투수층의 증가로 강수량의 대부분이 직접유출로 기여해 강우초기에 노면상의 오염물질을 급속히 하천으로 이동시켜 오염을 가중시키고 있다. 강우-유출수 처리에 있어서 처리용량 산정 등에 이용될 수 있는 초기강우의 기준은 비점오염원 유출 연구에 있어서 대단히 중요한 요소이며, 지금까지 많은 연구자들에 많은 연구가 이루어져왔다. 그러나 유역을 구성하고 있는 토지피복에 따라 유출특성이 다르고 각각의 연구자들이 제안한 초기강우 기준이 명확하게 제시되지 못하고 있는 실정이다. 따라서 본 연구에서는 SOM(Self-Organizing Map)이론을 도입하여 본 연구의 시험유역에서 측정된 유출 및 수질자료에 대해 패턴분류를 수행하여 분할구역별 자료의 특성분석을 분석한다.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회/대한산업공학회 2005년도 춘계공동학술대회 발표논문
• /
• pp.599-606
• /
• 2005

Novelty detection involves identifying novel patterns. They are not usually available during training. Even if they are, the data quantity imbalance leads to a low classification accuracy when a supervised learning scheme is employed. Thus, an unsupervised learning scheme is often employed ignoring those few novel patterns. In this paper, we propose two ways to make use of the few available novel patterns. First, a scheme to determine local thresholds for the Self Organizing Map boundary is proposed. Second, a modification of the Learning Vector Quantization learning rule is proposed so that allows one to keep codebook vectors as far from novel patterns as possible. Experimental results are quite promising.

• 한국전산유체공학회:학술대회논문집
• /
• 한국전산유체공학회 2005년도 추계 학술대회논문집
• /
• pp.1-10
• /
• 2005

A new approach, Multi-Objective Design Exploration (MODE), is presented to address Multidisciplinary Design Optimization (MDO) problems by CFD-CSD coupling. MODE reveals the structure of the design space from the trade-off information and visualizes it as a panorama for Decision Maker. The present form of MODE consists of Kriging Model, Adaptive Range Multi Objective Genetic Algorithms, Analysis of Variance and Self-Organizing Map. The main emphasis of this approach is visual data mining. An MDO system using high fidelity simulation codes, Navier-Stokes solver and NASTRAN, has been developed and applied to a regional-jet wing design. Because the optimization system becomes very computationally expensive, only brief exploration of the design space has been performed. However, data mining result demonstrates that design knowledge can produce a good design even from the brief design exploration.

• 경영과학
• /
• 제21권3호
• /
• pp.85-99
• /
• 2004

The retention of customer is an increasingly pressing issue in today's competitive environment. The proposes of this paper is a personalized defection detection and the procedure of prevention based on economic analysis of customer defection possibility, and behaviour state transition cost. This procedure is based on the observation that potential defectors have a tendency to take a couple of months or weeks to gradually change their behaviour before their eventual withdrawal. In this procedure, the SOM(Self-Organizing Map) is used to determine the possible states of customer behaviour from past behaviour data, and to prevent the defection of potential defectors, the proposed procedure recommends the desirable behaviour state for the next period based on the analysis of transition cost. and likelihood of defection. The case study has been conducted for a Korean on-line game provider to evaluate of this procedure.

• 한국멀티미디어학회논문지
• /
• 제10권6호
• /
• pp.699-706
• /
• 2007

This article is an approach to improving the clustering of Web images by using high-level semantic features from text information relevant to Web images as well as low-level visual features of image itself. These high-level text features can be obtained from image URLs and file names, page titles, hyperlinks, and surrounding text. As a clustering algorithm, a self-organizing map (SOM) proposed by Kohonen is used. To evaluate the clustering efficiencies of SOMs, we propose a simple but effective measure indicating the accumulativeness of same class images and the perplexities of class distributions. Our approach is to advance the existing measures through defining and using new measures accumulativeness on the most superior clustering node and concentricity to evaluate clustering efficiencies of SOMs. The experimental results show that the high-level text features are more useful in SOM-based Web image clustering.

• 전자공학회논문지C
• /
• 제35C권2호
• /
• pp.71-79
• /
• 1998

In this paper, we propose a new mART(modified ART) neural network by combining the winner neuron definition method of SOM(self-organizing map) and the real-time adaptive clustering function of ART(adaptive resonance theory) and construct it in a modular structure, for the purpose of organizing the feature maps of three dimensional targets. Being constructed in a modular structure, the proposed modular mART can effectively prevent the clusters from representing multiple classes and can be trained to organze two dimensional distortion invariant feature maps so as to recognize targets with three dimensional distortion. We also present the recognition result and self-organization perfdormance of the proposed modular mART neural network after carried out some experiments with 14 tank and fighter target models.