• Korean Journal of Cognitive Science
• /
• v.14 no.1
• /
• pp.11-16
• /
• 2003

This Paper extracts the edge of main components of face with Gator wavelets transformation in facial expression images. FCM(Fuzzy C-Means) clustering algorithm then extracts the representative feature points of low dimensionality from the edge extracted in neutral face. The feature-points of the neutral face is used as a template to extract the feature-points of facial expression images. To match point to Point feature points on an expression face against each feature point on a neutral face, it consists of two steps using a dynamic linking model, which are called the coarse mapping and the fine mapping. This paper presents an automatic extraction of feature-points by dynamic linking model based on Gabor wavelets and fuzzy C-means(FCM) algorithm. The result of this study was applied to extract features automatically in facial expression recognition based on dimension[1].

• The KIPS Transactions:PartB
• /
• v.11B no.4
• /
• pp.485-490
• /
• 2004

In data representation, the clustering performs a grouping process which combines given data into some similar clusters. The various similarity measures have been used in many researches. But, the validity of clustering results is subjective and ambiguous, because of difficulty and shortage about objective criterion of clustering. The fuzzy clustering provides a good method for subjective clustering problems. It performs clustering through the similarity matrix which has fuzzy membership value for assigning each object. In this paper, for objective fuzzy clustering, the clustering algorithm which joins principal components analysis as a dimension reduction model with bayesian learning as a statistical learning theory. For performance evaluation of proposed algorithm, Iris and Glass identification data from UCI Machine Learning repository are used. The experimental results shows a happy outcome of proposed model.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10b
• /
• pp.577-579
• /
• 2003

본 논문에서는 생체인식 분야 중 얼굴인식의 검색 정확성 향상 및 검색 시간을 단축하기 위한 단계로 인종별 얼굴영상 데이터베이스에 대한 군집화 기법을 연구하였다. 우선, 일반적으로 얼굴 및 이미지 검색에 사용되는 다양한 특징을 추출하고, 추출한 다차원의 특징 데이터들로부터 다 인종 얼굴 데이터를 유사한 인종별로 정확하게 군집화 하기 위해 최적의 특징벡터를 자동으로 선택 할 수 있는 방법을 제안하였다. 군집결과 분석을 위해 자기 조직화 지도 모형을 이용하였는데, 이는 2차원 분석 및 가시화에 유용하며, 학습 후 코드북벡터를 사용하여 유사한 의미간의 거리부터 검색할 수 있는 특징을 가지고 있다. 특징추출에 관한 실험결과 인종별 구분을 위한 특징벡터로는 웨이블릿 주파수 성분(lowpass 성분)과 CbCr 특징벡터가 인종별 군집화에 가장 유용한 특징으로 선택되었으며. 추출된 특징을 바탕으로 semantic map을 구성하여 제안방법의 효율성을 제시하였다.

• The Korean Journal of Applied Statistics
• /
• v.32 no.1
• /
• pp.149-160
• /
• 2019

The electricity consumption time series data of 'A' University from July 2016 to June 2017 is analyzed via nonparametric functional data clustering since the time series data can be regarded as realization of continuous functions with dependency structure. We use a Bouveyron and Jacques (Advances in Data Analysis and Classification, 5, 4, 281-300, 2011) method based on model-based functional clustering with an FEM algorithm that assumes a Gaussian distribution on functional principal components. Clusterwise analysis is provided with cluster mean functions, densities and cluster profiles.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.1
• /
• pp.27-35
• /
• 2021

In this paper, we propose a new stacking ensemble framework for deep learning models which reflects the distribution of label embeddings. Our ensemble framework consists of two phases: training the baseline deep learning classifier, and training the sub-classifiers based on the clustering results of label embeddings. Our framework aims to divide a multi-class classification problem into small sub-problems based on the clustering results. The clustering is conducted on the label embeddings obtained from the weight of the last layer of the baseline classifier. After clustering, sub-classifiers are constructed to classify the sub-classes in each cluster. From the experimental results, we found that the label embeddings well reflect the relationships between classification labels, and our ensemble framework can improve the classification performance on a CIFAR 100 dataset.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2005.11a
• /
• pp.229-233
• /
• 2005

모집단의 최적군집 수를 자동으로 결정하고 군집내의 분산은 최소로 하고 군집 간의 분산은 최대로 하는 최적 군집화에 대한 연구는 대부분의 지능형 시스템에서 필요로 하는 모형전략이다. 하지만 아직도 대부분의 군집화 과정에서 분석가의 주관적인 경험에 의존하여 군집수가 결정되어 군집화가 이루어지고 있다. 예를 들어 K-평균 군집화 알고리즘에서도 초기에 K 값을 결정해 주어야 한다. 모집단을 제대로 대표하지 못한 K 값에 의한 군집화 결과는 심각한 오류를 범하게 된다. 본 논문에서는 통계적 학습이론을 이용하여 이러한 문제점을 해결하려고 하였다. VC-차원에 의한 Support Vector를 이용하여 최적의 군집화 기법을 제안하였다. 제안 방법의 성능 평가를 위하여 UCI 기계학습 데이터를 이용하여 객관적인 실험을 수행하였다.

• The Korean Journal of Applied Statistics
• /
• v.27 no.1
• /
• pp.21-30
• /
• 2014

Ordinary survival analysis cannot be applied when a significant fraction of patients may be cured. A cure rate model is the combination of cure fraction and survival model and can be applied to several types of cancer. In this article, the cure rate model is considered in the interval censored data with a cluster effect. A shared frailty model is introduced to characterize the cluster effect and an EM algorithm is used to estimate parameters. A simulation study is done to evaluate the performance of estimates. The proposed approach is applied to the smoking cessation study in which the event of interest is a smoking relapse. Several covariates (including intensive care) are evaluated to be effective for both the occurrence of relapse and the smoke quitting duration.

• Korean Journal of Cognitive Science
• /
• v.14 no.1
• /
• pp.10-10
• /
• 2003

This paper extracts the edge of main components of face with Gabor wavelets transformation in facial expression images. FCM(Fuzzy C-Means) clustering algorithm then extracts the representative feature points of low dimensionality from the edge extracted in neutral face. The feature-points of the neutral face is used as a template to extract the feature-points of facial expression images. To match point to Point feature points on an expression face against each feature point on a neutral face, it consists of two steps using a dynamic linking model, which are called the coarse mapping and the fine mapping. This paper presents an automatic extraction of feature-points by dynamic linking model based on Gabor wavelets and fuzzy C-means(FCM) algorithm. The result of this study was applied to extract features automatically in facial expression recognition based on dimension[1].

• The Korean Journal of Financial Management
• /
• v.20 no.2
• /
• pp.95-126
• /
• 2003

This paper introduces multifractal processes and presents the empirical investigation of the multifractal asset pricing. The multifractal stock price process contains long-tails which focus on Levy-Stable distributions. The process also contains long-dependence, which is the characteristic feature of fractional Brownian motion. Multifractality introduces a new source of heterogeneity through time-varying local reqularity in the price path. This paper investigates multifractality in stock prices. After finding evidence of multifractal scaling, the multifractal spectrum is estimated via the Legendre transform. The distinguishing feature of the multifractal process is multiscaling of the return distribution's moments under time-resealing. More intensive study is required of estimation techniques and inference procedures.

• The Korean Journal of Applied Statistics
• /
• v.29 no.1
• /
• pp.193-203
• /
• 2016

This paper forecasts electricity demand as a critical element of a demand management system in Smart Grid environment. We present a prediction method of using a combination of predictive values by time series clustering. Periodogram-based normalized clustering, predictive analysis clustering and dynamic time warping (DTW) clustering are proposed for time series clustering methods. Double Seasonal Holt-Winters (DSHW), Trigonometric, Box-Cox transform, ARMA errors, Trend and Seasonal components (TBATS), Fractional ARIMA (FARIMA) are used for demand forecasting based on clustering. Results show that the time series clustering method provides a better performances than the method using total amount of electricity demand in terms of the Mean Absolute Percentage Error (MAPE).