• The Journal of the Acoustical Society of Korea
• /
• v.21 no.6
• /
• pp.585-591
• /
• 2002

While the multilayer perceptron(MLP) provides several advantages against the existing pattern recognition methods, it requires relatively long time in learning. This results in prolonging speaker enrollment time with a speaker verification system that uses the MLP as a classifier. This paper proposes a method that shortens the enrollment time through adopting the cohort speakers method used in the existing parametric systems and reducing the number of background speakers required to learn the MLP, and confirms the effect of the method by showing the result of an experiment that applies the method to a continuant and MLP-based speaker verification system.

• Investigative Magnetic Resonance Imaging
• /
• v.6 no.2
• /
• pp.137-146
• /
• 2002

Purpose : This paper introduces an improved classification approach which adopts a deterministic relaxation method and an agglomerative clustering technique for the classification of MRI using neural network. The proposed approach can solve the problems of convergency to local optima and computational burden caused by a large number of input patterns when a neural network is used for image classification. Materials and methods : Application of Hopfield neural network has been solving various optimization problems. However, major problem of mapping an image classification problem into a neural network is that network is opt to converge to local optima and its convergency toward the global solution with a standard stochastic relaxation spends much time. Therefore, to avoid local solutions and to achieve fast convergency toward a global optimization, we adopt MFA to a Hopfield network during the classification. MFA replaces the stochastic nature of simulated annealing method with a set of deterministic update rules that act on the average value of the variable. By minimizing averages, it is possible to converge to an equilibrium state considerably faster than standard simulated annealing method. Moreover, the proposed agglomerative clustering algorithm which determines the underlying clusters of the image provides initial input values of Hopfield neural network. Results : The proposed approach which uses agglomerative clustering and deterministic relaxation approach resolves the problem of local optimization and achieves fast convergency toward a global optimization when a neural network is used for MRI classification. Conclusion : In this paper, we introduce a new paradigm to classify MRI using clustering analysis and deterministic relaxation for neural network to improve the classification results.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.404-404
• /
• 2017

이상기후현상으로 인해 극치 수문 사상들이 빈번히 발생함에 따라 상대적으로 높은 재현기간에 해당하는 극치 수문 사상해석에 대한 관심이 높아지고 있다. 그러나 우리나라의 경우 이러한 극치 수문 사상을 추정하기 위한 표본의 수가 부족한 실정이다. 지역빈도해석은 지점의 표본 수가 적거나 수문자료의 수집이 불가능한 미계측지점인 경우, 해당 지점과 수문학적으로 동질하다고 여겨지는 주변 지점들의 자료를 확보하여 확률수문량을 추정함으로써 상대적으로 지점빈도해석 보다 roubst한 추정값을 얻을 수 있다는 장점을 가지고 있다. 따라서 최근 확률수문량 산정 기법으로 지역빈도해석 방법에 관한 관심이 높아지고 있다. 지역구분은 지역빈도해석이 지점빈도해석과 구분될 수 있는 큰 특징이고 지역구분 결과 따라 지역의 표본 크기가 결정되기 때문에 수문학적으로 동질한 지역을 나누는 방법은 매우 중요하다고 볼 수 있다. 인공신경망은 인간의 뇌가 학습하는 방식을 모사한 통계적 모델링 기법이다. 즉, 인간의 뇌가 일정한 반복 학습을 통해 어떠한 문제의 해법을 추론하거나 예측, 또는 패턴을 인식하는 일련의 과정을 알고리즘화 하여 목적함수의 해를 찾는 방식이다. 특히, 주어진 자료들로 부터 특징을 추출하고 그 특징을 학습하여 전체 자료의 분류나 군집화를 이루는데 널리 이용되고 있다. 본 연구에서는 낙동강유역을 대상으로 인공신경망을 이용한 군집분석을 수행하고 구분된 지역을 이용하여 지역빈도해석을 수행하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2002.11a
• /
• pp.393-396
• /
• 2002

최근 대용량의 데이터베이스로부터 유용한 정보를 발견하고 데이터간에 존재하는 연관성을 탐색하고 분석하는 데이터 마이닝에 관한 많은 연구들이 진행되고 있다. 실제 응용분야에선 수집된 데이터는 시간이 지날수록 데이터의 양이 늘어나게 되고, 중복되는 속성과 잡음을 갖게 되어 마이닝 기법을 이용하는데 많은 시간과 비용이 소요된다. 또한 어느 속성이 중요한지 알 수 없어 중요한 속성이 중요하지 않은 속성에 의해 왜곡되거나 제대로 분석되지 않을 수 있다. 본 연구는 이러한 문제점들을 해결하기 위해 GHSOM을 이용한 계층적 신경망 군집화 방법을 제안한다. 제안하는 방법은 비리 군집의 개수를 정해줄 필요가 없고, 다양한 레벨의 군집들을 얻을 수 있는 계층적 군집화를 이루어낸다는 장점을 갖는다. 본 논문에서는 신경망 GHSOM의 구조와 특성에 대해 간략히 살펴보고 시스템 처리과정에 대해 설명한다.

• Journal of the Korean Geotechnical Society
• /
• v.23 no.7
• /
• pp.65-76
• /
• 2007

Recently, there have been many construction projects on soft ground with growth of industry and various construction problems concerning soft soil behavior also have been reported. Especially, foundation piles of abutments and (or) buildings which were constructed on the soft ground have been suffering from a lot of stability problems of inordinary displacement due to lateral flow of soft ground. Although many researches for this phenomena have been carried out, it is still difficult to assess the mechanism of lateral flow on soft ground quantitatively. And reliable design method for judgement of lateral flow occurrence is not established yet. In this study, PNN (probabilistic neural network) and CNN (committee neural network) theories were applied for judgment of lateral flow occurrence based on eat data compiled from Korea and Japan. Predictions of PNN and CNN models for new data which were not used during model development are compared with those predicted by conventional empirical methods. It was found that the developed PNN and CNN models can predict more precise and reliable judgment of lateral flow occurrence than conventional empirical methods.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10c
• /
• pp.363-365
• /
• 1998

텍스트 마이닝(Text Mining)이란 텍스트형태의 문서들의 패턴 또는 관계를 추출하여 사용자가 원하는 새로운 정보를 가공하거나 기존의 정보를 변형하는 과정을 말한다. 텍스트 마이닝의 기능에는 문서 범주화(Document Categorization), 문서 군집화(Document Clustering), 그리고 문서 요약(Document Summarization)이 이에 해당된다. 문서 범주화란 문서에게 사전에 정의한 범주를 부여하는 과정을 말하고, 문서 군집화란 문서들을 계층적 구조로 형성하는 과정을 말하고, 문서 요약이란 문서의 전체 내용을 대표할 수 있는 내용의 일부만을 추출하는 과정을 말한다. 이 논문에서는 문서 범주화만을 다룰 것이며 그 대상으로는 신문기사로 설정하였다. 그의 범주는 4가지로 정치, 경제, 스포츠, 그리고 정보통신으로 설정하였다. 문서 범주화는 문서 분류(Document Classification)라고도 하며 문서에 범주를 자동으로 부여하여 기존에 인위적으로 부여함으로써 소요되는 시간과 비용을 절감하는 것이 목적이다. 문서 범주화에 대하여 k-NN(k-Nearest Neighbor)와 신경망을 이용하였으며, 신경망을 이용한 경우가 k-NN을 이용한 경우보다 성능이 우수하였다.

• Journal of IKEEE
• /
• v.16 no.2
• /
• pp.95-101
• /
• 2012

A novel approach for the reconstruction of 3D building model from aerial image data is proposed in this paper. In this approach, a Centroid Neural Network (CNN) with a metric of line segments is proposed for connecting low-level linear structures. After the straight lines are extracted from an edge image using the CNN, rectangular boundaries are then found by using an edge-based grouping approach. In order to avoid producing unrealistic building models from grouping lined segments, a hierarchical grouping method is proposed in this paper. The proposed hierarchical grouping method is evaluated with a set of aerial image data in the experiment. The results show that the proposed method can be successfully applied for the reconstruction of 3D building model from satellite images.

• The Journal of the Acoustical Society of Korea
• /
• v.22 no.5
• /
• pp.360-366
• /
• 2003

Although multilayer perceptrons (MLPs) present several advantages against other pattern recognition methods, MLP-based speaker verification systems suffer from slow enrollment speed caused by many background speakers to achieve a low verification error. To solve this problem, the quantitative discriminative cohort speakers (QnDCS) method, by introducing the cohort speakers method into the systems, reduced the number of background speakers required to enroll speakers. Although the QnDCS achieved the goal to some extent, the improvement rate for the enrolling speed was still unsatisfactory. To improve the enrolling speed, this paper proposes the qualitative DCS (QlDCS) by introducing a qualitative criterion to select less background speakers. An experiment for both methods is conducted to use the speaker verification system based on MLPs and continuants, and speech database. The results of the experiment show that the proposed QlDCS method enrolls speakers in two times shorter time than the QnDCS does over the online error backpropagation(EBP) method.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.3D
• /
• pp.347-356
• /
• 2009

This paper is described clustering analysis of traffic characteristics-based highway classification in order to deviate from methodologies of existing highway functional classification. This research focuses on comparing the clustering techniques performance based on the total within-group errors and deriving the optimal number of cluster. This research analyzed statistical clustering method (Hierarchical Ward's minimum-variance method, Nonhierarchical K-means method) and Kohonen self-organizing maps clustering method for highway characteristic classification. The outcomes of cluster techniques compared for the number of samples and traffic characteristics from subsets derived by the optimal number of cluster. As a comprehensive result, the k-means method is superior result to other methods less than 12. For a cluster of more than 20, Kohonen self-organizing maps is the best result in the cluster method. The main contribution of this research is expected to use important the basic road attribution information that produced the highway characteristic classification.

• Korean Journal of Cognitive Science
• /
• v.10 no.1
• /
• pp.13-24
• /
• 1999

In this paper, we propose a classifier based on modular networks using an unsupervised learning method. The structure of each module is designed through stochastic analysis of input data and each module classifier data independently. The result of independent classification of each module and a measure of the nearest distance are integrated during the final data classification phase to allow more precise c classification. Computation time is decreased by deleting modules that have been classified to be incorrect during the final classification phase. Using this method. a neural network sharing the best performance was implemented without considering. lots of of variables which can affect the performance of the neural network.