• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.9C
• /
• pp.861-866
• /
• 2007

A clustering algorithm for Gaussian Probability Distribution Function (GPDF) data called Centroid Neural Network with a Bhattacharyya Kernel (BK-CNN) is proposed in this paper. The proposed BK-CNN is based on the unsupervised competitive Centroid Neural Network (CNN) and employs a kernel method for data projection. The kernel method adopted in the proposed BK-CNN is used to project data from the low dimensional input feature space into higher dimensional feature space so as the nonlinear problems associated with input space can be solved linearly in the feature space. In order to cluster the GPDF data, the Bhattacharyya kernel is used to measure the distance between two probability distributions for data projection. With the incorporation of the kernel method, the proposed BK-CNN is capable of dealing with nonlinear separation boundaries and can successfully allocate more code vector in the region that GPDF data are densely distributed. When applied to GPDF data in an image classification probleml, the experiment results show that the proposed BK-CNN algorithm gives 1.7%-4.3% improvements in average classification accuracy over other conventional algorithm such as k-means, Self-Organizing Map (SOM) and CNN algorithms with a Bhattacharyya distance, classed as Bk-Means, B-SOM, B-CNN algorithms.

• Journal of the Korean Data and Information Science Society
• /
• v.8 no.2
• /
• pp.163-171
• /
• 1997

Function approximation from a set of input-output pairs has numerous applications in scientific and engineering areas. Multilayer feedforward neural networks have been proposed as a good approximator of nonlinear function. The back propagation(BP) algorithm allows multilayer feedforward neural networks to learn input-output mappings from training samples. It iteratively adjusts the network parameters(weights) to minimize the sum of squared approximation errors using a gradient descent technique. However, the mapping acquired through the BP algorithm may be corrupt when errorneous training data we employed. When errorneous traning data are employed, the learned mapping can oscillate badly between data points. In this paper we propose a robust BP learning algorithm that is resistant to the errorneous data and is capable of rejecting gross errors during the approximation process, that is stable under small noise perturbation and robust against gross errors.

• Journal of Korea Game Society
• /
• v.18 no.2
• /
• pp.89-98
• /
• 2018

The recent Neural Network technique has shown good performance in content generation such as image generation in addition to the conventional classification problem and clustering problem solving. In this study, we propose an image generation method using artificial neural network as a next generation content creation technique. The proposed artificial neural network model receives two images and combines them into a new image by taking color from one image and shape from the other image. This model is made up of Convolutional Neural Network, which has two encoders for extracting color and shape from images, and a decoder for taking all the values of each encoder and generating a combination image. The result of this work can be applied to various 2D image generation and modification works in game development process at low cost.

• The Journal of the Convergence on Culture Technology
• /
• v.2 no.1
• /
• pp.93-98
• /
• 2016

As information technologies are rapidly developed recently, social networking services through a variety of mobile devices and smart screen is becoming popular. SNS is a social networking based services which is online forms from existed offline. SNS can also be used differently which is confused with the online community. A modelling algorithm is a variety of techniques, which are assocoation, clustering, neural networks, and decision trees, etc. By utilizing this technique, it is necessary to study to effectively using the large number of materials. In this paper, we evaluate in particular the performance of the algorithm based on the results of the clustering using Facebook Insights data for the EM algorithm to be evaluated as a good performance in clustering. Through this analysis it was based on the results of the application of the experimental data of the change and the South Australian state library according to the performance of the EM algorithm.

• Journal of KIISE:Software and Applications
• /
• v.35 no.9
• /
• pp.547-553
• /
• 2008

We propose a two-stage document layout segmentation method. At the first stage, as top-down segmentation, morphological distance map algorithm extracts a collection of rectangular regions from a given input image. This preliminary result from the first stage is employed as input parameters for the process of next stage. At the second stage, a machine-learning algorithm is adopted RBF network, one of neural networks based on statistical model, is selected. In order for constructing the hidden layer of RBF network, a data clustering technique bared on the self-organizing property of Kohonen network is utilized. We present a result showing that the supervised neural network, trained by 300 number of sample data, improves the preliminary results of the first stage.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2001.12a
• /
• pp.251-254
• /
• 2001

본 논문에서는 Self Organizing Map을 이용하여 fMRI data를 분석해 보았다. fMRl (functional Magnetic Resonance Imaging)는 인간의 뇌에 대한 비 침투적 연구 방법 중 최근에 각광받고 있는 것이다. Motor task를 수행하고 있는 피험자로부터 image data를 얻어내어 SOM을 적용하여 clustering한 결과 motor cortex 영역이 뚜렷하게 clustering 되었음을 알 수 있었다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2010.05a
• /
• pp.1164-1168
• /
• 2010

수자원이 우리 생활의 전반적으로 중요한 역할을 차지하면서 댐의 효율적인 운영과 안정적인 용수공급에 대한 연구는 지속적으로 수행되어지고 있다. 1990년대 이후 비선형적인 특성을 잘 모의하는 장점을 가진 인공신경망(ANN)을 이용하여 유입량 예측에 대한 많은 연구가 수행되었다. 하지만 ANN 모형을 포함한 회귀모형은 월 강우 및 유입량의 예측에 대해 간편하게 사용을 할 수 있지만, 예측의 정확성에 한계를 가지고 있다. 본 연구에서는 ANN 모형과 회귀모형의 예측오차를 후처리 과정을 통하여 오차를 줄임으로써 예측모형의 성과를 향상시키는 방법을 제안하였다. 연구지역은 금강수계의 대청댐 유역으로, 1982년 9월부터 2005년 12월에 해당하는 유역 내 11개 지점의 강우관측소에서 관측한 월 강우와 댐 유입량을 수집하여 모형을 구축하였다. 강우량과 유입량 자료에 대해 자기상관함수와 교차상관함수를 이용하여 입력변수를 결정하였고, 정규화를 통한 전처리 과정을 거쳐 ANN 모형과 회귀모형을 이용한 예측모형을 구축하였으며, 예측성과의 향상을 위하여 군집 분석을 이용하여 오차를 재조정하였다. 이러한 오차 후처리 과정을 포함한 모형은 RMSE와 상관계수를 이용하여 비교 평가한 결과, 예측성과를 약 40% 정도 향상시켰다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.297-299
• /
• 2021

Predicting future productivity of tight oil is an important task for analyzing residual oil recovery and reservoir behavior. In general, productivity prediction is made using the decline curve analysis(DCA). In this study, we intend to propose an effective model for predicting future production using deep learning-based recurrent neural networks(RNN), LSTM, and GRU algorithms. As input variables, the main parameters are oil, gas, water, which are calculated during the production of tight oil, and the type curve calculated through various cluster analyzes. the output variable is the monthly oil production. Existing empirical models, the DCA and RNN models, were compared, and an optimal model was derived through hyperparameter tuning to improve the predictive performance of the model.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2010.07a
• /
• pp.169-172
• /
• 2010

본 논문에서는 기존의 ISODATA 알고리즘을 네트워크 공격탐지에 더욱 적합하도록 개선하여 공격을 탐지하는 새로운 방법을 제안한다. 수많은 인터넷상의 트래픽 정보들을 군집화하여 유사도를 비교하는 방법을 통해 공격을 판단한다. 기본적인 절차는 송신자 IP와 Port, 수신자 IP와 Port 정보를 이용하여 송신자와 수신자 사이의 관계를 분석하고 그 특징 값들을 이용하여 개선된 군집화 알고리즘을 이용하여 군집화를 수행한다. 그리고 얻어진 패턴의 특징값을 인공신경망에 학습하여 공격유형을 분류하고 탐지하도록 한다. 기존의 공격탐지 방법과 비교했을 때, 계산양이 적고 속도가 빠르다는 장점이 있으며 제안하는 방법의 우수성을 실험을 통해 증명하였다.

• Journal of Korean Society of Transportation
• /
• v.27 no.2
• /
• pp.95-105
• /
• 2009

Traffic accidents on highways are likely to happen when there is an imbalance in the complex relationships among key elements such as road geometries, driver related factors, and mechanical performances. The Demand-Effort Model (DEM), which evaluates highway safety, can be explained by the imbalance, which occurs when the level of demand of the driver's attention to the road environment exceeds that of the response from the driver. This study suggests a new model that improves the reliability of the current DEM through the reinterpretation on the physiological signals with the help of the Neural Network Model (NNM). The data were collected from 149 subjects, who drove a test vehicle on the Yongdong, Honam, and Seohaean Expressways in Korea. Three important results could be drawn from the recursive tests as follows; (1) Only 5 out of 10 parameters on the physiological signals which are currently used were proven to be meaningful through the Normality Test, Cluster Analysis, and Mann-Whitney Analysis. (2) The revised DEM, which internally uses the NNM, showed more reliable results than existing DEM. Group 1, which is based on the new DEM showed 80.0% of accuracy in measuring the level of driver's efforts, however, that of Group 2 based on the current DEM was 74.3%. (3) Field tests on the Honam Expressway showed lower 'type II error' with the new DEM (40.5%) than the old DEM (58.8%). The DEM is designed as a quick and easy way to determine highway safety prior to the minute road safety audit (RSA) by a professional audit team. Then a new DEM, which is based on the NNM, needs to be considered since it showed higher reliability and lower error.