• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.10a
• /
• pp.248-249
• /
• 2015

Clustering is one of the most important unsupervised learning methods that clusters data into homogeneous groups. However, cluster centers tend leaning to high density clusters because clustering is based on the distances between data points and cluster centers. In this paper, a modified clustering method forcing cluster centers to be apart by introducing a center-scattering term in the Fuzzy C-Means objective function is introduced. The proposed method converges more to real centers with small number of iterations compared to the original one. All the strengths can be verified with experimental results.

• Journal of KIISE
• /
• v.42 no.1
• /
• pp.86-92
• /
• 2015

A level set tree provides a useful representation of a multidimensional density function. Visualizing the data structure as a tree offers many advantages for data analysis and clustering. In this paper, we present a level set tree estimation algorithm for use with a set of data points. The proposed algorithm creates a level set tree from a family of level sets estimated over a whole range of levels from zero to infinity. Instead of estimating density function then thresholding, we directly estimate the density level sets using one-class support vector machines (OC-SVMs). The level set estimation is facilitated by the OC-SVM solution path algorithm. We demonstrate the proposed level set tree algorithm on benchmark data sets.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.05a
• /
• pp.76-78
• /
• 2014

Naive Bayes (NB, for shortly) learning is more popular, faster and effective supervised learning method to handle the labeled datasets especially in which have some noises, NB learning also has well performance. However, the conditional independent assumption of NB learning imposes some restriction on the property of handling data of real world. Some researchers proposed lots of methods to relax NB assumption, those methods also include attribute weighting, kernel density estimating. In this paper, we propose a novel approach called NB Based on Attribute Weighting in Kernel Density Estimation (NBAWKDE) to improve the NB learning classification ability via combining kernel density estimation and attribute weighting.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.19 no.3
• /
• pp.127-143
• /
• 2016

This study explored the variability in the accuracy of dasymetric population estimation with different grid cell sizes. Dasymetric population maps for Fulton County, Georgia in the US were generated from 30m to 420m at intervals of 30m using an automated intelligent dasymetric mapping technique, population data, and original and simulated land use and cover data. The accuracies of dasymetric population maps were evaluated using RMSE and adjusted RMSE statistics. Lumped fractal dimension values were calculated for the dasymetric population maps generated from resolutions of 30m to 420m using the triangular prism surface area (TPSA) method. The results show that a grid cell size of 210m or smaller is required to estimate population more accurately in terms of thematic accuracy, but a grid cell size of 30m is required to meet an acceptable spatial accuracy of dasymetric population estimation in the study area. The fractal analysis also indicates that a grid cell size of 120m is the optimal resolution for dasymetric population estimation in the study area.

• The KIPS Transactions:PartB
• /
• v.13B no.1 s.104
• /
• pp.27-34
• /
• 2006

In this study, we propose this new algorithm that generates score function in ICA(Independent Component Analysis) using entropy theory. To generate score function, estimation of probability density function about original signals are certainly necessary and density function should be differentiated. Therefore, we used kernel density estimation method in order to derive differential equation of score function by original signal. After changing formula to convolution form to increase speed of density estimation, we used FFT algorithm that can calculate convolution faster. Proposed score function generation method reduces the errors, it is density difference of recovered signals and originals signals. In the result of computer simulation, we estimate density function more similar to original signals compared with Extended Infomax and Fixed Point ICA in blind source separation problem and get improved performance at the SNR(Signal to Noise Ratio) between recovered signals and original signal.

• Spatial Information Research
• /
• v.16 no.2
• /
• pp.237-247
• /
• 2008

Statistical analysis of the coordinate information is regarded as one of the major GIS functions. Among them, one of the most fundamental analysis is density analysis of point features. For analyzing the density appropriately, determining the search radius, kernel radius, has critical importance. In this study, using L-index, known as its usefulness for choosing the kernel radius in previous researches, radius for density analysis of various point features are estimated, and the behavior of L-index is studied based on the estimated results. As results, L-index is not suitable to determine the search radius for the point features that are evenly distributed with small clusters, because the pattern of the L-index is depends on the size of the study area. But for the point features with small number of highly clustered areas, L-index is suitable, because the pattern of the L-index is not affected by the size of study area.

• The KIPS Transactions:PartA
• /
• v.9A no.1
• /
• pp.75-82
• /
• 2002

A robust signal transition density propagation method for a zero delay model is presented to obtain the signal transition density for estimating the power consumption. The power estimation for the zero delay model is a proper criteria for the lower boundary of power consumption. Since the input characteristics are generally unknown at design stage, robust estimation for wide range input characteristics is very important for the power consumption. In this paper, a conventional transition estimation method will be explored. And this exploration will be analyzed with the input/output signal transition behavior and used to propose the robust signal transition density propagation for the power estimation. In order to apply to practical circuits, the reconvergent path, which is crucial to affect the exactness of the power estimation, will be studied and an algorithm to take the reconvergent path into consideration will be presented. In experiment, the proposed methodology shows better robustness, comparable accuracy and elapsed time compared to the conventional methods.

• Journal of the Korean Data and Information Science Society
• /
• v.7 no.2
• /
• pp.211-217
• /
• 1996

본 논문은 확률밀도함수의 l 번째 도함수의 커널추정법에 관하여 다루고 있다. 확률밀도함수 도함수의 커널추정에 사용될 수 있는 두가지 평활량의 선택법, 교차타당성방법과 삽입방법에 의한 평활량의 점근분포를 규명하고 이들의 상대적 수렴속도를 각각 밝히고 삽입방법의 우수성을 소표본 모의실험을 통하여 확인하였다.

• Proceedings of the KSMRM Conference
• /
• 2002.11a
• /
• pp.99-99
• /
• 2002

목적： 인체에 전류를 주입하면 체내의 생체조직의 임피던스 분포에 따라서 전류밀도 분포가 결정된다. 이러한 전류밀도 분포에 대한 정보는 전기임피던스 단층촬영법과 유방암 진단, 체내 온도 분포의 영상화, 전기자극에 의한 체내 전류 경로의 시각화에 대한 연구에 응용될 수 있다. 한편 이러한 전류밀도 분포는 전류주입 자기공명영상기법에 의해 영상화할 수 있으며, 본 논문은 3차원 팬텀 내부의 전류밀도 분포를 영상화하는 전류주입 자기공명영상기법의 실험결과를 기술한다.

• 한국지구물리탐사학회:학술대회논문집
• /
• 2005.05a
• /
• pp.181-186
• /
• 2005

Gravity method produces subsurface density distribution, which is direct information of soundness of basement. Therefore, microgravity is one of the most effective method for detections of limestone cavities, abandoned mine-shafts and other tunnels, The paper show the effectiveness of microgravity by three different field cases.