• Communications for Statistical Applications and Methods
• /
• v.10 no.2
• /
• pp.507-517
• /
• 2003

Self-organizing map (SOM) has been developed mainly by T. Kohonen and his colleagues as a unsupervised learning neural network. Because of its topological ordering property, SOM is known to be very useful in pattern recognition and text information retrieval areas. Recently, data miners use Kohonen´s mapping method frequently in exploratory analyses of large data sets. One problem facing SOM builder is that there exists no sensible criterion for evaluating goodness-of-fit of the map at hand. In this short communication, we propose valid evaluation procedures for the Kohonen SOM of any size. The methods can be used in selecting the best map among several candidates.

• Phonetics and Speech Sciences
• /
• v.6 no.1
• /
• pp.77-83
• /
• 2014

This paper proposes a speaker and environment clustering method in order to overcome the degradation of the speech recognition performance caused by various noise and speaker characteristics. In this paper, instead of using the distance between Gaussian mixture model (GMM) weight vectors as in the Google's approach, the distance between the adapted mean vectors based on the modified maximum a posteriori (MAP) adaptation is used as a distance measure for vector quantization (VQ) clustering. According to our experiments on the simulation data generated by adding noise to clean speech, the proposed clustering method yields error rate reduction of 10.6% compared with baseline speaker-independent (SI) model, which is slightly better performance than the Google's approach.

• The Journal of Korea Robotics Society
• /
• v.6 no.2
• /
• pp.189-196
• /
• 2011

This paper presents a method of topological modeling using only low-cost sonar sensors. The proposed method constructs a topological model by extracting sub-regions from the local grid map. The extracted sub-regions are considered as nodes in the topological model, and the corresponding edges are generated according to the connectivity between two sub-regions. A grid confidence for each occupied grid is evaluated to obtain reliable regions in the local grid map by filtering out noisy data. Moreover, a convexity measure is used to extract sub-regions automatically. Through these processes, the topological model is constructed without predefining the number of sub-regions in advance and the proposed method guarantees the convexity of extracted sub-regions. Unlike previous topological modeling methods which are appropriate to the corridor-like environment, the proposed method can give a reliable topological modeling in a home environment even under the noisy sonar data. The performance of the proposed method is verified by experimental results in a real home environment.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.10a
• /
• pp.101-102
• /
• 2017

An orienteering is a sport that play with a map, a compass and a point map. However, it is difficult to measure the number of steps and the play time accurately. In this paper, we develop the Orienteering application to integrate the essential function for the orienteering sport. since the integrated UI display the map, the compass and the number of step at the same time, it is easy for an user to watch the progress of the orienteering and improve the accuracy by measuring the steps and the play time.

• Journal of the Korean Mathematical Society
• /
• v.36 no.4
• /
• pp.829-832
• /
• 1999

This to correct Section 4 of our previous work [1].

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2006.05a
• /
• pp.203-207
• /
• 2006

This paper deals with a problem of detecting BCC using confocal raman spectrum. Specially, we propose Fisher measure based filtering for rejection of frequency components being noisy or non-discriminative. we use PCA (principal component analysis) for reduction of feature space dimension. Also, we apply MAP detector for classification of BCC raman spectrum. The experimental results shows that our proposed method can reduce the feature dimension and also raise the detection ratio.

• Journal of the Chungcheong Mathematical Society
• /
• v.28 no.4
• /
• pp.647-651
• /
• 2015

Let (X, d) be a compact metric space, and let $f:X{\rightarrow}X$ be a continuous map. We consider that if a positively continuum-wise expansiveness continuous map f has the positively shadowing property in the nonwandering set, then the topological entropy is positive.

• Journal of Korean Society for Geospatial Information Science
• /
• v.19 no.4
• /
• pp.91-99
• /
• 2011

With increasing astronomically damage costs caused by frequent and large-sized flood, a hazard map containing comprehensive analysis results such as inundation trace investigation, flood possibility analysis, and evacuation plan establishment for flooded regions is a fundamental measure of non-structural flood prevention. Though an inundation trace map containing flood investigation results occurred by typhoon, rainfall and tsunami is a basic hazard map having close relationship with a flood possibility map as well as a hazard information map, it is often impossible to be produced because of financial deficiency, time delay of investigation, and the lack of maintenance for flood traces. Therefore, this study proposes the accuracy enhancement procedure of inundation trace map with flood damage information and three-dimensional Digital Elevation Model (DEM) for the past frequent flooded regions according to a guideline for inundation trace map of National Emergency Management Agency (NEMA).

• Spatial Information Research
• /
• v.20 no.5
• /
• pp.15-24
• /
• 2012

The purpose of this study was to establish an analytical framework and quantitative methodologies to analyze 'the cognitive effects of web map use' and to empirically test it. This experimental design was established based on the literature about spatial cognition. Accuracy, reaction time, and confidence within each participant were compared to measure the cognitive effects of web map use for wayfinding. Geometric accuracy of the cognitive maps was estimated and calculated based on the bidimensional regression. The experimental results showed that characteristics of map users and repetition of map use rather than types of map representation did significantly affect accuracy and reaction time of spatial cognition by using web maps. And confidence appeared to be low when the participants referred to both 2D and 3D maps for wayfinding tasks on the web maps. Understanding spatial cognition of web map users, which was suggested in the study, will help cartographers make more effectively-communicated maps.

• The Bulletin of The Korean Astronomical Society
• /
• v.39 no.2
• /
• pp.45.2-45.2
• /
• 2014

To measure the mass distribution of galaxy systems weak-lensing analysis has been widely used because it directly measures the total mass of a system regardless of its baryon content and dynamical state. However, the weak-lensing only provides a map of projected surface mass density. On the other hand, galaxy redshift surveys provide a map of the three-dimensional galaxy distribution. It thus can resolve the structures along the line of sight projected in the weak-lensing map. Therefore, the comparison of structures identified in the weak-lensing maps and in the redshift surveys is an important test of the issues limiting applications of weak-lensing to the identification of galaxy clusters. Geller et al. (2010) and Kurtz et al. (2012) compared massive clusters identified in a dense redshift survey with significant weak-lensing map convergence peaks. Both assessments of the efficiency of weak-lensing map for cluster identification did not draw a general conclusion, because the sample is so small. Thus, we additionally perform deep imaging observations of fields in a dense galaxy redshift survey that contain galaxy clusters at z~0.2-0.5, using CFHT Megacam.