• Proceedings of the CALSEC Conference
• /
• 2004.02a
• /
• pp.107-111
• /
• 2004

The purpose of the study is to develop an integrated knowledge management system for the domains of genome and nano-technology, in which terminology-based literature mining, knowledge acquisition, knowledge structuring, and knowledge retrieval are combined. The system supports integrating different types of databases (papers and patents, technologies and innovations) and retrieving different types of knowledge simultaneously. The main objective of the system is to facilitate knowledge acquisition from documents and new knowledge discovery through a terminology-based similarity calculation and a visualization of automatically structured knowledge. Implementation issue of the system is also mentioned.

• Communications for Statistical Applications and Methods
• /
• v.21 no.6
• /
• pp.539-559
• /
• 2014

We derive the exact distribution of summation for random samples from uniform distribution and then compare the exact distribution with the approximated normal distribution obtained by the central limit theorem. To check the similarity between two distributions, we consider five existing normality tests based on the difference between the target normal distribution and empirical distribution: Anderson-Darling test, Kolmogorov-Smirnov test, Cramer-von Mises test, Shapiro-Wilk test and Shaprio-Francia test. For the purpose of comparison, those normality tests are applied to the simulated data. It can sometimes be difficult to derive an exact distribution. Thus, we try two different transformations to find out which transform is easier to get the exact distribution in terms of calculation complexity. We compare two transformations and comment on the advantages and disadvantages for each transformation.

• Proceedings of the CALSEC Conference
• /
• 2003.09a
• /
• pp.197-203
• /
• 2003

The purpose of the study is to develop an integrated knowledge structuring system for the domain of engineering, in which ontology-based literature mining, knowledge acquisition, knowledge integration, and knowledge retrieval are combined using XML-based tag information and ontology management. The system supports combining different types of databases (papers and patents, technologies and innovations) and retrieving different types of knowledge simultaneously. The main objective of the system is to facilitate knowledge acquisition and knowledge retrieval from documents through an ontology-based dynamic similarity calculation and a visualization of automatically structured knowledge. Through experimentations we conducted using 100,000 words economic documents reported in the "Go！ Japan" project for analyzing Japanese industrial situation, and 100,000 words molecular biology Papers, we show the system is Practical enough for accelerating knowledge acquisition and knowledge discovery from the information sea.

• Annual Conference on Human and Language Technology
• /
• 2016.10a
• /
• pp.201-204
• /
• 2016

본 논문에서는 문서의 공기관계를 통해 추출된 문서의 특징을 이용하여 유사 보고서를 판별하는 시스템을 제안한다. 국가 R&D 보고서의 XML형식 파일에서 텍스트를 추출 후, 문장 단위로 나누어 각 문장의 공기 관계를 추출한다. 그 후 공기관계의 노드와 엣지를 문서에 추가하고, 노드로 사용된 단어만 남기고 나머지 단어는 제외한다. 그리고 이것을 문서의 특징으로 삼고 유사도 계산을 한다. 이 때, 유사도 계산은 코사인 유사도를 사용한다. 실험결과, 국가 R&D문서 유사도 계산에서 제안된 방법이 기존의 방법보다 높은 분류율을 보여주었다.

• Proceedings of the KSR Conference
• /
• 2006.11b
• /
• pp.387-391
• /
• 2006

Insight into behaviour of pedestrians as well as tools to assess passenger flow conditions are important in for instance planning and geometric design of railway station under regular and safety-critical circumstances. Algorithm for passenger flow analysis based on DEM(Discrete Element Method) is newly developed. There are lots of similarity between particle-laden two phase flow and passenger flow. The velocity component of 1st phase corresponds to the unit vector of calculation cell, each particle to passenger, volume fraction to population density and the particle velocity to the walking velocity, etc. And, the walking velocity of passenger is also represented by the function of population density. Key algorithms are developed to determine the position of passenger, population density and numbering to each passenger. To verify the effectiveness of new algorithm, passenger flow analysis for the basic models of railway station is conducted.

• Advances in nano research
• /
• v.1 no.3
• /
• pp.171-182
• /
• 2013

The alternating current (AC) conductivity in semiconductor crystals with an open-core screw dislocation is studied in the current work. The screw dislocation in crystalline media results in an effective potential field which affects the electronic transport properties of the system. Therefore, from a technological view point, it is interesting to investigate properties of AC conductivity at frequencies of a few terahertz. To quantify the screw-induced potential effect, we calculated the AC conductivity of dislocated crystals using the Kubo formula. The conductivity showed peaks within the terahertz frequency region, where the amplitude of the AC conductivity was large enough to be measured in experiments. The measurable conductivity peaks did not arise in dislocation-free crystals threaded by a magnetic flux tube. These results imply different conductivity mechanisms in crystals with a screw dislocation than those threaded by a magnetic flux tube, despite the apparent similarity in their electronic eigenstates.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2006.11a
• /
• pp.325-328
• /
• 2006

최근 DTD (Document Type Descriptor)를 포함하고 있지 않은 XML 문서의 사용이 증가하고 있다. 따라서 서로 다른 구조를 갖는 많은 양의 XML 문서를 관계형 DBMS에 저장하거나, 인덱스를 이용하여 매핑하는 등 보다 효율적으로 관리하기 위한 다양한 인덱싱 기법에 대한 연구가 진행되고 있다. 이러한 연구들 중 경로 비트맵 인덱싱 기법은 경로 구성 유사도를 기반으로 3차원 비트맵 클러스터를 생성하고, 클러스터 단위의 검색을 수행함으로서 빠른 검색 속도를 보여주었다. 그러나 이 기법은 비교하려는 두 경로 중 항상 짧은 경로가 기준 경로가 되는 한계점과, 같은 노드 구성을 가지는 두 경로에서도 노드의 위치에 따라 그 유사도가 크게 변하는 등의 여러 문제점을 가지고 있었다. 이러한 문제점을 해결하고, 정확한 클러스터링을 수행하기 위해서는 합리적인 경로 유사도 계산식이 필요하게 되었다. 본 논문에서는 기존 방법의 문제점을 해결하고, 보다 정확한 클러스터링을 수행할 수 있는 새로운 경로 유사도 계산식을 제안한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.10a
• /
• pp.1144-1145
• /
• 2015

협업 필터링 기반 추천 시스템에서는 이웃 사용자를 정확하게 찾는 것이 추천 정확도에 핵심적인 영향을 미친다. 그러나 기존의 유사도 척도는 사용자가 공통으로 평가한 아이템만을 고려하여 유사도를 계산하기 때문에 이러한 아이템이 적은 사용자 간의 유사도가 부정확하게 계산되는 문제가 있다. 본 논문에서는 이러한 문제를 극복하기 위해 공통으로 평가하지 않은 아이템을 함께 고려하여 유사도를 계산하는 방안을 제안한다. 또한, 실험을 통해 제안하는 방안이 협업 필터링 기반 추천 시스템의 정확도 향상에 기여함을 보인다.

• Korean Journal of Remote Sensing
• /
• v.3 no.1
• /
• pp.25-39
• /
• 1987

The purpose of this research is to develop the Automatic Fingerprint Verfication System by digital computer based on specially in PC level. Fingerprint is used as means of personal identity verification in view of that it has the high reliability and safety. Fingerprint pattern recognition algorithm is constitute of 3 stages, namely of the preprocessing, the feature extraction and the recognition. The preprocessing stage includes smoothing, binarization, thinning and restoration. The feature extraction stage includes the extraction of minutiae and its features. The recognition stage includes the registration and the matching score calculation which measures the similarity between two images. Tests for this study with 325 pairs of fingerprint resulted in 100% of separation which which in turn is turned out to be the reliability of this algorithm.

• Journal of Information Processing Systems
• /
• v.15 no.4
• /
• pp.785-796
• /
• 2019

Due to the view point, illumination, personal gait and other background situation, person re-identification across cameras has been a challenging task in video surveillance area. In order to address the problem, a novel method called Joint Bayesian across different cameras for person re-identification (JBR) is proposed. Motivated by the superior measurement ability of Joint Bayesian, a set of Joint Bayesian matrices is obtained by learning with different camera pairs. With the global Joint Bayesian matrix, the proposed method combines the characteristics of multi-camera shooting and person re-identification. Then this method can improve the calculation precision of the similarity between two individuals by learning the transition between two cameras. For investigating the proposed method, it is implemented on two compare large-scale re-ID datasets, the Market-1501 and DukeMTMC-reID. The RANK-1 accuracy significantly increases about 3% and 4%, and the maximum a posterior (MAP) improves about 1% and 4%, respectively.