• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.2
• /
• pp.206-211
• /
• 2011

In this paper, the evolutionary algorithm was specifically formulated for optimization with continuous parameter space. The proposal was motivated by the fact that the genetic algorithms have been most intensively reported for parameter identification problems with continuous search space. The difference of primary characteristics between genetic algorithms and the proposed algorithm, discrete or continuous individual representation has made different areas to which the algorithms should be applied. Results obtained by optimization of some well-known test functions indicate that the proposed algorithm is superior to genetic algorithms in all the performance, computation time and memory usage for continuous search space problems.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.10b
• /
• pp.226-228
• /
• 2004

최근에 이동객체의 위치정보를 활용한 위치기반서비스(L8S, Location Based Services)에 대한 관심이 증가하고 있다. 전통적으로 정적인 위치정보를 갖는 공간 객체는 GIS(Geographic Information System) 서버에 저장, 관리되었다. 이동객체는 시간에 따라 위치의 변화가 매우 빈번하여 위치 정보가 계속 갱신되기 때문에, 전통적인 GIS 서버로는 관리가 어렵다. 본 논문에서는 기존의 연속적인 최근접 이웃탐색 기법에서 데이터의 처리 순서에 따라 탐색공간과 계산비용이 증가하는 문제점을 슬랩을 사용하여 해결한다. 최근접 이웃의 수직연장선 사이의 공간인 슬랩 내부영역에 대해서만 탐색하도록 하여 탐색영역을 줄이고, 그 내부에 있는 점들에 대해서만 처리하여 계산비용을 줄인다.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.06c
• /
• pp.75-78
• /
• 1998

이 논문에서는 대용량 연속음성 인식에서 인식 속도를 향상시키기 위한 방법들에 대해서 연구하였다. 음성인식에 있어서 많은 양의 계산을 요하는 부분은 관측 확률의 계산과 탐색에 필요한 계산이다. 탐색에 필요한 계산을 줄이기 위하여 빔 탐색법과 phoneme look-ahead기법을 통해 탐색 공간을 줄였으며, 관측 확률을 계산하는데 소요되는 시간을 줄이기 위하여 입력 특징 벡터와 이웃 관계에 있는 가우시안 성분들만 정확한 계산을 하는 VQ에 의한 계산량 감축 방법과 tree-structured pdf 방법을 구현하였다. 3천개의 어휘와 2천여개의 트라이폰 모델로 구성된 연속 음성인식 시스템에서 보통의 Viterbi 빔 탐색법을 적용한 경우에 실시간의 2.73배의 인식 속도로 93.39%의 단어 인식률을 얻을 수 있는데 phoneme look-ahead 기법과 tree-structured pdf 방법을 추가 적용함으로써 비슷한 인식 성능에서 1.55배의 인식 속도를 얻을 수 있었다.

• The Journal of the Acoustical Society of Korea
• /
• v.21 no.6
• /
• pp.543-551
• /
• 2002

In this paper, we present an efficient decoding method that performs in real time for 20k word continuous speech recognition task. Basic search method is a one-pass Viterbi decoder on the search space constructed from the novel language model network. With the consistent search space representation derived from various language models by the LM network, we incorporate basic pruning strategies, from which tokens alive constitute a dynamic search space. To facilitate post-processing, it produces a word graph and a N-best list subsequently. The decoder is tested on the database of 20k words and evaluated with respect to accuracy and RTF.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2010.05a
• /
• pp.927-930
• /
• 2010

Recently, continuous k-nearest neighbor query(CkNN) which is defined as a query to find the nearest points of interest to all the points on a given path is widely used in the LBS(Location Based Service) and ITS(Intelligent Transportation System) applications. It is necessary to acquire results quickly in the above applications and be applicable to spatial network databases. This paper proposes a new method to search nearest POIs(Point Of Interest) for moving query objects on the spatial networks. The method produces a set of split points and their corresponding k-POIs as results. There is no order between the POIs. The analysis show that the proposed method outperforms the existing methods.

• Journal of the Korean Geographical Society
• /
• v.42 no.4
• /
• pp.632-647
• /
• 2007

Lots of geographic phenomena have dynamic spatial patterns with time changes, and there have been lots of researches on exploring these dynamic spatial patterns. However, most of these researches focused on the static pattern analysis in a given period, rather than dealing with dynamic changes in the spatial pattern over time with the continual or cumulative perspective. For this reason, investigation of the inertia of spatial process in terms of temporal changes is needed. From this background, the purpose of this paper is to propose the methodology to explore the changes in spatial pattern cumulatively by considering the inertia of the spatial statistics over time, and to apply it to the case study That is, we introduce the new spatial statistic, and produce the z-values of the statistic using Monte Carlo Simulation, and then to explore the changes in spatial patterns over time cumulatively. To do this, the method to combine the J statistic with CUSUM statistic for exploring spatial patterns, and to apply it to the changes in the commercial landuse in Erie County, New York State. Through the proposed method for spatio-temporal Patterns, we could explore continual changes effectively in the spatial patterns reflecting the statistics by temporal spot cumulatively.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.36 no.2
• /
• pp.67-74
• /
• 2018

This study has proposed a methodology to detect suspicious facilities that occupy national and public land by using the cadastral and digital maps. First, we constructed a spatial database of national & public land based on the cadastral maps by linking its management ledger. Using the PNU (Parcel Number) code as a key field, the data managed by different institutions are integrated into a single spatial information DB (database) and then, the use or nonuse state of each parcel is confirmed on the cadastral map. Next, we explored the suspicious facilities that existed in the unused parcel by utilizing the digital topographical map. Then, the proposed methodology was applied for various regions and tested its feasibility. Through this study, it will be possible to improve the utilization of digital maps and to manage the national and public land efficiently and economically.

• Journal of Advanced Marine Engineering and Technology
• /
• v.21 no.2
• /
• pp.108-116
• /
• 1997

유전알고리즘은 진화원리에서 발견된 몇몇 특징들을 컴퓨터 알고리즘과 결합시켜 복잡한 최적화 문제를 해결하려는 도구로서 1975년 미국의 Holland 교수에 의해 처음으로 개발되었다. 주어진 문제에서 탐색환경이 다변수 또는 다봉(multi-modal)이 되어 대단히 복잡하거나 또는 부분적으로 알려질 경우는, 구배(gradient)에 기초한 재래식 방법을 사용하여 최적화하는 것은 매우 어렵게 되고 경우에 따라서는 불가능할 수도 있다. 이러한 이유로 유전알고리즘과 같은 강인한 탐색법이 요구된다. 유전알고리즘의 장점은 연속성(continuity), 미분가능성(differentiability), 단봉성(unimodality) 등과 같이 탐색공간에 대한 제약으로부터 자유롭다는 것이다. 다시 말하면 목적함수 외 탐색공간에 대한 사전지식을 필요로 하지 않고, 매우 크고 복잡한 공간일지라도 전역해 쪽으로 수렴해 갈수 있다는 것이다. 이러한 특성 때문에 유전알고리즘은 실제 환경에서 많은 복잡한 최적화 문제를 해결하는 방법으로 인정을 받고 있으며, 함수의 최적화, 신경회로망의 학습, 동적시스템의 식별및 제어, 신호처리등 여러 분야에 성공적으로 응용되고 있다. 이러한 중요성에 비해 유전알고리즘에 대한 연구는 국내적으로는 아직 미진한 수준이나 최근 이에 대한 관심이 고조되고 있으며, 또한 그 응용분야도 점점 넓어져 이론 개발과 실질적인 응용에 확산되리라 생각된다. 따라서 본 해설기사는 유전알고리즘의 원리와 응용 사례를 살펴봄으로서 최적화 문제를 해결하려는 독자들에게 조금이나마 도움을 주고자 한다.

• The Journal of the Korea Contents Association
• /
• v.20 no.11
• /
• pp.636-643
• /
• 2020

The differential evolution algorithm is one of the meta-heuristic techniques developed to solve the real optimization problem, which is a continuous problem space. In this study, in order to use the differential evolution algorithm to solve the traveling salesman problem, which is a discontinuous problem space, a random key representation method is applied to the differential evolution algorithm. The differential evolution algorithm searches for a real space and uses the order of the indexes of the solutions sorted in ascending order as the order of city visits to find the fitness. As a result of experimentation by applying it to the benchmark traveling salesman problems which are provided in TSPLIB, it was confirmed that the proposed differential evolution algorithm based on the random key representation method has the potential to solve the traveling salesman problems.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.12 no.5
• /
• pp.2383-2389
• /
• 2011

Ubiquitous applications require hybrid continuous query processing which processes both on-line data stream and spatial data in the disk. In the hybrid continuous spatial query processing, disk access costs for the high-volume spatial data should be minimized. However, previous indexing methods cannot reduce the disk seek time, because it is difficult that the data are stored in contiguity with others. Also, existing methods for the space-filling curve considering data cluster have the problem which does not cluster available data for queries. Therefore, we propose the region storage structure for efficient data access in hybrid continues spatial query processing. This paper shows that there is an obvious improvement of query processing costs through the contiguous data storing method and the group processing for user queries based on the region storage structure.