• Journal of KIISE:Databases
• /
• v.27 no.4
• /
• pp.651-662
• /
• 2000

멀티미디어 데이터베이스에서 k-최근접 질의는 가장 일반적이며, 비공간 검색 조건이 포함된 경우가 많다. 현재까지 이러한 질의를 위한 여러 기법 중에서 Hjaltason과 Samet이 제안한 점증적 최근접 알고리즘에 가장 유용하다고 알려져 있다. 질의 처리를 위해 상위 연산자가 k보다 많은 객체를 요구할 때, 이 알고리즘은 처음부터 질의를 재실행하지 않고 다음 객체를 전달할 수 있기 때문이다. 그런데, 이 알고리즘에서 사용하는 R-트리는 결국에는 비공간 검색조건을 만족시키지 않을 투플 후보들을 부분적으로 제거할 수가 없기 때문에 비효율적이다. 본 논문에서 우리는 이 알고리즘을 보완한 RS-트리 기반 점증적 최근접 알고리즘을 제안한다. RS-트리는 R-트리와, 그 보조 트리로서 계층적 시스니쳐 파일을 기반으로 하는 S-트리로 구성된다. S-트리는 R-트리를 탐색하는 과정에서 많은 불필요한 투플을 제거하는 역할을 수행한다. 본 논문에서는 실험을 통해 RS-트리가 Hjaltason과 Samet의 알고리즘의 성능을 향상시킬 수 있음을 보인다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06c
• /
• pp.376-380
• /
• 2008

사회의 복잡화와 인터넷의 성장으로 폭발적으로 늘어나고 있는 정보들을 사용자가 모두 검토한 후 여과하기는 어려운 일이다. 이러한 문제를 보완하기 위해서 자동화된 정보 여과 기술이 사용되는데, k-최근접 이웃(k-nearest neighbor) 알고리즘은 그 구현이 간단하며 비교적 정확하여 가장 널리 쓰이고 있는 알고리즘 중 하나이다. k 개의 최근접 이웃들로부터 평가값을 계산하는 데 흔히 쓰이는 방법은 상관계수를 이용한 가중치에 기반하는 것이다. 본 논문에서는 이를 보완하여 대규모 데이터에 대해서도 속도는 크게 저하되지 않으며 정확도는 대폭 향상시킬 수 있는 방법을 적용하였다. 또한, 최근접 이웃을 구하는 거리함수로 다양한 방법을 시도하였다. 영화추천을 위한 실제 데이터에 대한 실험 결과, 속도의 저하는 미미하였으나 정확도에 있어서는 크게 향상된 결과를 가져올 수 있었다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.10a
• /
• pp.83-85
• /
• 2000

최근접 질의(nearest neighbor query)는 멀티미디어 데이터베이스에서 주어진 질의 객체와 가장 유사한 객체를 찾기 위한 매우 중요한 연산으로 사용된다. 대부분의 최근접 질의 처리 기법들은 객체의 효과적인 인덱싱을 위하여 다차원 인덱스(multidimensional index)를 사용한다. 그러나 N차원 시각형 혹은 원을 사용하여 객체 클러스터의 캡슐을 표현하는 기존의 다차원 인덱스들은 차원 수가 높아짐에 따라 검색 성능이 크게 떨어진다. 본 논문에서는 이러한 문제를 해결하는 새로운 인덱스 구조를 제시하고, 이를 이용하는 최근접 질의 처리 방안을 제안한다. 또한, 다양한 실험에 의한 성능 평가를 통하여 제안된 기법의 우수성을 검증한다.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.1
• /
• pp.11-20
• /
• 2008

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.10
• /
• pp.73-81
• /
• 2015

The paper proposes a prototype selection method and evaluates the generalization performance of standard algorithms and prototype based classification learning. The proposed prototype classifier defines multidimensional spheres with variable radii within class areas and generates a small set of training data. The nearest-neighbor classifier uses the new training set for predicting the class of test data. By decomposing bias and variance of the mean expected error value, we compare the generalization errors of k-nearest neighbor, Bayesian classifier, prototype selection using fixed radius and the proposed prototype selection method. In experiments, the bias-variance changing trends of the proposed prototype classifier are similar to those of nearest neighbor classifiers with all training data and the prototype selection rates are under 27.0% on average.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.9 no.1
• /
• pp.1-8
• /
• 2010

For three and more rapidly moving vehicles, they want to search the nearest location for meeting. Each vehicle has a different velocity and a efficient method is needed for shifting a short distance. It is observed that the existing group nearest-neighbor query has been investigated for static query points; however these studies do not extend to highly dynamic vehicle environments. In this paper, we propose a novel Dynamic Nearest-Neighbor query processing for Multiple Vehicles (DNN_MV). Our method retrieves the nearest neighbor for a group of moving query points with a given vector and takes the direction of moving query points with a given vector into consideration for DNN_MV. Our method efficiently calculates a group nearest neighbor through a centroid point that represents the group of moving query points. The experimental results show that the proposed method operates efficiently in a dynamic group nearest neighbor search.

• The Korean Journal of Applied Statistics
• /
• v.34 no.3
• /
• pp.401-410
• /
• 2021

Popular in discriminant classification analysis, k-nearest neighbor classification methods have limitations that do not reflect the local characteristic of the data, considering only the number of fixed neighbors. Considering the local structure of the data, the adaptive nearest neighbor method has been developed to select the number of neighbors. In the analysis of high-dimensional data, it is common to perform dimension reduction such as random projection techniques before using k-nearest neighbor classification. Recently, an ensemble technique has been developed that carefully combines the results of such random classifiers and makes final assignments by voting. In this paper, we propose a novel discriminant classification technique that combines adaptive nearest neighbor methods with random projection ensemble techniques for analysis on high-dimensional data. Through simulation and real-world data analyses, we confirm that the proposed method outperforms in terms of classification accuracy compared to the previously developed methods.

• Journal of Korea Spatial Information System Society
• /
• v.9 no.2
• /
• pp.67-79
• /
• 2007

Recently, to support LBS(location-based services) and telematics applications efficiently, there have been many researches which consider the spatial network instead of Euclidean space. However, existing range query and k-nearest neighbor query algorithms show a linear decrease in performance as the value of radius and k is increased. In this paper, to increase the performance of query processing algorithm, we propose materialization-based range and k-nearest neighbor algorithms. In addition, we make the performance comparison to show the proposed algorithm achieves better retrieval performance than the existing algorithm.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.5C
• /
• pp.305-312
• /
• 2005

Three classification algorithms are tested using musical instruments. Several classification algorithms are introduced and among them, Bayes rule, NN and k-NN performances evaluated. ZCR, mean, variance and average peak level feature vectors are extracted from instruments sample file and used as data set to classification system. Used musical instruments are Violin, baroque violin and baroque cello. Results of experiment show that the performance of NN algorithm excels other algorithms in musical instruments classification.

• The Journal of the Korea Contents Association
• /
• v.14 no.10
• /
• pp.454-462
• /
• 2014

With the development of location aware technologies and mobile devices, location-based services have been studied. To provide location-based services, many researchers proposed methods for processing various query types with Mapreduce(MR). One of the proposed methods, is a Reverse k-nearest neighbor(RkNN) query processing method with MR. However, the existing methods spend too much cost to process the continuous RkNN query. In this paper, we propose an efficient continuous RkNN query processing method with MR to resolve the problems of the existing methods. The proposed method uses the 60-degree-pruning method. The proposed method does not need to reprocess the query for continuous query processing because the proposed method draws and monitors the monitoring area including the candidate objects of a RkNN query. In order to show the superiority of the proposed method, we compare it with the query processing performance of the existing method.