• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07b
• /
• pp.157-159
• /
• 2005

본 논문에서는 단일 색인을 사용하는 정규화 변환 지원 서브시퀀스 매칭 방법을 제안한다. 기존의 정규화 변환 지원 서브시퀀스 매칭 방법은 질의 시퀀스 길이가 커질수록 성능이 저하되고, 이를 해결하기 위하여 여러개의 색인을 사용하는 방법을 취하였다. 본 논문에서는 하나의 색인을 사용하면서도 다양한 길이의 정규화 변환 지원 서브시퀀스 매칭을 수행하는 효율적인 방법을 제시한다. 이를 위하여, 본 논문에서는 정규화 변환의 정의를 확장하여 일반화 정규화 변환 개념을 제시한다. 또한, 이러한 일반화 정규화 변환 개념을 기존 서브시퀀스 매칭 방법들에 적용하는 방안에 대한 이론적 근거를 각각의 정리로서 제시하고 증명하였다. 그리고, 이들 방안을 구현하기 위한 색인 구성 알고리즘 및 서브시퀀스 매칭 알고리즘을 각각 제시하였다. 본 논문에서 제안한 정규화 변환 지원 서브시퀀스 매칭은 다른 변환을 지원하는 서브시퀀스 매칭으로 일반화 될 수 있는 우수한 연구결과라 사료된다.

• The KIPS Transactions:PartD
• /
• v.9D no.4
• /
• pp.555-560
• /
• 2002

Subsequence matching, which consists of index searching and post-processing steps, is an operation that finds those subsequences whose changing patterns are similar to that of a given query sequence from a time-series database. This paper discusses optimization of post-processing for subsequence matching. The common problem occurred in post-processing of previous methods is to compare the candidate subsequence with the query sequence for discarding false alarms whenever each candidate subsequence appears during index searching. This makes a sequence containing candidate subsequences to be accessed multiple times from disk, and also have a candidate subsequence to be compared with the query sequence multiple times. These redundancies cause the performance of subsequence matching to degrade seriously. In this paper, we propose a new optimal method for resolving the problem. The proposed method stores ail the candidate subsequences returned by index searching into a binary search tree, and performs post-processing in a batch fashion after finishing the index searching. By this method, we are able to completely eliminate the redundancies mentioned above. For verifying the performance improvement effect of the proposed method, we perform extensive experiments using a real-life stock data set. The results reveal that the proposed method achieves 55 times to 156 times speedup over the previous methods.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.11a
• /
• pp.444-447
• /
• 2000

본 논문에서는 종래의 정적인 디지털 패턴 매칭을 행하는 연상메모리와는 달리 아놀로그의 시계열정보를 직접 처리하여 시간축 방향으로 설정하는 것으로 강인성이 뛰어난 연상기억시스템을 제안하였다. 시스템의 기본적인 능력을 조사하기 위하여 기억패턴을 주기계열로 그리고 하중은 전부 고정하는 조건으로 단순화하여 시뮬레이션을 행하여 오류 정정능력을 갖는 것을 확인하였다. 시간축 방향의 하중을 적절하게 설정하면 기억용량의 증대나 상기 오류의 저감 등의 효과가 기대된다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10a
• /
• pp.184-186
• /
• 2001

본 논문에서는 시계열 데이터베이스에서 서브시퀸스 매칭을 효과적으로 처리하는 방안에 관하여 논의한다. 본 논문에서는 먼저, 기존의 이원성 기반 서브시퀸스 매칭 기법에서 발생하는 성능상의 문제점들을 지적하고, 이들을 해결할 수 있는 방법을 제시한다. 제안된 기법은 서브시퀸스 매칭 시 요구되는 인덱스 검색을 윈도우-조인이라는 일종의 공간 조인 문제로 새롭게 해석하는 것에서 출발한다. 제안된 기법에서는 효과적인 윈도우-조인의 처리를 위하여 질의 윈도우 점들을 위한 R＊-트리를 주기억장치 내에 on-the-fly로 구성하는 방법을 사용한다. 또한, 데이터 윈도우 점들을 위한 디스크 상의 R*-트리와 질 윈도우 점들을 위한 주기억장치 상의 R＊-트리를 효과적으로 조인할 수 있는 새로운 알고리즘을 제안한다. 제안된 기법은 R＊-트리 페이지들을 착오 채택 없이 단 한번만 디스크로부터 액세스 측면에서 이원성 기반 서브시퀸스 매칭을 위한 최적의 기법이다.

• Journal of KIISE:Databases
• /
• v.33 no.1
• /
• pp.42-55
• /
• 2006

We propose a single Index approach for subsequence matching that supports moving average transform of arbitrary order in time-series databases. Using the single index approach, we can reduce both storage space overhead and index maintenance overhead. Moving average transform is known to reduce the effect of noise and has been used in many areas such as econometrics since it is useful in finding overall trends. However, the previous research results have a problem of occurring index overhead both in storage space and in update maintenance since tile methods build several indexes to support arbitrary orders. In this paper, we first propose the concept of poly-order moving average transform, which uses a set of order values rather than one order value, by extending the original definition of moving average transform. That is, the poly-order transform makes a set of transformed windows from each original window since it transforms each window not for just one order value but for a set of order values. We then present theorems to formally prove the correctness of the poly-order transform based subsequence matching methods. Moreover, we propose two different subsequence matching methods supporting moving average transform of arbitrary order by applying the poly-order transform to the previous subsequence matching methods. Experimental results show that, for all the cases, the proposed methods improve performance significantly over the sequential scan. For real stock data, the proposed methods improve average performance by 22.4${\~}$33.8 times over the sequential scan. And, when comparing with the cases of building each index for all moving average orders, the proposed methods reduce the storage space required for indexes significantly by sacrificing only a little performance degradation(when we use 7 orders, the methods reduce the space by up to 1/7.0 while the performance degradation is only $9\%{\~}42\%$ on the average). In addition to the superiority in performance, index space, and index maintenance, the proposed methods have an advantage of being generalized to many sorts of other transforms including moving average transform. Therefore, we believe that our work can be widely and practically used in many sort of transform based subsequence matching methods.

• The KIPS Transactions:PartD
• /
• v.11D no.7 s.96
• /
• pp.1385-1398
• /
• 2004

This paper discusses an effective processing of subsequence matching under time warping in time-series databases. Time warping is a trans-formation that enables finding of sequences with similar patterns even when they are of different lengths. Through a preliminary experiment, we first point out that the performance bottleneck of Naive-Scan, a basic method for processing of subsequence matching under time warping, is on the CPU processing step. Then, we propose a novel method that optimizes the CPU processing step of Naive-Scan. The proposed method maximizes the CPU performance by eliminating all the redundant calculations occurring in computing the time warping distance between the query sequence and data subsequences. We formally prove the proposed method does not incur false dismissals and also is the optimal one for processing Naive-Scan. Also, we discuss the we discuss to apply the proposed method to the post-processing step of LB-Scan and ST-Filter, the previous methods for processing of subsequence matching under time warping. Then, we quantitatively verify the performance improvement ef-fects obtained by the proposed method via extensive experiments. The result shows that the performance of all the three previous methods im-proves by employing the proposed method. Especially, Naive-Scan, which is known to show the worst performance, performs much better than LB-Scan as well as ST-Filter in all cases when it employs the proposed method for CPU processing. This result is so meaningful in that the performance inversion among Nive- Scan, LB-Scan, and ST-Filter has occurred by optimizing the CPU processing step, which is their perform-ance bottleneck.

• Journal of KIISE:Databases
• /
• v.34 no.5
• /
• pp.409-419
• /
• 2007

Database community focuses on the similar music retrieval systems for music database when a humming query is given. One of the approaches is converting the midi data to time series, building their indices and performing the similarity search on them. Queries based on humming can be transformed to time series by using the known pitch detection algorithms. The recently suggested algorithm, scaled and warped matching, is based on dynamic time warping and uniform scaling. This paper proposes Humming BIRD(Humming Based sImilaR mini music retrieval system) using sliding window and center-aligned scaled and warped matching. Center-aligned scaled and warped matching is a mixed distance measure of center-aligned uniform scaling and time warping. The newly proposed measure gives tighter lower bound than previous ones which results in reduced search space. The empirical results show the superiority of this algorithm comparing the pruning power while it returns the same results.

• Proceedings of the Korean Information Science Society Conference
• /
• 2010.06c
• /
• pp.49-52
• /
• 2010

본 논문에서는 윤곽선 이미지 매칭에서 회전-불변 거리를 계산하는 효율적 방법을 제안한다. 회전-불변 거리 계산은 이미지 시계열을 한 칸씩 회전하면서 매번 유클리디안 거리를 계산해야 하는 고비용의 연산이다. 본 논문에서는 엔빌로프 기반 하한을 사용하여 불필요한 회전-불변 거리 계산을 크게 줄이는 효율적인 해결책을 제시하다. 이를 위해, 먼저 질의 시퀀스 대상의 엔빌로프 작성과 이의 하한 개념을 제시한다.다음으로, 엔빌로프 기반 하한을 회전-불변 거리 계산에 사용하면 많은 수의 회전-불변 거리계산을 줄일 수 있음을 보인다. 실험 결과, 제안한 엔빌로프 기반 매칭 기법은 기존 기법에 비해 최대 수배에서 수십배까지 매칭 성능을 향상시킨 것으로 나타났다.

• The KIPS Transactions:PartD
• /
• v.12D no.3 s.99
• /
• pp.355-364
• /
• 2005

Previous researches on subsequence matching have been focused on how to make indexes in order to speed up the matching time, and do not take into account the effectiveness issues of subsequence matching methods. This paper considers the effectiveness of subsequence matching methods and proposes two metrics for effectiveness evaluations of subsequence matching algorithms. We have applied the proposed metrics to Korean stock data and five known matching algorithms. The analysis on the empirical data shows that two methods (i.e., the method supporting normalization, and the method supporting scaling and shifting) outperform the others in terms of the effectiveness of subsequence matching.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.5
• /
• pp.655-660
• /
• 2009

The context prediction algorithms are not suitable to provide real-time personalized service for users in context-awareness environment. The algorithms have problems like time delay in training data processing and the difficulties of implementation in real-time environment. In this paper, we propose a prediction algorithm with user modeling to shorten of processing time and to improve the prediction accuracy in the context prediction algorithm. The algorithm uses moving path of user contexts for context prediction and generates user model by time-series analysis of user's moving path. And that predicts the user context with the user model by sequence matching method. We compared our algorithms with the prediction algorithms by processing time and prediction accuracy. As the result, the prediction accuracy of our algorithm is similar to the prediction algorithms, and processing time is reduced by 40% in real time service environment.