• Journal of applied mathematics & informatics
• /
• 제8권2호
• /
• pp.641-649
• /
• 2001

A procedure of detecting change-points is considered with split linear fitting idea from Hall and Titterington(1992). At each given point, left, central and right linear fits are compared to detect the discontinuities or change-points. A simulation study is done with various types of change models and shows that the suggested technique can be a flexible data-analytic tool.

• 음성과학
• /
• 제8권4호
• /
• pp.59-72
• /
• 2001

In this paper, we experimented with speaker change detection that uses a statistical method for NOD (News On Demand) service. A specified speaker's change can find out content of each data in speech if analysed because it means change of data contents in news data. Speaker change detection acts as preprocessor that divide input speech by speaker. This is an important preprocessor phase for speaker tracking. We detected speaker change using GLR(generalized likelihood ratio) distance base division and BIC (Bayesian information criterion) base division among matrix method. An experiment verified speaker change point using BIC base division after divide by speaker unit using GLR distance base method first. In the experimental result, FAR (False Alarm Rate) was 63.29 in high noise environment and FAR was 54.28 in low noise environment in MDR (Missed Detection Rate) 15% neighborhood.

• 대한음성학회지:말소리
• /
• 제47호
• /
• pp.97-107
• /
• 2003

Speaker change detection is to detect automatically a point of time at which speaker was replaced. Since feature parameters used for speaker change detection depend not only on speaker characteristics but also on phonetic characteristics, spoken contents included in the feature parameters inevitably causes performance degradation of speaker change detection. In this paper, to alleviate this problem, a method to normalize phonetic variations in speech feature parameters is proposed for emphasizing changes due to speaker characteristics. Experimental results show that the proposed method improves the performance of speaker change detection.

• Journal of the Korean Data and Information Science Society
• /
• 제16권2호
• /
• pp.173-183
• /
• 2005

This paper studies quick detection of variance change point for iid data. For development of sensitive and adaptive variance change point detector, moving variance ratio is employed as a variance ratio estimator. It is shown that selection of tuning parameters of detector, (i.e., information and lag tuning parameters) is critical for detector to achieve desirable sensitivity and adaptiveness. Interestingly our simulation result reveals limitations of the commonly used change ratio against the previous day. Our results will provide useful insight when the detector is applied to time series data.

• Journal of the Korean Data and Information Science Society
• /
• 제16권2호
• /
• pp.289-300
• /
• 2005

In this article quick variance change point (VCP) detection problem for time series is considered. For this variance VCP detector equipped with tuning parameters is proposed. A major tool for the detector is moving variance ratio (MVR) which monitors variance change of a given time series. Tuning process of detector is investigated via simulation, which shows that tuning parameters are critical in achieving sensitivity and adaptiveness of detector.

• 응용통계연구
• /
• 제26권2호
• /
• pp.291-305
• /
• 2013

네트워크상에서 분산 서비스 거부(DDoS) 공격 탐지를 위해 수집되는 트래픽 자료(BPS, PPS 등)는 시간 순서대로 발생하는 대용량 자료이다. 대용량 자료에서 공격 탐지를 위한 변화점 탐지 알고리즘은 정확성 뿐 아니라 시간과 공간적인 계산 수행의 효율성이 확보되어야 한다. 본 연구에서는 대용량자료에서 변화점 탐지에 대한 Ross 등(2011)이 연구한 순차적인 Sliding Window and Discretization(SWD) 방법을 확장하였다. 그리고 경험적 분포함수와 순위를 이용한 다섯 종류의 검정방법을 사용하면서 자료의 분포에 대한 가정없이 DDoS 공격을 탐지할 수 있도록 새로운 비모수 모형을 제안한다. 다양한 확률밀도 함수와 이에 대응하는 모평균과 분산을 변화시키면서 모의실험하여 본 연구에서 제안한 비모수적 검정방법을 SWD 방법에 적용하여 모형의 효율성을 탐색하고 토론한다. 그리고 실증 분석을 통해 공격 탐지율 및 공격 탐지의 정확성을 기준으로 성능을 측정하고 비교하였다.

• International Journal of Quality Innovation
• /
• 제2권1호
• /
• pp.30-49
• /
• 2001

This paper considers a recent environment in the manufacturing process in which data in large amounts can be obtained on-line in real-time. Under this environment an on-line real-time Statistical Process Control (SPC) scheme equipped with detection of a process change, change-point estimation, and recognition of the change pattern is proposed. The proposed SPC scheme is composed of a Cusum chart, filtering methods and Akaike Information Criterion (AIC). We examine the performance of this scheme by Monte Carlo simulation and show its usefulness.

• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 2006년도 춘계공동학술대회 논문집
• /
• pp.1214-1220
• /
• 2006

In this article, it will be shown that a nonparametric and data-adaptive approach to the variance change point (VCP) detection problem is possible by formulating it as a pattern classification problem. Technical aspects of the VCP detector are discussed, which include its training strategy and selection of proper classification tool.

• 품질경영학회지
• /
• 제35권4호
• /
• pp.159-165
• /
• 2007

Using statistical methods a change-point analysis of oil supply disruption is conducted. The statistical distribution of oil supply disruption is a weibull distribution. The detection of the change-point is applied to Bayesian method and weibull parameters are estimated through Markov chain monte carlo and parameter approach. The statistical approaches to the estimation for the change-point and weibull parameters is implemented with the sets of simulated and real data with small sizes of samples.

• 한국컴퓨터정보학회논문지
• /
• 제4권3호
• /
• pp.28-34
• /
• 1999

본 논문에서는 DCT 영역에서 DCT블록내의 DC성분과 매크로블록내의 움직임 벡터성분을 이용하여 동영상 장면전환 자동 검출에 관한 방법을 제안한다. 제안하는 방법은 압축된 MPEG 신호열내의 픽쳐 종류와 상관없이 임의의 신호열에서 장면전환을 검출한다. 이를 위해 MPEG 신호열의 DCT 영역에서 픽쳐구성요소의 특징을 이용하여 새로운 비교잣대를 정의하고 움직임 벡터에 예민한 B-picture 의 특성을 이용하여 정확한 장면전환을 검출한다. 이때. I-picture를 가지고 급격한(cut) 장면전환을 검출하고 P. B picture 를 가지고 점진적인(gradual) 장면 전환을 검출한다. 제안한 방법은 다양한 MPEG 신호열에 대하여 시험결과 그 효율성이 우수함을 보인다.