지원사업
학술연구/단체지원/교육 등 연구자 활동을 지속하도록 DBpia가 지원하고 있어요.
커뮤니티
연구자들이 자신의 연구와 전문성을 널리 알리고, 새로운 협력의 기회를 만들 수 있는 네트워킹 공간이에요.
논문 기본 정보
- 자료유형
- 학술저널
- 저자정보
- 발행연도
- 2025.3
- 수록면
- 80 - 93 (14page)
- DOI
- 10.33162/JAR.2025.3.25.1.080
이용수
초록· 키워드
오류제보하기
Purpose: Traditional change-point detection algorithms often overlook trends and seasonality, particularly in multivariate time series. This paper proposes a method that addresses this limitation by integrating trend and seasonality decomposition into change-point detection.
Methods: The proposed algorithm is based on N-BEATS, a deep learning model for time-series analysis. The algorithm first decomposes each time series into trend and seasonal components. It separately detects change points for these components and then employs a frequency-based approach to determine the final change points via integration across multiple time series.
Results: The proposed algorithm effectively performs trend and seasonality decomposition while accurately identifying diverse change points in multivariate time-series data. It outperforms existing methods, particularly in detecting subtle changes that traditional statistical and neural network-based models often fail to capture.
Conclusion: The proposed N-BEATS-based change point detection algorithm effectively detects change points in multivariate time series data, while considering trends and seasonality.
Methods: The proposed algorithm is based on N-BEATS, a deep learning model for time-series analysis. The algorithm first decomposes each time series into trend and seasonal components. It separately detects change points for these components and then employs a frequency-based approach to determine the final change points via integration across multiple time series.
Results: The proposed algorithm effectively performs trend and seasonality decomposition while accurately identifying diverse change points in multivariate time-series data. It outperforms existing methods, particularly in detecting subtle changes that traditional statistical and neural network-based models often fail to capture.
Conclusion: The proposed N-BEATS-based change point detection algorithm effectively detects change points in multivariate time series data, while considering trends and seasonality.
목차
1. 서론
2. 관련 연구
3. N-BEATS를 이용한 시계열 변화점 탐지
4. 실험 및 결과
5. 결론 및 향후 과제
References
참고문헌 (0)
참고문헌 신청함께 읽어보면 좋을 논문
논문 유사도에 따라 DBpia 가 추천하는 논문입니다. 함께 보면 좋을 연관 논문을 확인해보세요!
-
Development of a deep learning-based anomaly detection model using multivariate time series manufacturing data
대한산업공학회 추계학술대회 논문집
2022 .11
-
Contrastive learning for time series anomaly detection using decomposition methods
대한산업공학회 춘계공동학술대회 논문집
2023 .05
-
A Survey on Time-Series Anomaly Detection focusing on Features, Approaches, and Challenging
대한전자공학회 학술대회
2023 .06
-
주요 악기가 다른 음악에서 에너지 기반과 위상 기반의 Beat Tracking 알고리즘 성능 비교 및 분석
대한전자공학회 학술대회
2017 .06
-
Multiple Encoder based Memory Network for Multivariate Time-series Anomaly Detection
대한산업공학회 추계학술대회 논문집
2022 .11
-
A Comparative Study on Anomaly Detection in Multivariate Time Series Data with Correlation Dependencies
대한산업공학회 추계학술대회 논문집
2021 .11
-
비정상심박 검출을 위해 영상화된 심전도 신호를 이용한 비교학습 기반 딥러닝 알고리즘
한국정보통신학회논문지
2022 .01
-
시계열 분해와 GRU를 결합한 시계열 예측 방법 연구
Proceedings of KIIT Conference
2023 .11
-
Shallow CNN을 활용한 주가 예측 방법론
한국통신학회논문지
2023 .06
-
Anomaly Detection in Multivariate Time Series for Variable Refrigerant Flow System
대한산업공학회 춘계공동학술대회 논문집
2022 .06
-
다변량 시계열 데이터 분류를 위한 특징 선택 알고리즘
대한산업공학회 춘계공동학술대회 논문집
2017 .04
-
움직임 정보와 칼라정보 분석을 통한 화재검출 알고리즘
멀티미디어학회논문지
2016 .02
-
Target BPM을 이용한 Beat Tracking 알고리즘 구현
한국통신학회 학술대회논문집
2016 .11
-
Self-Supervised Anomaly Detection using Outliers for Multivariate Time Series
대한산업공학회 추계학술대회 논문집
2024 .10
-
Review of Data-Driven Multivariate and Multiscale Methods
IEIE Transactions on Smart Processing & Computing
2015 .04
-
Beat 주파수 성분을 이용한 병렬 DC-DC 컨버터의 온라인 커패시턴스 추정 방법
전력전자학회논문지
2024 .08
-
An optimal window time length for starting points detection of long/short-term tasks in multivariate time series
대한전자공학회 학술대회
2024 .06
-
손목형 혈압계 방식의 Beat-by-Beat 혈압 측정에 관한 실험적 연구
전기학회논문지
2024 .09
-
Design of an Enhanced Object Detection Algorithm through Image Scaling
한국통신학회 학술대회논문집
2021 .06
-
A Survey on Deep Learning-based Anomaly Detection Models for Time Series Data
한국정보과학회 학술발표논문집
2019 .06
이 논문의 저자 정보
최근 본 자료
댓글(0)
0
UCI(KEPA) : I410-151-25-02-092513971