Exploring distance measures for time series data: A comparative analysis

Time series similarity search is a method used to identify the identical pattern within two sets of time series data, finds widespread utility in clustering, anomaly detection, and forecasting. In real-world scenarios, vibration data are often vast, intricate, and noisy, with adjustments in time, am...

Full description

Bibliographic Details
Main Author: Lee, Jia Yee
Format: Final Year Project / Dissertation / Thesis
Published: 2024
Subjects:
Online Access:http://eprints.utar.edu.my/6490/
http://eprints.utar.edu.my/6490/1/LeeJiaYee_Full_Report_Finalised.pdf
_version_ 1848886694247399424
author Lee, Jia Yee
author_facet Lee, Jia Yee
author_sort Lee, Jia Yee
building UTAR Institutional Repository
collection Online Access
description Time series similarity search is a method used to identify the identical pattern within two sets of time series data, finds widespread utility in clustering, anomaly detection, and forecasting. In real-world scenarios, vibration data are often vast, intricate, and noisy, with adjustments in time, amplitude, and phase shifting direct influence on search outcomes. Through a systematic evaluation, various distance measurement methods including Euclidean distance, Dynamic Time Warping, Fast Fourier Transform, Symbolic Aggregate Approximation, and Matrix Profile are performed under diverse conditions such as frequency shifting, amplitude scaling, state change, and noise. The comparative study encompasses not only quantitative assessments of accuracy but also considerations of computational efficiency and robustness. The findings reveal Matrix Profile generally outperforms classic measures like Euclidean distance, Dynamic Time Warping, and Fast Fourier Transform in accuracy, but performs poorly compared to Symbolic Aggregate Approximation. While Matrix Profile exhibits shorter computational time than Symbolic Aggregate Approximation, it slightly extends beyond other classic measures. Thus, Matrix Profile presents competitive advantages among ii distance measurement methodologies. By providing a comprehensive examination of similarity measurement techniques, this study equips the idea for the strength and weaknesses of distance measures, providing valuable insight for decision-making in time series data mining activities.
first_indexed 2025-11-15T19:42:34Z
format Final Year Project / Dissertation / Thesis
id utar-6490
institution Universiti Tunku Abdul Rahman
institution_category Local University
last_indexed 2025-11-15T19:42:34Z
publishDate 2024
recordtype eprints
repository_type Digital Repository
spelling utar-64902024-10-25T00:43:33Z Exploring distance measures for time series data: A comparative analysis Lee, Jia Yee Q Science (General) QA75 Electronic computers. Computer science T Technology (General) Time series similarity search is a method used to identify the identical pattern within two sets of time series data, finds widespread utility in clustering, anomaly detection, and forecasting. In real-world scenarios, vibration data are often vast, intricate, and noisy, with adjustments in time, amplitude, and phase shifting direct influence on search outcomes. Through a systematic evaluation, various distance measurement methods including Euclidean distance, Dynamic Time Warping, Fast Fourier Transform, Symbolic Aggregate Approximation, and Matrix Profile are performed under diverse conditions such as frequency shifting, amplitude scaling, state change, and noise. The comparative study encompasses not only quantitative assessments of accuracy but also considerations of computational efficiency and robustness. The findings reveal Matrix Profile generally outperforms classic measures like Euclidean distance, Dynamic Time Warping, and Fast Fourier Transform in accuracy, but performs poorly compared to Symbolic Aggregate Approximation. While Matrix Profile exhibits shorter computational time than Symbolic Aggregate Approximation, it slightly extends beyond other classic measures. Thus, Matrix Profile presents competitive advantages among ii distance measurement methodologies. By providing a comprehensive examination of similarity measurement techniques, this study equips the idea for the strength and weaknesses of distance measures, providing valuable insight for decision-making in time series data mining activities. 2024-01 Final Year Project / Dissertation / Thesis NonPeerReviewed application/pdf http://eprints.utar.edu.my/6490/1/LeeJiaYee_Full_Report_Finalised.pdf Lee, Jia Yee (2024) Exploring distance measures for time series data: A comparative analysis. Final Year Project, UTAR. http://eprints.utar.edu.my/6490/
spellingShingle Q Science (General)
QA75 Electronic computers. Computer science
T Technology (General)
Lee, Jia Yee
Exploring distance measures for time series data: A comparative analysis
title Exploring distance measures for time series data: A comparative analysis
title_full Exploring distance measures for time series data: A comparative analysis
title_fullStr Exploring distance measures for time series data: A comparative analysis
title_full_unstemmed Exploring distance measures for time series data: A comparative analysis
title_short Exploring distance measures for time series data: A comparative analysis
title_sort exploring distance measures for time series data: a comparative analysis
topic Q Science (General)
QA75 Electronic computers. Computer science
T Technology (General)
url http://eprints.utar.edu.my/6490/
http://eprints.utar.edu.my/6490/1/LeeJiaYee_Full_Report_Finalised.pdf