A Novel Method for Estimating Similarity of Multivariate Time Series

Comparison metrics for multivariate time series data are needed to determine if physics-based experiments are repeating instead of relying of visual comparison. While there are many techniques for determining similarity between univariate time series, a single metric is needed for quantitative analysis that can handle multivariate time series. Machine Learning was used to apply clustering techniques to unlabeled data to assign groups and a K-Nearest Neighbor model was optimized for accuracy to verify the clustering. Several similarity metrics were evaluated and combined for a given data set of repeated physical experiments to determine an appropriate method for comparing multivariate time series as compared to the assigned groups.

Author

• Singleton, Elizabeth

Language

• en

Keyword

• similarity
• time series
• multivariate

Date created

• 2023

Type of Work

• Dissertation

License

• All rights reserved

Rights statement

• In Copyright

GW Unit

• Engineering Management

Degree

• D.Engr.

Advisor

• Etemadi, Amir

Committee Member(s)

• Etemadi, Amir
• Blackford, Joseph P.
• Biemer, Steven

Persistent URL

•  https://scholarspace.library.gwu.edu/etd/9z903083f