Computer Science Department, PhD Defense, Ramoza Ahsan " Efficient Time Series Data Analytics"

Thursday, April 25, 2019
10:00 am to 11:00 am

Location:

Floor/Room #: 
115

Committee members:

Prof.Elke A. Rundensteiner (Advisor), WPI Computer Science

Prof. Gabor Sarkozy (Co-advisor), WPI Computer Science

Prof. Xiangnan Kong, WPI Computer Science

Prof.Vassilis Athitsos, University of Texas at Arlington – External member

Abstract:

Finding similar trends and patterns among time series data is critical for a wide range of applications. My dissertation addresses challenges related to exploring time series using a rich variety of similarity distance measures. With time series being high-dimensional objects, detection of similar trends especially at the granularity of subsequences  or among time series of different lengths and temporal misalignments incurs prohibitively high computation costs. Finding trends using non-metric correlation measures further compounds the complexity, as traditional pruning techniques cannot be directly applied.

To address these challenges while meeting the need to achieve near real-time responsiveness, data reduction strategies are designed leveraging the inexpensive Euclidean distance with subsequent time warped matching on the reduced data. To support a rich diversity of correlation analytics, we compress time series into Euclidean-based clusters augmented by a compact overlay graph encoding correlation relationships. Finally, for retrieving exact similarity results using Lp-norm distances, we design a two-layered time series index for subsequence matching.  Its powerful pruning capability greatly reduces the number of time series as well as subsequence comparisons, resulting in a several order of magnitude speed-up. Comprehensive experimental studies using real-world and synthetic datasets demonstrate the efficiency, effectiveness and quality of the results achieved by our proposed techniques as compared to the state-of-the-art methods.