Date on Master's Thesis/Doctoral Dissertation

8-2025

Document Type

Doctoral Dissertation

Degree Name

Ph. D.

Department

Computer Engineering and Computer Science

Degree Program

Computer Science and Engineering, PhD

Committee Chair

Zhang, Hui

Committee Member

Elmaghraby, Adel

Committee Member

Altiparmak, Nihat

Committee Member

Park, Juw Won

Committee Member

Ruan, Guangchen

Author's Keywords

visual analytics; multimodal time series; anomaly exploration; dynamic change analysis; user-centric interaction; scalable analysis; personalized modeling; cohort-level comparison

Abstract

In today's data-intensive landscape, rapid advances in digital sensing and recording technologies have enabled the acquisition of high-resolution multimodal time series data, capturing intricate real-world dynamics across various domains such as healthcare, behavioral science, and environmental monitoring. However, the complexity and scale of these datasets present significant analytical challenges, particularly in understanding dynamic changes at both individual and cohort levels. This dissertation introduces EvoMetric, a novel visual analytics framework designed to support scalable exploration and analysis of large-scale multimodal time series data with dynamic changes. EvoMetric seamlessly integrates individual-level temporal dynamics with population-level comparative insights, enabling users to visually analyze, explore, and interpret dynamic temporal changes. Leveraging interactive visualization techniques, scalable cloud-based infrastructure, and rigorous multi-scale analytical methods, EvoMetric facilitates intuitive and responsive data exploration. The contributions of this research include: (1) EvoMetric employs an Elasticsearch-based backend infrastructure to efficiently manage, index, and retrieve extensive multimodal datasets. Its interactive ribbon-based visual interface provides users with seamless navigation across temporal scales and multimodal data streams, enabling intuitive hypothesis generation and pattern discovery. (2) Multi-scale Dynamic Change Analysis: EvoMetric integrates personalized temporal modeling through the Individual-Specific Temporal Dynamics Modeling (ISTDM) module with comprehensive cohort-level analysis via the Cohort-Wide Comparative Analysis Engine (CCAE). This unified approach allows users to assess individual deviations within statistically validated cohort trends, fostering deeper insights into complex temporal dynamics. Through detailed case studies involving representative subjects from distinct cohorts, EvoMetric demonstrated its capabilities in identifying meaningful temporal changes in circadian rhythm deviations and long-term behavioral differences. This integrated, scalable approach advances the field of visual analytics by providing researchers and domain experts with actionable insights into multimodal time series data and enabling informed decision-making based on dynamic temporal behaviors.

Download

Share

COinS