ORCID

https://orcid.org/0009-0007-3298-5093

Date of Award

Summer 2024

Language

English

Embargo Period

6-25-2024

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

College/School/Department

Department of Computer Science

Program

Computer Science

First Advisor

Petko Bogdanov

Committee Members

Petko Bogdanov, Frank Takes, Mariya Zheleva, Shaghayegh Sahebi

Keywords

Machine Learning, Sparse representation, Dictionary basis, Tensor factorization, Temporal networks

Subject Categories

Applied Statistics | Artificial Intelligence and Robotics | Data Science | Longitudinal Data Analysis and Time Series | Other Applied Mathematics | Signal Processing | Statistical Models

Abstract

Temporal networks arise in many domains including activity of social network users, sensor network readings over time, and time course gene expression within the interaction network of a model organism. Data of this type contains a wealth of prior information such as the connectivity among nodes (e.g., a friendship graph), and prior knowledge of expected temporal patterns (e.g., periodicity). Modeling these temporal and network patterns jointly is essential for state-of-the-art performance in temporal network data analysis and mining. Sparse dictionary encoding is one modeling approach for such underlying patterns. However, most classical approaches consider only one dimension of the data (i.e., network or temporal

priors but not both). To address this shortcoming of existing work I propose

novel frameworks for representation learning of temporal networks via appropriate dictionary bases. The methods which comprise this thesis utilize and incorporate network and temporal priors to learn novel representations in two keyways: data aggregation and multidictionary representation. The learned representations achieve state-of-the-art performance on a wide variety of tasks such as missing value imputation, future value forecasting, community detection, periodicity analysis, change point and anomaly detection, node classification, and link prediction. Beyond their quantitative performance, my proposed methodologies reveal insights into the underlying graph-time behavioral patterns in datasets from diverse domains. I demonstrate this through various case studies including discovery of communities

in global air traffic, identification of meaningful changes in the activity of users on the social network of reddit, and extraction of common interest groups among streamers on the Twitch platform.

License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Share

COinS