Social media is a virtual community where users share news, ideas, interests, and information. Learning the information diffusion dynamics and making decisions correspondingly, e.g., selecting the seed nodes to maximize the influence, have been widely applied to the areas of viral marketing and cyber security. In this dissertation, we study the problem of learning diffusion process, i.e., infection prediction, in social media networks utilizing both feature-based machine learning methods and mathematical model-based methods. For feature-based machine learning methods, the neighborhood information is treated as an important feature together with user profile and content similarity features. For model-based methods, two distinctive mathematical models, i.e., Linear Threshold Learning Model and Random Walk Learning Model, are proposed to learn the information diffusion dynamics. Neural networks are implemented to train the proposed models for all aforementioned methods. In this dissertation, we also study the problem of choosing seed nodes to maximize the influence in social media networks. In one project, the problem is addressed through solving the tiered influence and activation thresholds target set selection problem, which is to find the seed nodes that can influence the most users within a limited time frame. Both the minimum influential seeds and maximum influence within budget problems are considered in this study. In addition, we study the impacts arising from the uncertainties in network structures, user behavior and activation prices via two-stage stochastic optimization as well.
If this is your thesis or dissertation, and want to learn how to access it or for more information about readership statistics, contact us at STARS@ucf.edu
Doctor of Philosophy (Ph.D.)
College of Engineering and Computer Science
Industrial Engineering and Management Systems
Length of Campus-only Access
Doctoral Dissertation (Open Access)
Qiang, Zhecheng, "Learning and Decision Making in Social Media Networks" (2022). Electronic Theses and Dissertations, 2020-. 1073.