Description
TitleLearning of network dynamics: mobility, diffusion and evolution
Date Created2022
Other Date2022-01 (degree)
Extent179 pages : illustrations
DescriptionWithin network science and network theory, there has been a lot of success in the analysis of complex networks in a static setting or at a certain snapshot. The analysis of network dynamics takes interactions of social features and temporal information into account. Different from static networks, dynamic networks consider larger, dynamic, multi-mode, multi-plex networks, and may contain varying levels of uncertainty. Due to the heterogeneity of networks, agent-based modeling and other forms of simulations are often used to explore how networks evolve and adapt as well as the impact of interventions on those networks. In this thesis, we discuss two aspects of this topic. In the first aspect, we consider human mobility and dynamic networks that are enabled by human trajectories. A human trajectory is a sequence of spatial-temporal data from an individual. Knowledge mined from human trajectories can help us improve a lot of real-world applications. However, the increasing size of the human trajectory dataset brings challenges to our analysis. At the same time, for human trajectory, we still lack for a comprehensive understanding of the relationship among the human trajectories, such as commonality and individuality. Thus, we focused on the statistic tools to measure the similarity between human trajectories. Different from the traditional geometric similarity measures, such as Hausdorff distance and Fr'{e}chet distance, we proposed three novel partial similarity measures, that are more suitable to human trajectories. Using the partial similarity measures can also reduce the storage demand and save computation time. Based on these similarity measures, we designed a advanced unsupervised clustering algorithm, leveraging with the conformal prediction framework. It performs better in varied trajectory datasets compared with the classical clustering algorithms. In the second aspect, we investigate two problems in the social interactions of the social networks, diffusion and evolution. For the first problem, previous works mainly focused on the information spreading speed in static social networks. Taking mobility into the consideration brings a new dimension to the problem, where interactions between different individuals happen over time. The heterogeneity of mobility plays an important role in the diffusion process. We utilize human trajectory dataset to simulate the physical interactions among individuals which allowed us to predict the diffusion process and dependency on parameters. . At the same time, we also investigate the impact of targeted interventions on the social networks. In the second problem, we study the evolution process of opinion dynamics (people's view on a particular topic). We proposed a co-evolution model, including the opinion dynamics and social tie dynamics, to investigate the community structure and structural balance in opinion dynamics (i.e., consistency of opinion and signs of ties along cycles) in the final state with rigorous theoretical analysis. We also show applications of this model in predicting evolution in a real world data set.
NotePh.D.
NoteIncludes bibliographical references
Genretheses
LanguageEnglish
CollectionSchool of Graduate Studies Electronic Theses and Dissertations
Organization NameRutgers, The State University of New Jersey
RightsThe author owns the copyright to this work.
Statistical Profile