Understanding Automatic Identification System Data as Applied to Social Network Relationships and Activities
Abstract
This thesis considers the problem of how to infer a social network of entities based solely on periodic location information as they move in space and time. Specifically, we consider social networks implied by vessel traffic within the South China Sea as captured by Automatic Identification System data consisting of vessel identifier, position coordinates (latitude and longitude), heading, speed, and other information about its course. We create customized data structures in the Python programming language and implement analgorithm to quickly and efficiently create a social network of entities based on proximity parameters for both space and time. As we see from our analysis, there is no single network depiction of our data. Rather, the topology is largely influenced by the parameters we use to define a connection between entities. Because we can efficiently query the data, we quickly draw conclusions about the most active time of day and mostactive location within the region. From this, we are able to look into specific entities and examine the relationships, activities, and behaviors that we wish to understand.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 2018
- Accession Number
- AD1059814
Entities
People
- Rachel Cline
Organizations
- Naval Postgraduate School