Inferring Missing Information On A Social Network
Abstract
Networks have long been a subject of study, but with an explosion in the amount of available data to describe them, machine learning (ML) methods have become a popular complement to traditional network analysis techniques. When the challenge of uncertainty enters the picture, many difficulties can be attacked with ML methods. Social network analysis is an analysis of networks among people using machine learning techniques. For example, social networks can be networks of friends or followers in social media, academic collaboration networks, or networks among workers. In this thesis, we propose a novel machine learning method to analyze social networks among people from a personal database via connecting similarities among people in order to get a better characterization of their relationships. This machine learning method is a non-parametric method so that we can utilize data sets containing categorical variables as well as data sets with small sample sizes or sparse networks. Under our method, we assume that nodes in a social network represent people of interest, and an edge is defined as a connection between two people in the group. To verify our methodology, we apply our method with the social network data collected in the Teenage Friends and Lifestyle Study and demonstrate its efficacy.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 2020
- Accession Number
- AD1114376
Entities
People
- Ross Spinelli
Organizations
- Naval Postgraduate School