Learning Social Networks from Text Data using Covariate Information

Accurately describing the lives of historical figures can be challenging, but unraveling their social structures perhaps is even more so. Historical social network analysis methods can help in this regard and may even illuminate individuals who have been overlooked by historians, but turn out to be influential social connection points. Text data, such as biographies, are a useful source of information for learning historical social networks but the identification of links based on text data can be challenging. Traditional methods directly

use the number of name co-mentions in the text to infer relations. The use of a conditional independence

structure reduces the tendency to overstate the relationship between \friends of friends". However, this method does not take into account the abundance of covariate information that is often available in text data.

In this work, we first explore the effect of multiple conditional independence structures on reconstructing social network from the text. Then we extend the Local Poisson Graphical Lasso model with a (multiple) penalty structure that incorporates covariates, opening up the opportunity for similar individuals to have a higher probability of being connected. We propose both greedy and Bayesian approaches to estimate the penalty parameters and present results on data simulated with characteristics of historical networks and show that this type of penalty structure can improve network recovery as measured by precision and recall. We also illustrate the approach on biographical data of individuals who lived in early modern Britain between

1500 to 1575. Finally, we show the model can also incorporate continuous covariates and discuss several

applications of how to create continuous covariates from text data.