Module 22: Information Networks in the Moral Narrative Analyzer (MoNA)

Faculty Contact: Rene Weber

Research Area(s): 

Abstract: The goal of MoNA is to classify the fundamental moral domains that permeate mediated narratives, to identify systematic differences in these domains across sources and cultures, and to reveal the underlying temporal dynamics of moral perspectives in narratives that drive judgment and decision making in various types of audiences (e.g. movie goers, news viewers). We are using advanced language processing methodology for the analysis of large quantities of unstructured narratives from sources across the globe. Since topics are often subjective and difficult to identify using computerized language processing methods alone, MoNA provides a hybrid between automated computational methods and evaluations from people who are exposed to moral narratives.1

In this module, IGERT trainees assist with the further development of the MoNA News Narrative Analyzer project. Jacob Fisher writes,

I am hoping to continue to work on ingesting more data and refining our filtering methods to create a more robust network. Primarily I’m interested in the dynamics of cluster centrality in the network. More central clusters correspond to more highly co-mentioned events. Do highly central clusters exhibit patterns of event type, tone of discourse, etc.? How do these patterns change based on how the news data is filtered (e.g. more liberal or more conservative news sources, US vs. non-US sources, etc.)

 

Active Quarters:

  • Winter 2017: Jacob Fisher and Frederick Hopp

 

References:

1. The Moral Narrative Analyzer website (https://mnl.ucsb.edu/mona/)