Degree
Master of Science
Program
Computer Science
Supervisor
Mercer, Robert E.
2nd Supervisor
Xiao, Lu
Affiliation
Syracuse University
Co-Supervisor
Abstract
Tension is an emotional experience that can occur in different contexts. This phenomenon can originate from a conflict of interest or uneasiness during an interview. In some contexts, such experiences are associated with negative emotions such as fear or distress. People tend to adopt different hedging strategies in such situations to avoid criticism or evade questions.
In this thesis, we analyze several survivor interview transcripts to determine different characteristics that play crucial roles during tension situation. We discuss key components of tension experiences and propose a natural language processing model which can effectively combine these components to identify tension points in text-based oral history interviews. We validate the efficacy of our model and its components with experimentation on some standard datasets. The model provides a framework that can be used in future research on tension phenomena in oral history interviews.
Recommended Citation
Islam, Jumayel, "Tension Analysis in Survivor Interviews: A Computational Approach" (2018). Electronic Thesis and Dissertation Repository. 5878.
https://ir.lib.uwo.ca/etd/5878