Date of Completion

2019

Document Type

Honors College Thesis

Department

College of Engineering and Mathematical Sciences

Thesis Type

Honors College

First Advisor

Margaret J. Eppstein

Second Advisor

Chris Danforth

Keywords

natural language processing, palliative care, sentiment analysis, temporal reference, story arc

Abstract

Palliative care is an approach to improving the quality of life for patients with a serious, most likely terminal, condition. Palliative care conversations are often referred to by professionals as ‘narratives’, as the conversations are guided dynamically to best fit patient needs. Using transcribed text conversations of 354 palliative care consultations from 225 patients, we investigate trends in word usage over narrative time. Using crowdsourced sentiment rankings of the most common terms in the English language, we find that decreasing references to illness terms increases sentiment over narrative time. We then explore temporal references by looking at the usage of yesterday, today, and tomorrow, as well as variations in verb tense more generally. We find that discussion of the past decreases throughout the conversation, while discussion about the future increases. Our findings provide clinically-relevant insight into the storyline of a palliative care conversation, helping professionals to better understand these critical discussions.

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.

Download

Share

COinS