Electronic Theses and Dissertations

Date of Award

1-1-2023

Document Type

Thesis

Degree Name

M.S. in Engineering Science

First Advisor

Thai Le

Second Advisor

Feng Wang

Third Advisor

Timothy Holston

School

University of Mississippi

Relational Format

dissertation/thesis

Abstract

Recently, a lot of new natural language models have been developed which help individuals with writing. One of the widely used tools is Google’s Smart Compose which is a text-predictive system that helps individuals to write by reducing the need for repetitive typing. Though this tool has been well established and used by a lot of people, there is a lack of studies on its impact on open-ended writing.

In this thesis, we investigate the effect of Google’s Smart Compose in open-ended writing. To do this, we built a custom software that collects data while the users are writing on Google Docs web application. We recruited 119 individuals who wrote on Google Docs where 55 of the individuals had smart compose enabled while the other 59 had the smart compose disabled. We then compared the writings of those two groups. Additionally, we also compared the participants’ writing process under different suggestions.

Our results show that Google’s Smart Compose does not have a significant quantitative or qualitative effect on open-ended writing. However, one positive impact of Google’s Smart Compose is that it reduced the time required to write a character by 35.3 milliseconds. This was measured by comparing the time it took to write 10 characters before and 10 characters after the suggestion appeared.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.