Date of Graduation

5-2021

Document Type

Thesis

Degree Name

Bachelor of Science

Degree Level

Undergraduate

Department

Computer Science and Computer Engineering

Advisor/Mentor

Gauch, Susan

Committee Member/Reader

Gauch, Susan

Committee Member/Second Reader

Nelson, Alexander

Committee Member/Third Reader

Le, Thi Hoang Ngan

Abstract

The purpose of this project is to explore the effectiveness of emotional analysis as a means to automatically moderate content or flag content for manual moderation in order to reduce the workload of human moderators in moderating toxic content online. In this context, toxic content is defined as content that features excessive negativity, rudeness, or malice. This often features offensive language or slurs. The work involved in this project included creating a simple website that imitates a social media or forum with a feed of user submitted text posts, implementing an emotional analysis algorithm from a word emotions dataset, designing a system to configure tolerance thresholds on a per-emotion basis, implementing the process of determining violations of incoming text posts using the configuration, and testing the effectiveness of the emotional analysis algorithm at determining toxic posts using a dataset of posts that have been manually reviewed for toxicity by a group of human moderators.

Keywords

software, moderation, moderator, emotional analysis, sentiment analysis

Download

Share

COinS