Social Media Safety Practices and Flagging Sensitive Posts

Today, social media has uniquely become a force for positive change, community building, and sharing ideas. But with millions of people, of all ages, tuned into social media many hazards and mishaps can arise due to inadequate content monitoring. Therefore, our research problem deals with social media safety practices, flagging sensitive posts, and effective content monitoring. In this paper, an examination will be made of the most impactful hazards and mishaps that arise from poor social media content monitoring, incorrect flagging, public safety, and the spread of misinformation. In addition, this paper will discuss the shortcomings of content monitoring tools currently on the market and how they can be improved. Finally, this paper goes over the findings and areas of research related to the University of Texas Dallas 2022 Software Safety REU program. Through the synthesis and culmination of exploring social media safety practices, we have curated an application to manually label social media posts according to a pre-established corpus of violent phrases.

I. Introduction

Social media is a ubiquitous force for positive change and community building. However, with millions of people tuned into social media [1], many hazards and mishaps can arise due to inadequate content monitoring. Lack of effective safety protocols and flagging algorithms on social media platforms lead to incorrectly flagged or unflagged posts, the spread of misinformation, and threats to public safety.