Lab Home | Phone | Search
Center for Nonlinear Studies  Center for Nonlinear Studies
 Colloquia Archive 
 Postdoc Seminars Archive 
 Quantum Lunch 
 Quantum Lunch Archive 
 CMS Colloquia 
 Q-Mat Seminars 
 Q-Mat Seminars Archive 
 P/T Colloquia 
 Kac Lectures 
 Kac Fellows 
 Dist. Quant. Lecture 
 Ulam Scholar 
 CNLS Fellowship Application 
 Student Program 
 Past Visitors 
 History of CNLS 
 Maps, Directions 
 CNLS Office 
Thursday, February 15, 2024
2:00 PM - 2:00 PM
CNLS Conference Room (TA-3, Bldg 1690)

Postdoc Seminar

Constructing a textual profile of social media posts that gain traction in online communities

Margaret Wood (A-1 / CNLS)
A-1 / CNLS

Social media has democratized both the consumption and dissemination of information online, creating opportunities to spread mis/disinformation with unprecedented ease. As a community-driven discussion platform with little top-down content moderation, Reddit is particularly hospitable to the dissemination of mis/disinformation. To better understand how information spreads on a platform relying on "collaborative filtering" of content, this study first seeks to identify the textual and linguistic characteristics of posts that successfully gain traction within the Reddit structure. As a first step, I employ grammatical analysis and various natural language processing techniques to explore a variety of features of these texts (i.e., author sentiment and subjectivity, topical content, and linguistic complexity). Through these analyses, I construct a textual profile of submission posts that generate a greater response from users on the platform. In the following phase, I will use this textual profile to train a machine learning model to identify social media posts that are more likely to spread efficiently through user-regulated online communities.

Host: Nidhi Parikh (A-1), Johnny Seales Jr. (A-1)