Comparison of Toxicity Among Female and Male Active Politicians in Social Media
Name
Devrim Nesipoglu
Abstract
Toxicity analysis on social media is critical in understanding and addressing of hateful behaviors and discourse. This master's thesis aims to comprehensively compare toxicity levels in online social media discourse among American male and female politicians. The study uses a multifaceted machine learning approach and natural language processing (NLP) techniques. Leveraging sentiment analysis, we measure the sentiment of posts and comments made by politicians while using models for toxicity detection to classify text as toxic or non-toxic. In addition, we separate the data into male and female categories, thus enabling a detailed comparison. Statistical analysis is then applied to assess and compare toxicity levels between the two groups, shedding light on possible gender-based differences in online discourse. Through visualization and interpretation of results, we aim to contribute to understanding toxicity patterns in social media on political communication and gender dynamics.
Graduation Thesis language
English
Graduation Thesis type
Master - Innovation and Technology Management
Supervisor(s)
Uku Kangur, Rajesh Sharma
Defence year
2024