It can be said that women are one of the most affected groups by digital hate speech. Violence against women encompasses attitudes and behaviors that affect women and lead to gender-based discrimination, as well as human rights violations. With the increasing digitalization, individuals with access to unmonitored environments on social media can resort to digital violence against women. Such posts may include digital violence against women or positive messages supporting women. This study examines digital hate speech related to women in the context of Twitter (X). The aim of this research is to analyze emotions and thoughts regarding social events like violence against women through text mining methods and determine the extent to which campaigns against such social events influence emotions and thoughts.
In this research, the contents of messages within the #violenceagainstwomen hashtag on Twitter (X) were examined. Necessary libraries were included in the Python programming language to extract the data. Discourse and sentiment analyses were conducted using text mining methods on tweets about violence against women.
A total of 5,000 tweets from March 13, 2021, to January 17, 2023, were collected, and sentiment analyses were performed. BERT model was used for sentiment analysis. The research results indicate that the content consisted of 63% negative, 9% neutral, and 28% positive tweets. It was found that negative content mostly came from individuals inclined towards violence, whereas positive messages were supportive of women, not endorsing violence against women.
Anahtar Kelimeler: Violence Against Women, Twitter (X), Text Mining Analysis, Sentiment Analysis, BERT Model.