Analysis of the Tone and Framing of the LGBTQ+ Community in Turkish Media
Presentation Type
Poster
Faculty Advisor
Larissa Goulart da Silva
Access Type
Event
Start Date
26-4-2024 9:45 AM
End Date
26-4-2024 10:44 AM
Description
This research aims to provide insights into how the LGBTQ+ (Lesbian, Gay, Bisexual, Transsexual, Queer, Questioning and Ace) community is represented in Turkish media. Previous studies suggest that the media representation of the LGBTQ+ community in Turkey has been negative, reinforcing the stigma and discrimination faced by this community (Cetin, 2016). To conduct this research, a Turkish news articles dataset, which was created by Kemik Doğal Dil İşleme Grubu (Kemik Natural Language Processing Group), was utilized. The dataset was last updated in 2013 and it contained 41,992 Turkish news articles, classified into 13 topics. AntConc (Anthony, 2023) was utilized to extract 11 keywords and their contexts from the news articles. The chosen keywords were “LGBT, LGBTQ, LGBTQ+, eşcinsel (homophile), gay, gey (gay), homoseksüel (homosexual), lezbiyen (lesbian), transseksüel (transsexual), travesti (transvestite), kuir (queer)”. Extracted sentences were analyzed, unrelated sentences were deleted, and the remaining sentences were organized and turned into a dataset to perform preprocessing and sentiment analysis in Python. Two models from Hugging Face, BERTurk, and XLM-RoBERTa were utilized to conduct sentiment analysis. The results were analyzed and turned into a dataset to visualize through the ggplot2 package of R. When the findings were analyzed, I found that the tone and framing of the LGBTQ+ community in Turkish media is mainly neutral and negative, meaning that the Turkish media preferred words that had neutral or negative connotations. This study reinforces the previous research and suggests that it is likely for media organizations to influence people’s opinions, beliefs, and conceptions towards the LGBTQ+ community negatively in Turkey.
Analysis of the Tone and Framing of the LGBTQ+ Community in Turkish Media
This research aims to provide insights into how the LGBTQ+ (Lesbian, Gay, Bisexual, Transsexual, Queer, Questioning and Ace) community is represented in Turkish media. Previous studies suggest that the media representation of the LGBTQ+ community in Turkey has been negative, reinforcing the stigma and discrimination faced by this community (Cetin, 2016). To conduct this research, a Turkish news articles dataset, which was created by Kemik Doğal Dil İşleme Grubu (Kemik Natural Language Processing Group), was utilized. The dataset was last updated in 2013 and it contained 41,992 Turkish news articles, classified into 13 topics. AntConc (Anthony, 2023) was utilized to extract 11 keywords and their contexts from the news articles. The chosen keywords were “LGBT, LGBTQ, LGBTQ+, eşcinsel (homophile), gay, gey (gay), homoseksüel (homosexual), lezbiyen (lesbian), transseksüel (transsexual), travesti (transvestite), kuir (queer)”. Extracted sentences were analyzed, unrelated sentences were deleted, and the remaining sentences were organized and turned into a dataset to perform preprocessing and sentiment analysis in Python. Two models from Hugging Face, BERTurk, and XLM-RoBERTa were utilized to conduct sentiment analysis. The results were analyzed and turned into a dataset to visualize through the ggplot2 package of R. When the findings were analyzed, I found that the tone and framing of the LGBTQ+ community in Turkish media is mainly neutral and negative, meaning that the Turkish media preferred words that had neutral or negative connotations. This study reinforces the previous research and suggests that it is likely for media organizations to influence people’s opinions, beliefs, and conceptions towards the LGBTQ+ community negatively in Turkey.