ScholarWorks@숙명여자대학교

초록

This research examines the gender stereotypes portrayed in South Korea’s “Thanks To” Challenge social media campaign, which aimed to express gratitude to healthcare workers during the COVID-19 crisis. Utilizing cutting-edge transfer-learning techniques for image processing, this study found that gender stereotypes persist in terms of age and the emotions extracted from the social media posts of those who participated in the challenge. The results indicate that among the 4,943 postings randomly selected from those who participated in the challenge, men were three times more likely to appear than women, whereas their extracted ages did not differ. Additionally, for those classified as posting male images, negative emotion scores (sadness and anger) were higher than was the case for female images, whereas female images exhibited higher levels of happiness. Moreover, we found that the emotionality score for happiness was extremely high through the quantifying process using transfer learning, thus reaffirming emotional stereotypes for women in the social media realm.

키워드