How Hateful are Movies? A Study and Prediction on Movie Subtitles

How Hateful are Movies? A Study and Prediction on Movie Subtitles Niklas von Boguszewski ∗ nvboguszewski@googlemail.com Sana Moin ∗ moinsana77@gmail.com Anirban Bhowmick ∗ anirbanbhowmick88@gmail.com Seid Muhie Yimam seid.muhie.yimam@uni-hamburg.de Chris Biemann christian.biemann@uni-hamburg.de Language Technology Group Universität Hamburg, Germany Abstract In this research, we investigate techniques to detect hate speech in movies. We introduce a new dataset collected from the subtitles of six movies, where each utterance is annotated ei- ther as hate, offensive or normal. We apply transfer learning techniques of domain adapta- tion and fine-tuning on existing social media datasets, namely from Twitter and Fox News. We evaluate different representations, i.e., Bag of Words (BoW), Bi-directional Long short- term memory (Bi-LSTM), and Bidirectional Encoder Representations from Transformers (BERT) on 11k movie subtitles. The BERT model obtained the best macro-averaged F1- score of 77%. Hence, we show that trans- fer learning from the social media domain is efficacious in classifying hate and offensive speech in movies through subtitles. Cautionary Note: The paper contains exam- ples that many will find offensive or hateful; however, this cannot be avoided owing to the nature of the work. 1 Introduction Nowadays, hate speech is becoming a pressing is- sue and occurs in multiple domains, mostly in the major social media platforms or political speeches. Hate speech is defined as verbal communication that denigrates a person or a community on some characteristics such as race, color, ethnicity, gen- der, sexual orientation, nationality, or religion (Nockleby et al., 2000; Davidson et al., 2017). Some examples given by Schmidt and Wiegand (2017) are: • Go fucking kill yourself and die already a use- less ugly pile of shit scumbag. • The Jew Faggot Behind The Financial Col- lapse. ∗ Equal contribution • Hope one of those bitches falls over and breaks her leg. Several sensitive comments on social media platforms have led to crime against minorities (Williams et al., 2020). Hate speech can be con- sidered as an umbrella term that different authors have coined with different names. Xu et al. (2012); Hosseinmardi et al. (2015); Zhong et al. (2016) re- ferred it by the term cyberbully-ing, while David- son et al. (2017) used the term offensive language to some expressions that can be strongly impolite, rude or use of vulgar words towards an individual or group that can even ignite fights or be hurtful. Use of words like f**k, n*gga, b*tch is common in social media comments, song lyrics, etc. Although these terms can be treated as obscene and inappro- priate, some people also use them in non-hateful ways in different contexts (Davidson et al., 2017). This makes it challenging for all hate speech sys- tems to distinguish between hate speech and of- fensive content. Davidson et al. (2017) tried to dis- tinguish between the two classes in their Twitter dataset. These days due to globalization and online me- dia streaming services, we are exposed to different cultures across the world through movies. Thus, an analysis of the amount of hate and offensive content in the media …

Content truncated to 500 words