Online Hate Speech Detection and Management from Bystander Intervention Perspective based on ETPB Model

Abstract

The negative impact of hate speech spreading on social media, such as hatred towards targets’ sexual orientation, ethnicity, refugee and gender etc., has received increasing attention in recent years. Encouraging users to intervene as bystanders, such as reporting or flagging the hatred speech and making counter-speech, has gradually become a new trend for Internet governance by SNS providers. This study is based on the extension of the theory of planned behavior (ETPB) to identify the predictive factors for bystander intervention intention from a cognitive sight. Research was conducted through 486 online social media user questionnaires; the conclusion is perceived behavioral control, moral norms, and behavioral attitudes can positively predict the behavior intention, while the effect of subjective normative is not significant. The results could be piloted and implemented by SNS providers to encourage more active intervention from users to improve the efficiency of online hate speech detection and management.