Please join us for a talk being given by Mai ElSherief from the University of California, Santa Barbara.
Computational Methods for Next Generation Online Media Ecosystems
While social media has become an empowering agent to individual voices and freedom of expression, it also facilitates anti-social behavior including online harassment, cyberbullying, and hate speech. In a Pew Research Center study, 60% of Internet users said they had witnessed offensive name calling, 25% had seen someone physically threatened, and 24% witnessed sustained harassment of an individual. This talk focuses on speech that denigrates a person because of their innate and protected characteristics, which is also known as hate speech. The majority of prior work tackling online hate speech has focused on classification without analyzing the nuances of hate speech. In this talk, I present our state of the art research on hate speech characterization and detection. I will begin by discussing our linguistic analysis, the first to differentiate between different forms of hate speech based on the target of hate -- either directed towards a person or entity or generalized towards a group of people. I also note that hate speech is not just individual posts but also an organized effort. I will discuss NLP methods that enable us to understand hate group linguistic temporal evolution. I will conclude the talk by introducing a novel design of a hate speech detection system that leverages inter and intra-user representation to minimize the impact of noise present in social media posts on hate speech detection systems.
The Stanford Institute for Human-Centered Artificial Intelligence