Surfacing contextual hate speech words within social media
This work addresses the problem of improving hate speech detection for social media platforms by moving beyond fixed keyword lists, though it is incremental in building on existing classification methods.
The paper tackled the problem of detecting hate speech on social media by addressing the challenge of code words with alternate meanings that evade keyword-based detection, developing a community detection approach and a word embedding model to learn these meanings, and reporting inter-annotator agreement rates of K=0.871 and K=0.676 for their method versus keyword-based approaches.
Social media platforms have recently seen an increase in the occurrence of hate speech discourse which has led to calls for improved detection methods. Most of these rely on annotated data, keywords, and a classification technique. While this approach provides good coverage, it can fall short when dealing with new terms produced by online extremist communities which act as original sources of words which have alternate hate speech meanings. These code words (which can be both created and adopted words) are designed to evade automatic detection and often have benign meanings in regular discourse. As an example, "skypes", "googles", and "yahoos" are all instances of words which have an alternate meaning that can be used for hate speech. This overlap introduces additional challenges when relying on keywords for both the collection of data that is specific to hate speech, and downstream classification. In this work, we develop a community detection approach for finding extremist hate speech communities and collecting data from their members. We also develop a word embedding model that learns the alternate hate speech meaning of words and demonstrate the candidacy of our code words with several annotation experiments, designed to determine if it is possible to recognize a word as being used for hate speech without knowing its alternate meaning. We report an inter-annotator agreement rate of K=0.871, and K=0.676 for data drawn from our extremist community and the keyword approach respectively, supporting our claim that hate speech detection is a contextual task and does not depend on a fixed list of keywords. Our goal is to advance the domain by providing a high quality hate speech dataset in addition to learned code words that can be fed into existing classification approaches, thus improving the accuracy of automated detection.