Euphemistic Phrase Detection by Masked Language Model
This addresses a content moderation challenge for social media platforms by enabling automated detection of multi-word euphemisms used by fringe groups, which is an incremental advance over existing single-word detection methods.
The paper tackles the problem of automatically detecting multi-word euphemisms, such as 'blue dream' for marijuana, by proposing a method that mines phrases, selects candidates using word embeddings, and ranks them with SpanBERT, achieving 20-50% higher detection accuracies compared to baselines.
It is a well-known approach for fringe groups and organizations to use euphemisms -- ordinary-sounding and innocent-looking words with a secret meaning -- to conceal what they are discussing. For instance, drug dealers often use "pot" for marijuana and "avocado" for heroin. From a social media content moderation perspective, though recent advances in NLP have enabled the automatic detection of such single-word euphemisms, no existing work is capable of automatically detecting multi-word euphemisms, such as "blue dream" (marijuana) and "black tar" (heroin). Our paper tackles the problem of euphemistic phrase detection without human effort for the first time, as far as we are aware. We first perform phrase mining on a raw text corpus (e.g., social media posts) to extract quality phrases. Then, we utilize word embedding similarities to select a set of euphemistic phrase candidates. Finally, we rank those candidates by a masked language model -- SpanBERT. Compared to strong baselines, we report 20-50% higher detection accuracies using our algorithm for detecting euphemistic phrases.