Abstract | ||
---|---|---|
Most social media platforms grant usersfreedom of speech by allowing them tofreely express their thoughts, beliefs, andopinions. Although this represents incredible and unique communication opportunities, it also presents important challenges. Online racism is such an example. In this study, we present a super-vised learning strategy to detect racist language on Twitter based on word embeddings that incorporate demographic (Age, Gender, and Location) information. Our methodology achieves reasonable classification accuracy over a gold standard dataset (F1=76.3%) and significantly improves over classification performance of demographic-agnostic model |
Year | Venue | Field |
---|---|---|
2017 | IJCNLP | Social media,Racism,Computer science,Supervised learning,Natural language processing,Artificial intelligence,Word embedding |
DocType | Citations | PageRank |
Conference | 1 | 0.35 |
References | Authors | |
12 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Mohammed Hasanuzzaman | 1 | 52 | 13.52 |
Gaël Dias | 2 | 354 | 41.95 |
Andy Way | 3 | 881 | 126.78 |