Twitter2015-Urdu

The Twitter2015-Urdu Dataset is a multimodal resource designed to advance Multimodal Named Entity Recognition (MNER) research in Urdu, a low-resource language. It adapts the widely used Twitter2015 English dataset with culturally grounded annotations tailored to Urdu's unique linguistic complexities. Featuring a balanced split across training, validation, and test sets, this dataset addresses the limitations of existing resources like CRULP, UCREL, and WikiDiverse, making it a critical tool for advancing Urdu-specific MNER research and multimodal applications.