Urdu News
Research on Urdu news currently focuses heavily on automated fake news detection, driven by the lack of readily available resources and the significant societal impact of misinformation in this language. This involves developing and benchmarking machine learning models, primarily employing transformer-based architectures like BERT and its variants, to classify news articles as real or fake. These efforts are crucial for improving information credibility and are contributing to broader advancements in low-resource language processing within the NLP community. Additionally, research is exploring improved news recommendation systems using NLP techniques like TF-IDF and BERT to enhance user experience and information access.
Papers
UrduFake@FIRE2021: Shared Track on Fake News Identification in Urdu
Maaz Amjad, Sabur Butt, Hamza Imam Amjad, Grigori Sidorov, Alisa Zhila, Alexander Gelbukh
Overview of the Shared Task on Fake News Detection in Urdu at FIRE 2021
Maaz Amjad, Sabur Butt, Hamza Imam Amjad, Alisa Zhila, Grigori Sidorov, Alexander Gelbukh
Exploiting Transliterated Words for Finding Similarity in Inter-Language News Articles using Machine Learning
Sameea Naeem, Dr. Arif ur Rahman, Syed Mujtaba Haider, Abdul Basit Mughal
Urdu News Article Recommendation Model using Natural Language Processing Techniques
Syed Zain Abbas, Dr. Arif ur Rahman, Abdul Basit Mughal, Syed Mujtaba Haider