Automatic identification of Urdu fake news using Logistic Regression Model

Abstract

Social media offers a platform to disseminate information with family and friends quickly. The spread of fake news on social media has a significant social and economic impact. With the ever-increasing amount of social media data, it is challenging to quickly differentiate between real and fake news. In previous years, the research community focused on Fake news classification for the English language. However, many resource-poor languages, such as Urdu, still require efficient methods to classify and contain fake news. This study proposes a methodology to identify Urdu fake news based on machine learning techniques. Our proposed methodology uses the TF-IDF feature extraction technique and Logistic regression classifier to classify Urdu fake news automatically. The proposed approach outperforms the baseline with a 72%f1 score.