Investigating the Effectiveness of Detecting Misinformation on Social Media using Tshivenda Language

Abstract

The spread of misinformation on social media poses a major challenge to information integrity and public discourse. This study examines the effectiveness of detecting misinformation in the Tshivenda language, an underrepresented linguistic context, on social media platforms. Tshivenda is a Bantu language spoken primarily in the Limpopo province of South Africa [1]. This research analysed misinformation patterns, adapted existing detection techniques, and examined the influence of Tshivenda language linguistics. Through an extensive literature review, the investigation examined the state of the art in misinformation detection and its applicability to languages with limited datasets. To address this gap, we used LSTM models, a type of RNN known for capturing long-range dependencies, for misinformation detection. This research involved training and evaluating the LSTM model on the Tshivenda and English datasets. This comparative analysis provided valuable insights into the challenges and opportunities that linguistic diversity presents in detecting misinformation. The results shed light on the effectiveness of using Authorized licensed use limited to: CSIR Information Svcs. Downloaded on July 16,2024 at 15:25:37 UTC from IEEE Xplore. Restrictions apply. Copyright © 2024 The authors www.IST-Africa.org/Conference2024 Page 2 of 5 LSTM models to detect misinformation in underrepresented languages. By analysing the results from the Tshivenda and English datasets, we were able to gain valuable insights into the differences in performance and the impact of linguistic variation on the accuracy of misinformation detection.