•  
  •  
 

Abstract

In today's age of excessive information, the proliferation of fake news has emerged as a significant challenge, eroding the trustworthiness of news sources and posing serious threats to society. With the advent of social media and the internet, false information spreads more easily, necessitating the development of effective methods to identify and prevent its dissemination.While machine learning models have been widely employed to classify text data as authentic or fake, the majority of existing research has primarily focused on well-resourced languages such as English, leaving low-resourced languages like Kurdish largely overlooked. To address this gap, our proposed work introduces a novel approach: a hard voting ensemble method that combines the insights of four weak learners. By fine-tuning these weak learners for optimal performance, our approach achieves enhanced accuracy compared to the current state-of-the-art methods. Specifically, our findings demonstrate the effectiveness of the hard voting approach using a combination of Support Vector Machines (SVM), Decision Trees (DT), and Naive Bayes (NB) classifiers, resulting in an impressive accuracy rate of 89.73%. This outperforms the individual SVM classifier and underscores the potential of ensemble techniques in fake news detection

Download

Share

COinS