Identifying nuances in sentiment and claimed words to spot yellow journalism

Abstract

This research delves into the realm of media integrity, focusing on the detection of yellow journalism through nuanced sentiment analysis and claim identification. A manual dataset preparation approach ensures granularity and accuracy. Leveraging Support Vector Machines (SVM) and Bidirectional Long Short-Term Memory (BiLSTM) networks, the study endeavours to elucidate the intricate interplay between claimed words and sentiment in journalistic content. The methodology involves meticulous claim and sentiment identification, followed by a rigorous analysis of their relationships. By employing both traditional SVM classifiers and cutting-edge BiLSTM architectures, the research aims to not only detect but also understand the underlying mechanisms of yellow journalism. Through meticulous evaluation and interpretation, the study sheds light on the effectiveness of these techniques in combating misinformation and preserving media integrity. This interdisciplinary approach advances the field of sentiment analysis and claim identification, underscoring the significance of nuanced understanding in combating the spread of misleading information.