Answers are grounded in this paper and cite their sources.
ABSTRACTSociety grapples with an overwhelming and ceaseless flow of information,posing challenges to a media environment already afflictedby eroding trust in news. The use of Supervised Learning modelsfor Fake News classification is widespread, yet their effectivenesshinges on the quality of labeled data. Constructing datasets thatencompass the intricate nuances of disinformation across diversecontexts remains a formidable task. This study presents a comparativeanalysis of various Supervised Learning models for detectingand classifying misinformation. Leveraging the LIAR dataset, whichemploys six different classes to characterize the veracity of statements,our findings align with the accuracy benchmarks establishedby the LIAR authors. Specifically, the logistic regression model withstemming achieves an accuracy of 25%. The study suggests potentialenhancements through the application of Deep Learningtechniques, revealing a positive correlation between accuracy andthe number of training epochs. Despite current accuracy levels,notably lower than datasets with binary classifiers, it is crucialto underscore the meticulous manual verification and annotationprocess executed by the LIAR dataset authors.
No takes yet. Share an insight, caveat, or question.
Pinto et al. (2024) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: