WEIGHTED ACCURACY ALGORITHMIC APPROACH IN COUNTERACTING FAKE NEWS AND DISINFORMATION

Kwadwo Osei Bonsu

Economic and Regional Studies (Studia Ekonomiczne i Regionalne), 2021, vol. 14, issue 01

Abstract: Subject and purpose of work: Fake news and disinformation are polluting information environment. Hence, this paper proposes a methodology for fake news detection through the combined weighted accuracies of seven machine learning algorithms. Materials and methods: This paper uses natural language processing to analyze the text content of a list of news samples and then predicts whether they are FAKE or REAL. Results: Weighted accuracy algorithmic approach has been shown to reduce overfitting. It was revealed that the individual performance of the different algorithms improved after the data was extracted from the news outlet websites and 'quality' data was filtered by the constraint mechanism developed in the experiment. Conclusions: This model is different from the existing mechanisms in the sense that it automates the algorithm selection process and at the same time takes into account the performance of all the algorithms used, including the less performing ones, thereby increasing the mean accuracy of all the algorithm accuracies.

Keywords: Research and Development/Tech Change/Emerging Technologies; Research Methods/Statistical Methods (search for similar items in EconPapers)
Date: 2021
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://ageconsearch.umn.edu/record/310404/files/0 ... nr1-2021%20Bonsu.pdf (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:ags:plecrs:310404

DOI: 10.22004/ag.econ.310404

Access Statistics for this article

More articles in Economic and Regional Studies (Studia Ekonomiczne i Regionalne) from John Paul II University of Applied Sciences in Biala Podlaska
Bibliographic data for series maintained by AgEcon Search ().