FaD-CODS Fake News Detection on COVID-19 Using Description Logics and Semantic Reasoning

FaD-CODS Fake News Detection on COVID-19 Using Description Logics and Semantic Reasoning

Kartik Goel, Charu Gupta, Ria Rawal, Prateek Agrawal, Vishu Madaan
DOI: 10.4018/IJITWE.2021070101
OnDemand:
(Individual Articles)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

COVID-19 has affected people in nearly 180 countries worldwide. This paper presents a novel and improved Semantic Web-based approach for implementing the disease pattern of COVID-19. Semantics gives meaning to words and defines the purpose of words in a sentence. Previous ontology approaches revolved around syntactic methods. In this paper, semantics gives due priority to understand the nature and meaning of the underlying text. The proposed approach, FaD-CODS, focuses on a specific application of fake news detection. The formal definition is given by depiction of knowledge patterns using semantic reasoning. The proposed approach based on fake news detection uses description logic for semantic reasoning. FaD-CODS will affect decision making in medicine and healthcare. Further, the state-of-the-art method performs best for semantic text incorporated in the model. FaD-CODS used a reasoning tool, RACER, to check the consistency of the collected study. Further, the reasoning tool performance is critically analyzed to determine the conflicts between a myth and fact.
Article Preview
Top

1. Introduction

COVID-19 (aka SARS-COV-2 or n-Coronavirus) originated from Wuhan food markets in China in December 2019 (Qin et. al, 2020), has now spread to different parts of the world (Holmes & K.V, 2003). It has infected more than 40 million people worldwide and still this number is increasing exponentially. This pandemic has mis-balanced and disrupted ecology and life cycle, beyond repairs. In order to combat this virus, proper study of its disease patterns, source of infection, diagnosis procedure (Li et. al, 2020), virus structure information (Anand et. al, 2003) and the method of actions performed by various medicines and drugs against it, different domain experts are working hard and examining carefully. COVID-19 virus has presented many challenges to global health (Phelan et. al, 2020) due to the inherent uncertainty involved with the disease outbreak.

Since ages, it is known that when situations are grave and there is a disease outbreak, many kinds of unwanted and unethical formulations are floated in the society by various anti-social elements. They do so in order to get attention, money, profit or they think of a long-time money-making plan. Such situations also give rise to a very lethal challenge of Fake News. Fake news are low quality information pieces with false data (Rubinet. al, 2016). These fake news data are propagated purposefully through various channels to the general public in order to foresee a profitable gain (Hardalov et. al, 2016) related to social, economic or cultural diaspora. Fake news is also known to influence and deceive the reader about a situation / issue. In the current scenario, fake news is spread about the COVID-19 disease and its associated modalities. It is observed that fake news is necessary to analyze (especially in pandemic times) because fake news often leads to uncontrolled circumstances and is very dangerous.

In this paper, the challenges arising due to the fake news are addressed using Description Logic. Description logic is a formal knowledge representation that is used for semantic reasoning.

This paper aims to give useful insights to the following:

  • Understanding the effect of Fake News on the spread of a deadly disease outbreak

  • The functionality of anti-social elements floating the fake news

The Proposed FaD-CODS (Fake news Detection on COVID-19 using Description logics and Semantic reasoning) framework gathers and studies the information about COVID-19 from various online sources and presents several facts and myths about this disease. Description logic (DL) helps in formulating axioms and provides semantics to the facts. Implementation of DL (Grosof et. al, 2003) axioms helps in classifying the fake news spread about COVID-19. Several myths and facts which were quick to spread with the outbreak of COVID-19 are explicitly listed and explained. We aim to use DL and ontological axioms (Baader et. al, 2012) to present our COVID-19 model in a concise and unequivocal manner. Ontological approach adopted to formulate a representational structure of axioms. Several fake news articles which began to spread with the pandemic, had gained huge attention. For instance, Chloro-quinine began to sell in large quantities, with the thought of people thinking them being the ultimate solution of COVID-19. Semantics (Lewis & D., 1997) the branch of web-based ontology has been adopted to differentiate myths from facts. Web-based ontology (OWL) is a formal language for representation. It allows data to be shared among different applications and acts as an integrator of facts and data. Thus, the features collectively focus on the sole purpose of making the treatment easier by seeing it through the perspective of reasoning in semantics. All this can be considered to be a boon in the field of medicine and healthcare. DL or description logic presents a formal way of representation of reasoning of statements and facts. DL gives a formal way of representation for expression. Previous studies and research papers revolving around the prime subject of ontologies have shelled around and presented a tractable relationship between description logic and ontology. DL has been used by researchers for defining the concepts and roles from ontologies.

Complete Article List

Search this Journal:
Reset
Volume 19: 1 Issue (2024)
Volume 18: 1 Issue (2023)
Volume 17: 4 Issues (2022): 1 Released, 3 Forthcoming
Volume 16: 4 Issues (2021)
Volume 15: 4 Issues (2020)
Volume 14: 4 Issues (2019)
Volume 13: 4 Issues (2018)
Volume 12: 4 Issues (2017)
Volume 11: 4 Issues (2016)
Volume 10: 4 Issues (2015)
Volume 9: 4 Issues (2014)
Volume 8: 4 Issues (2013)
Volume 7: 4 Issues (2012)
Volume 6: 4 Issues (2011)
Volume 5: 4 Issues (2010)
Volume 4: 4 Issues (2009)
Volume 3: 4 Issues (2008)
Volume 2: 4 Issues (2007)
Volume 1: 4 Issues (2006)
View Complete Journal Contents Listing