B AC, Mahesh K. Ontology is what makes data interesting: Interestingness framework
for COVID-19 corpora.
J Inf Sci 2023. [PMCID:
PMC10076162 DOI:
10.1177/01655515231161137]
[Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 04/08/2023]
Abstract
The COVID-19 pandemic has already shown to be a worldwide threat, demonstrating
how susceptible humans may be. It has also inspired experts from a range of
aspects and countries to find the potential solution to control the widespread.
In line with this, our research proposes a novel framework for finding
interesting facts from COVID-19 corpora using domain ontology. Since data mining
with domain knowledge provides semantically rich facts, we use ontology in our
proposed approaches. Most of the state-of-the-art methods rely on instance level
or user intervention. These methods do not entirely exploit the richness of
ontology. In this work, we demonstrate how to extract exciting rules from data
at ontology’s schema and instance levels. Our experiments were carried out on
two COVID-19 corpora that depict COVID-19 patients’ symptoms and drug
information. The proposed framework outperformed the traditional methods by
reducing the number of rules by 70% and generating semantic-rich rules that are
more user-readable and quickly adopted by decision-makers. Furthermore, to
support our claims, we compared the outcomes of the proposed framework with the
most recent approach in the field. Also, statistically significant tests and
domain expert evaluations are conducted to validate our framework.
Collapse