biswanathdutta / CODO

CODO is an ontology for the semantic representation and annotation of COVID-19 data in a machine-readable form for tracking history of the disease and patient.
9 stars 4 forks source link

CODO: COviD-19 Ontology for collection and analysis of data

An Ontological Framework for collection, representation, and analysis of COVID-19 pandemic data

It is a formal data model (an ontological framework) to facilitate the semantic representation and publication of COVID-19 data in a machine-processable format (as a Knowledge Graph). The model primarily has two dimensions: representation of COVID-19 cases and patient-oriented information at the granular level. The first dimension refers to data representation, such as active cases, recovered, deceased, migrated, ICU beds needed, invasiveVentilatorsNeeded, etc. distributed across the geolocation (district level, state (province), country) and time (e.g., date, time). The latter one refers to the patient data representation, such as symptoms, level of COVID-19, travel information, suspected reasons for catching the disease, inter-patient relationships, diagnosis results on a daily basis, COVID-19 medical facilities, etc. The model has been produced following the RDF directed level graph data model (https://www.w3.org/TR/rdf11-primer/) and Linked Data (LD) principles (https://www.w3.org/standards/semanticweb/data). The ontology has been expressed in OWL (https://www.w3.org/TR/owl2-primer/). On top of it, SWRL (Semantic Web Rule Language) rules (https://www.w3.org/Submission/SWRL/) have been created to enable the inferencing.

Some of the use cases of the proposed data model are:

(1) The organization and representation of COVID-19 case data on a daily basis following the LD Principles and OWL Language. So, the produced data can be queried and retrieved semantically, and can also be taken as an input to carry out advanced analytics (e.g., trend study, growth projection).

(2) The representation of patient (anonymous) data, the relationships between patients (if any), between patient and locations, time, etc. This network data may help in analysing the behavior of the disease, possible route of disease spread, various factors of disease transmission, etc.

(3) The data published following the proposed Knowledge Graph (KG) will also help the policymakers, for example, in understanding the state of the infrastructure, failure study, in resource allocation for a similar kind of pandemic in the future.

Persistent URI

https://w3id.org/codo redirects to https://www.isibang.ac.in/ns/codo/index.html

Contacts

Biswanath Dutta (Indian Statistical Institute)

(@biswanathdutta) dutta2005@gmail.com, bisu@drtc.isibang.ac.in