JMIR Publications

ABSTRACT

Background:

Since the beginning of the COVID-19 pandemic, more than one million studies have been collected within the COVID-19 Open Research Dataset Challenge (CORD-19), a corpus of manuscripts created to accelerate the research against the disease. Their related abstracts represent a wealth of information, which is – in many cases – yet to be explored, as well as unstructured and thus hardly searchable. Keyword-based search is the standard approach, which allows users to retrieve the documents of a corpus that contain (all or some of) the words in a target list. This kind of search, however, does not provide visual support to the task and is not suited to expressing complex queries, nor to compensating for missing specifications.

Objective:

As graphs are increasingly used to represent and query scientific knowledge, this paper proposes to consider small graphs of concepts and exploit them for expressing graph searches over existing COVID-19-related literature, providing a user-friendly search and exploration experience.

Methods:

We consider the CORD-19 corpus and summarize its content by annotating publications’ abstracts using terms selected from the Unified Medical Language System (UMLS) and the Ontology of Coronavirus Infectious Disease (CIDO). Then, we build a co-occurrence network that includes all relevant concepts mentioned in the corpus, establishing connections when their mutual information is relevant. A sophisticated graph query engine is built to allow the identification of the best matches of graph queries on the network, allowing as well partial matches and proposing candidate query completions (through shortest paths).

Results:

We build a large co-occurrence network, consisting of 128,249 entities and 47,198,965 relationships; the GRAPH-SEARCH interface allows users to explore the network by formulating or adapting graph queries; it produces a bibliography of publications, globally ranked; each publication is further associated with the specific parts of the query that it explains, thereby allowing the user to understand each aspect of the matching.

Conclusions:

Our approach supports the process of query formulation and evidence search upon a large text corpus; it can be reapplied to any scientific domain where documents corpora and curated ontologies are made available.