Keyword Search over the COVID-19 Data

Abstract

Keyword search is typically associated with information retrieval systems. However, recently, keyword search has been expanded to relational databases and RDF datasets, as an attractive alternative to traditional database access. With this motivation, this paper first introduces a platform for data and knowledge retrieval, called DANKE, concentrating on the keyword search component. It then describes an application that uses DANKE to implement keyword search over two COVID-19 data scenarios.

SSM: A Semantic Metasearch Platform for Scientific Data retrieval

“…This prompted research geared to the coronavirus (rather than generic search mechanisms). An example of COVID-directed data research is the work of Izquierdo et al [12], who proposed a platform to search COVID-19 Scientific Semantic Metasearch (SSM): Scientific Data Retrieval Platform data. Their platform receives natural language queries, extracts keywords, and applies the search over the contents of two COVID-19 repositories, the Brazilian NSG (Notificac ¸ões de Síndrome Gripal) and the one provided by Johns Hopkins University.…”

“…Figure 5 presents an example of the output. The example explores ontologies available in AgroPortal 12 . Figure 5 illustrates the stepby-step processing of term "Soybeans" resulting in 5 concepts in the final query expansion.…”

Addressing search in scientific open data repositories: A semantic metasearch platform

“…This prompted research geared to the coronavirus (rather than generic search mechanisms). For instance, Izquierdo et al [Izquierdo et al 2020] proposed a platform to search COVID-19 data. Their platform receives natural language queries, extracts keywords, and applies the search over the contents of two COVID-19 repositories, the Brazilian NSG (Notificações de Síndrome Gripal) and the one provided by Johns Hopkins University.…”

A Novel Solution to Empower Natural Language Interfaces to Databases (Nlidb) to Handle Aggregations

“…Another experiment that is being carried out is the integration of GLAMORISE with another NLIDB called DANKE [García 2020;Izquierdo et al 2018Izquierdo et al , 2020Torres Izquierdo et al 2020], which is a NLIDB that lacks the ability to answer aggregation questions. DANKE is an NLIDB of the Keyword Search (KwS) type and, as such, are unable to answer questions with aggregation [Affolter et al 2019].…”

“…We also wanted to integrate GLAMORISE with an NLIDB that processed queries using RDF/SPARQL to prove that GLAMORISE works independently of the structure of the NLIDB database chosen, despite the fact that GLAMORISE uses a relational database (SQLite). DANKE [García 2020;Izquierdo et al 2018Izquierdo et al , 2020Torres Izquierdo et al 2020] A keyword query is just a list of keywords that the user wishes to search for without the need of stop words and all the elements involved in a complete NLQ.…”