Cognitive Search: A Free Information Retrieval Web Service to Coronavirus Scientific Papers

Abstract

Coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is an ongoing global health emergency. With millions of people affected by the COVID-19 pandemic, many questions arise concerning transmission, diagnosis, treatment, development of vaccines, and viral pathogens. Bearing that in mind, the dangers of wrong and inaccurate information represent a socio-economic could do more damage than the disease itself. To help fight this ongoing outbreak, we present Cognitive Search - a friendly deployed service application IR exploring the latest language processing advance. The service provides access to CORD-19, a resource of scholarly articles about COVID-19 and related coronaviruses. The system allows rending documents retrieval by Term-Frequency and Semantic Neural Search and the Hybrid Term-Neural. The retrieval performance can often be significantly improved by using several different retrieval algorithms and allowing the user to combine the results instead of just one. Additionally, the Hybrid Term-Neural approach supports the exploitation of temporal information in documents and the usage of such information to anchor search results along a well-defined timeline. So it can generate insights through an intuitive and easy-to-use interface.