SAS specialists created Scientific Literature Search and Text Analysis, a free environment with visual text analysis capabilities that uses AI, machine learning, natural language processing, linguistic rules and complex modeling methods.
The scientific community is constantly searching for a vaccine and the most effective treatments for COVID-19. Leading research groups have already collected and published more than 50 thousand articles on these topics in the Open Research Dataset (CORD-19).
SAS complements CORD-19 with models developed by linguists and health and pharmaceutical experts, using SAS Visual Text Analytics and SAS Visual Data Mining and Machine Learning on the SAS Viya platform.
Researchers can interactively study materials on various topics, including the effectiveness of various personal protection tools, social distancing, infection methods, treatment methods, virus mutations etc. The models also extract and visualize quantitative data, for example the incubation period and the number of COVID-19 infections from one person. The system allows to visualize generalized quantitative data, quickly determine the authority of sources and make a search for keywords in the original text.
"The new environment for analyzing scientific texts has no analogues on the market and provides an intuitive and accessible way to search for complex relationships in disparate texts. This cloud solution is at no charge and it should allow the research community to deal with the COVID-19 crisis faster," - said Alexander Zhukov, head of the health solutions department, SAS Russia/CIS.
As part of the COVID-19 counteraction effort, which includes many initiatives, SAS also released COVID-19 Epidemiological Scenario Analysis, a set of interactive analytical models created together with the Cleveland Clinic. The solution allows to model various scenarios to predict the consequences of the virus and also to evaluate the effectiveness of the health system at the regional level.