Sberbank has improved the methodology of the Sber COVID-19 symptom checker, a free online service that allows the user to assess the risk of coronavirus disease and the likelihood of a mild course of the disease based on symptoms and external factors. The service was developed on the basis of a probabilistic artificial intelligence model and improved with the latest medical data.
In accordance with the latest data from world statistics, several new symptoms of COVID-19 were added to the calculation of the results and the specific weights added earlier were clarified. The questions of the online test were corrected, new ones were included in place of the outdated ones, including those about antibodies.
More than 36 thousand users have already passed the updated version of the test, information about it is available to the bank's clients in the stories in Sberbank Online. Since the launch of the Sber COVID-19 symptom checker, more than 250,000 people have been tested with it.
Similar questionnaires exist in many countries of the world, including the United States and the European Union, but Sberbank was the first to offer a service that instantly provides not only results, but also personalized recommendations for the prevention of coronavirus, taking into account the age and habits of each user.
Alexander Vedyakhin, First Deputy Chairman of the Management Board of Sberbank:
“In the first wave of the pandemic, our online test, or symptom checker, as we call it, proved to be effective. Then the test showed a high risk of coronavirus disease in about 10% of respondents. They consulted a doctor, and those who were already ill began to undergo treatment on time, and those whose suspicions were not confirmed began to take preventive measures more seriously. Equally important, in completely healthy, but suspicious people, our symptom-checker removed irrational anxiety and allowed them to live a fuller and more fulfilling life. Soon we will present the statistics of the results of the new version of the online test and compare it with the spring statistics."