Data Science links about Covid-19 | Combine

Data Science links about Covid-19

As researches around the world rush to gain as much information and knowledge about the new Covid-19 virus as possible, it is of high importance to be able to convey any new findings to politicians, experts, and the public in general. This task may sound simple enough, but with lacking visualization techniques and a very high speed in the data flow, the important information could easily get lost. In this blog post you can read about some updates in the Data Science field concerning Covid-19.

Here is an example of an intuitive, easy to use interactive dashboard which shows the number of cases for different countries, and cumulative in the world: https://www.gohkokhan.com/corona-virus-interactive-dashboard-tweaked/ 

Another issue for the scientific world is how to gather all the information from different sources and regions. To date, around 29 000 scientific papers have been published focused on Covid-19 (https://www.theregister.co.uk/2020/03/17/ai_covid_19/). To go through and categorize, analyse and join the information from all these papers is a huge task if done manually. But, this is where AI could come into play. Using Natural Language Processing (NLP) techniques, an AI model could go through and gather all important information from all papers in a fraction of the time it would take a human to do the same. Without a doubt, more research on the subject will emerge and the dataset will only grow from here, so utilizing the power of our computers and AI methods will be essential to keep growing our knowledge globally going forward. In fact, this work has already been started in the U.S, and an open dataset is now available (https://pages.semanticscholar.org/coronavirus-research). This dataset is meant to be machine readable, to enable machine learning algorithms to utilize the information within.   

Lastly, we would like to mention that the largest social networks have announced that they will pick up the fight against misinformation on their platforms. As history already has proven, advertisements and misinformation online can really affect the populations behaviour, so this announcement is well received.  

Contact us