Responding to COVID-19

In the midst of a global health crisis like COVID-19, how could data be more helpful? The main reason for slow responses on the part of leadership, both in businesses and in governments worldwide, is the lack of a clear picture of the current and projected future situation. What if there was a way for billions of analysts with the ears of Presidents, CEOs and Prime Ministers, to have simultaneous access to a global database with real-time updates on an evolving situation that is actively impacting lives and economies?

Most companies are guilty of underutilizing the troves of data sitting at their fingertips. How much faster could we react to a global situation like the coronavirus if we had access to all of that data and were able to interact with it via simple, user-friendly data-visualization dashboards that provide sub-second query response times?

While we may not yet have a system in place that provides a global data service on this level, the technology is already here. We can prepare for our future by implementing it and taking full advantage of the stores of data we already collect on a daily basis. Given the range of global responses to COVID-19, the need for a collaborative and universal data source is clear.

Our best chance of avoiding the next pandemic is to learn from our mistakes and our ignorance in the current rapidly developing international crisis and use that knowledge to improve our data infrastructures and emergency protocols so that we can be ready to share critical data and perform concurrent sub-second analysis en masse around the world.

Fortunately, businesses and governments are now taking action. A great example of this is China Mobile, which played a major role in providing critical data and insights to China’s government, businesses, and citizens as they dealt with the initial outbreak of the coronavirus.

Telecom Giant Uses Kyligence to Aid Government and Businesses in COVID-19 Crisis

China Mobile, one of the top telecommunication operators in China (the equivalent of AT&T or T-Mobile in the U.S.), built a coronavirus data analysis dashboard using Kyligence that helped guide the decisions of the Chinese government, businesses, landlords, and citizens and enabled them to act quickly when it came to determining who to quarantine, how long those people should remain in isolation and what regions were most likely to need additional support to contain and recover from the outbreak.

Utilizing cellphone data to track the location history of individuals across the country, they were able to help identify anyone who had been to the high-risk areas the government had identified as Zone A or Zone B. This allowed government officials and business leaders to make on-the-spot, data-driven decisions to reduce the spread of the virus. Those who had been to Zone A or B in the last few months were asked to stay home and self-isolate for 14 days, whereas those who had only been to Zone C were considered low-risk and free to move about as needed.

The way this tool works is by calculating the location of a given cellphone signal over a specified period of time. It then identifies the people who were near this individual, including those that live or work nearby or were visiting the designated area, based on their cellphone signals, providing the user with a list of people that may have been exposed to the virus if the individual in question tested positive for COVID-19.

For example, a user of this dashboard can choose a city that has had many confirmed cases of the virus and send an SQL script to a Kyligence OLAP cube. The Kyligence data service will then locate and identify anyone currently in that designated city who has traveled to areas that are known to have had an outbreak of the virus within the last month based on their cellphone signal’s location history. The high-risk individuals are then displayed as red dots.

The rapid response of China Mobile to this emergent health crisis was invaluable for those reacting to the situation. Having access to this information gave people critical information to keep their communities safe.

How Kyligence Monitored COVID-19

At the end of January 2020, families across China prepared to travel around the country to see their friends and families for the biggest holiday of the year, Chinese New Year. Leading into this holiday, cases of coronavirus spiked in China and the government responded by extending the holiday for an extra week to keep businesses closed and people safe at home.

During this extended holiday, Luke Han, CEO of Kyligence, set up a project to monitor the progress of the virus in China. He quickly pulled a team together and within 6 hours they had completed the provisioning of the entire cluster with Kyligence Cloud. Over the next 2-3 days, the team worked to prepare the data and build the report.

The team built a dashboard pulling from open source data that reported a daily breakdown of how many patients in China:

• Were confirmed to have the virus
• Were suspected to be infected
• Had been cured
• Had died

This data could be broken down and viewed by date, province, and city as shown in the two graphs below.

Our team of 6 included a cloud architect, two ETL developers, two data modelers and one data analyst. From the data they analyzed, they were able to derive a range of additional metrics regarding the amount of confirmed cases as well as confirmed death rates and cured rates:

• Per hospital, per hospital bed, per doctor, and per nurse
• Per million in the population
• Per square kilometer

Their analysis showed that Hubei Province was experiencing the highest pressure by far on medical resources based on the number of confirmed cases per million and the number of cases confirmed per hospital (see the two graphs below).

The team was also able to conclude that the tests to confirm if a person has the illness are administered and processed faster than the virus itself spreads. On February 5^th, the total suspected cases fell below the total confirmed cases (graph below). As soon as someone was suspected of being infected, they were able to receive the test and their results rapidly, allowing for timely isolation and prevention of further spreading of the virus.

The importance of testing and isolating positive cases has since been re-emphasized by communities battling this virus around the world. Most notably one small town in Italy, Vo Euganeo, the location of the first death in that country. The people of this town aggressively tested and isolated patients with COVID-19 to the point of eliminating the virus from their town completely.

This technology is key to our ability to collaborate globally on rapidly developing emergency situations. Having timely access to this data empowered the senior leadership at Kyligence to act quickly to protect employees and their families by implementing remote-work policies based on the knowledge gained from this analysis.

Thanks to the rapid and decisive acts of Kyligence’s leadership, not a single employee or member of their families has contracted the virus. This dashboard focused on China, but this technology could be used for any country or even on a global scale.

An Overview of Kyligence's Tracking Solution for COVID-19

Implications for the Future – How We Can Use Big Data to Prepare and React

These two Kyligence-powered dashboards that were built to monitor the status of this viral outbreak are only the tip of the iceberg when it comes to the potential applications for this technology on our ability to prepare for, and react more quickly to, the next global crisis.

With Kyligence’s technology, China Mobile was able to provide accessible data to the government, organizations and the general public that enabled them to respond efficiently and effectively to the epidemic (now pandemic) situation.

Kyligence provides a data service that can handle any size dataset so that analysts can do what they do best without needing to jump through hoops or wait minutes to hours between queries. The key to avoiding and surviving the next pandemic will be our ability to analyze data on a global scale, with teams from all countries able to utilize the same dashboards simultaneously.

This data service is capable of being accessed by millions, or billions, or people simultaneously. It provides all kinds of rich and standard interfaces including SQL, MDX and API so that any data scientist or analyst can perform self-service analysis as needed.

While Kyligence can build and provide the data visualization and reporting dashboards, our core product offering delivers a data service layer that can be consumed by many people, organizations or entire governments. In the case of a global health crisis like the sudden rise of COVID-19, there could potentially be billions of people accessing data at the same time.

That level of access requires very strong backend support with essentially non-existent query latency to ensure users are able to understand the data and act immediately. Leaders need to be able to quickly view and assess their data so they are able to make crucial decisions and act in the best interest of their families, their employees, and their citizens.

Many thanks to Joanna He, Dong Li, and Richard Zhou for their contributions to this article.

About the Author

Samantha Berlant is the Marketing Communications Manager at Kyligence and a big fan of AI, machine learning, and science-fiction. She spent several years leading content analytics projects at Facebook and Instagram and has been a writer and editor for over a decade. Samantha believes in the power of accessible data and her favorite Star Trek character is, coincidently, Data.