How We Can Use Social Media To Predict COVID Contact Tracing

--

Source

Countries like Singapore have been leading the charge in creating contact tracing applications. The app TraceTogether uses Bluetooth technology to try to detect when individuals are near each other and “make contact”. There’s also an app made in Israel that can detect if you’ve come into contact with someone infected with COVID-19.

The issue is that these apps are reactionary, they only help after someone’s been infected with COVID-19. Don’t get me wrong, these apps help a lot. But these apps don’t do much to actually prevent yourself from contracting the virus, they’re there to help make sure that people around you won’t be infected by you.

Tech needs to start taking a proactive stance against the virus

But to truly beat the COVID-19, we need to prevent as many infections as possible. To do this, tech can’t just be reactive, tech needs to start taking a proactive stance against the virus.

If we can predict where an infected person might go, or who they might meet up with. We can stop it from happening in the first place.

Pre-crime and Pre-infection

In the movie Minority Report, the world is built upon this concept of Pre-Crime. The idea is that the police in this world is so advanced, that they developed the technology to be able to predict crime before it even happened.

The Pre-Crime system was so accurate, that they used it to arrest murderers before they’ve even committed the murder that they would be arrested for. Quite simply, they could predict your future, and they’d arrest you for that future.

Now, this sounds like a fantasy, and you’d be mostly right. We’re many years away from having tech that’s accurate enough to predict your future completely accurately. But our current AI tech is good enough to make plausible guesses.

And we’re at the point now where AI can detect if you have depression based on how you use your phone.

Predicting Possible Infection

Photo by Christian Wiediger on Unsplash

If we want to reduce a person’s possibility of coming into contact with an individual infected with COVID-19, we need to be able to tell where an infected individual is likely going to go to and at what times.

This means we need to be able to tag individuals as infected and also learn from their habits and behaviors. Then we can use that to predict the possibility of them visiting certain locations.

To prevent infection, we don’t need to predict the future with absolute certainty, we just need a possibility.

If we can tell people, there’s a 70% chance that someone infected will go to this grocery, or go to this area, then that alone could stop some people from making the trip in the first place. If the infected person doesn’t end up going to that place, then it’s alright. What matters more is that if the infected person actually went, we prevented some at-risk individuals from going.

What looking for a McDonalds could look like

In this case, we don’t need to be as completely accurate like with Pre-Crime. To prevent infection, we don’t need to predict the future with absolute certainty, we just need a possibility.

All we need is to show users the possibilities and to let them know what the chances are that they might get infected if they go to certain places. We just need to show them the risk of going somewhere, and we’ll let them decide on whether or not they will take it

The Tech is Already Here

As far fetched as all this might sound, similar technologies are actually already being used by different companies, mainly by tech giants like Facebook and Google.

Both of these companies have enough behavioral data on you to predict which places you’ll go to and at which times. As weird as it seems if you can combine their massive amounts of data, you can use it to train an AI on predicting movement during COVID-19.

Trending AI Articles:

1. MS or Startup Job — Which way to go to build a career in Deep Learning?

2. TOP 100 medium articles related with Artificial Intelligence

3. Neural networks for algorithmic trading. Multimodal and multitask deep learning

4. Artificial Intelligence Conference

Google keeps track of your movements. Through Google Maps, it knows which restaurants you visit, where you like to go on Friday nights, and so on. It knows exactly where you’re going, and at what times.

Facebook looks at your messages to prevent abusive behavior. Through Messenger and Whatsapp, they can look at your messages to prevent people from sending things like spam. Of course, Facebook also tracks your behavior through your posts and your location through location tracking.

If you used Facebook’s power in tracking your messages and Google’s power in tracking location-based behavior, we can start to predict a user’s movements via the messages they send, and their normal day-to-day behaviors.

But of course, we haven’t accounted for changes in behavior due to the existence of COVID-19, and that’s where the Contact Tracing apps come in. With the Contact Tracing apps, we have a large data pool of each user’s movements during the coronavirus situation. We know exactly where they go, who they come into contact with, and at what times.

With these three pools of data, we can use it to train AI that we can use to predict the probability that you will come into contact with an infected individual.

Some of you may see this as absolutely terrifying, and you’d be obviously right to think so. Yes, companies collect this much data on you, and due to advances in data privacy laws like the GDPR, we’ve become more mindful about our data privacy.

But there is definitely an opportunity here. An opportunity to take a new approach towards combating viruses and pandemics. Tech like this could save millions of lives everywhere.

With this, we can start making new proactive measures on preventing the spread of disease and maybe even prevent future viruses from getting this far.

Innovation in transportation has increased the reach and strength of viruses, but innovation in artificial intelligence can help us push back against this formidable foe.

Don’t forget to give us your 👏 !

--

--