The media update team explores how machine learning will improve location-based services.

Location-based services use real-time geographical data from mobile devices like a smartphones to provide the device user with useful information. In simple terms, these are apps and computer programs that give you directions, traffic updates, weather reports, and recommendations for nearby locations to visit.

Machine learning, a component of artificial intelligence, is being applied to these apps and programs to make the information they deliver more accurate and relevant.

Quick recap: Machine learning is when humans train machines to create or improve algorithms, based on historical data that the machines process.

There are many ways in which machine learning is making location-based apps and technology even better. Here are three that we found:

1. These apps will predict where you’re going next

Predictive analytics rely on techniques such as machine learning to find patterns within historical data. Location apps that use predictive analytics should be able to tell where you’ll want to go to, based on places you’ve been to in the past.

Apps will basically be able to suggest destinations and routes that are relevant to you – without the need for you to search for that information.

2. It will predict high-traffic areas

Machine learning enables navigation apps to predict how long it will take a traveller to get to their destination by taking complex factors like parking difficulty into account.

This is a feature that Google Maps launched in 2017. Google product manager Jeff Albertson writes that a parking difficulty icon was introduced to the Google Maps Android app in the United States. This feature predicts parking difficulty close to a destination so you can plan accordingly.

The engineers and scientists behind the project write that they used a unique combination of crowdsourcing and machine learning to build this system.

And this is only the first step, the developers say: “We’re excited about the opportunities to continue to improve the model quality based on user feedback. If we are able to better understand parking difficulty, we will be able to develop new and smarter forms of parking assistance.”

3. Your app will make better recommendations

Where you want to dine out or which hotel you want to stay at depends on a large number of factors, from preferences and previous experiences to your life stage and even the time of year. Location-based apps that rely on algorithms can recommend restaurants and accommodation based on a number of factors, but rarely on all of the above.

Machine learning allows a location-based app to learn from your previous movements across areas, and incorporates that knowledge with data that it has about your preferences and profile.

The benefit of learning from your daily movements, and not only from online data about you, is significant. Just because you like the Facebook Page of a business or brand, doesn’t always mean you prefer it, or that you’re interested in all its products. Your movements, from shop to shop and destination to destination, is the real proof of your preferences.

Michael Garvin, CEO at RoamingAround, explains the benefit of location data for personalisation: “It can supplement declared interests, spend events, and all other types of online data to grow exponentially better profiles over time. The patterns it identifies are more meaningful and representative of what the user really prefers.”

“Adding location to the equation replaces assumptions with accurate conclusions.”

The applications of machine learning in location-based services are set to grow as more data becomes available and machine learning becomes more mainstream.

From there, the sky’s the limit: “Mr Jones, there is heavy traffic on your route. Would you like me to book you a driverless taxi drone?”
Location-based services is not the only field that will be affected by AI. Find out which other sectors will benefit from this technological advance in our article, Three industries AI will change in 2018.
Image designed by Freepik