December 12, 2018 weblog
Tabs on Roger headed for coffee shop could spill ad-worthy beans
Location, location. Retailers see gold at the end of the location-tracking rainbow. They seek the advantage of targeting potential footsteps with relevant ads and offers. Questions: Where are you headed? When will you go offline? Facebook in patent filings proposes how they would get some answers.
Tyler Lee in Ubergizmo wrote on Tuesday that in a patent discovered by BuzzFeed News, Facebook has a patent discussion of how to predict where you might go in the future. The patent was filed last year in May. The title is "Offline Trajectories."
The technology that Facebook addressed would calculate a "transition probability based at least in part on previously logged location data associated with a plurality of users who were at the current location." In other words, the technology could also use the data of other people you know, as well as that of strangers, to make predictions, said BuzzFeed News.
Lee explained how this would work, based on current location and locations in the past, and people you know, it will try to predict where you might be next. Also, "According to the patent, this feature can be used to determine if where you're going might not have internet connectivity, and will try to prefetch data so that you can still access some content on Facebook."
Long and short, Sean Keane in CNET said that "Offline Trajectories" described tech that predicts where you're headed based on previously logged locations and those of others. Leaving work? Facebook would know if you are off to a store, not home.
Actually, according to BuzzFeed News, Facebook parked several patent applications, not just one, with the US Patent and Trademark Office for location-data technology, to predict where a person goes and when the person will be offline. Nicole Nguyen wrote about the application trio on Monday.
Another Facebook patent application titled "Location Prediction Using Wireless Signals on Online Social Networks" described how tracking the strength of Wi-Fi, Bluetooth, cellular, and near-field communication (NFC) signals could be used to estimate the person's location, to anticipate where next.
Mariella Moon in Engadget discussed that one, too. "Location Prediction Using Wireless Signals on Online Social Networks" is "for a technology that can use the strength of WiFi, Bluetooth, cellular and near-field communication (NFC) signals. If people of the same age as you typically visit a set of specific locations in your city in one outing, it can make a prediction based on that behavior. For example, if users with a similar profile usually go to a Starbucks next to the movie theater after seeing a movie, Facebook will anticipate that you're going to that Starbucks after watching a film, as well."
A Facebook patent application, "Predicting Locations and Movements of Users Based on Historical Locations for Users of an Online System," talks about how location data from multiple people would be used for location and movement trends, and to model location chains.
What would Facebook gain from such patent ideas if ever brought to light? Lee offered a speculation of how Facebook might leverage such technology. Lee wrote, "we wouldn't be surprised if such predictions could be used for advertising purposes, where if it detects you're in a restaurant and based on your penchant for desserts in the past, it might start to show you ads for desserts around you."
Nguyen also was on the mark with her assessment: "Location data is valuable. It could help Facebook prove to businesses that it's sending them lots of foot traffic, or it could improve Facebook's ad-targeting mechanism."
Companies have a now-familiar sermon when it comes to patent talk, something like "this is just exploration, folks, and it does not mean it will our product." Facebook was no exception. Reports quoted Facebook spokesperson Anthony Harrison. "We often seek patents for technology we never implement, and patent applications—such as this one—should not be taken as an indication of future plans."
© 2018 Science X Network