Taylor Swift and Machine Learning

What do Taylor Swift, Atlanta US airport and London police have in common?

Sure, Atlanta Airport has little ‘Blank Space’ in its busy air traffic schedule, but there’s more to it than that.

When we discuss facial recognition technologies, we immediately think to futuristic scenarios or some of Netflix’s Black Mirror episode, but this technology is now real and being used. In December 2018 alone we saw 3 major examples of facial recognition applied:

• Taylor Swift used it to spot stalkers during her concerts
• Atlanta US airport used it to allow quicker check-in and security checks
• London police used it to detect people wanted by the courts

To simplify, this technology is recording multiple factors that make your (almost) unique face, creating a 3D version of your face and categorizing it into recurrent patterns.

To manage those patterns, Machine Learning has been introduced and now this technology is able to overcome the four main challenges related to facial recognition: illumination, pose, aging and emotion with an accuracy of 97%.

As you may know, the key success factor for Machine Learning is the size of the data set available. Having a large data set of pictures, videos, recordings and selfies will increase the facial recognition accuracy by matching (or mismatching with false positives) your photo with a biometric database. If it matches, thumbs up, otherwise thumbs down and it learns from its mistakes.

At Aspire Technology, we’re applying Machine Learning algorithms to simplify, automate, and make our operations more efficient. Using those algorithms, we are able to run complex network activities in parallel and optimize network performances.

Thanks to Machine Learning we are also minimizing human errors, reducing operational costs and avoiding repetitive tasks assigned to our engineers.

Aspire Technology with Machine Learning allows mundane network activities to evolve into smarter operations, just as facial recognition with Machine Learning increases your personal security.

Security and smarter living are just two examples of Machine Learning applications and IT giants like Facebook, Google and Amazon are currently investigating how to apply this technology to other areas such as personalized advertising, efficient e-commerce or even predicting what type of burger you will order at McDonald’s.

Unfortunately, it comes at a cost, and facial recognition can cost something priceless: your privacy. In fact, this technology is currently facing huge backlash and criticism by privacy campaigners, but just as many people are accepting it and willingly sacrificing some privacy in exchange for increased or perceived security.

What’s your view, and in what situation would you agree to allow facial recognition?