A constant motif in Facebook’s journey thus far is bringing content relevant to the user to the surface. For this they have relied on algorithms, and some trial and error, but it’s next step could totally change the way Facebook chooses to show you content.
The company is now working on a system that uses a simulated network of brain cells to process data in a more human way. According to a report in the MIT Technology Review , Facebook has internally put together a team of eight to form a research group known as the AI group. They’re working on a system called ‘deep learning’ that will help Facebook figure out posts that may not seem too pertinent to the algorithms but have a great personal value to the user.
Deep learning reportedly shows potential to enable software to do things such as work out emotions and events described in text. It will be able to do so even when the emotions are not explicitly referenced. Not just this, deep learning can also recognise objects in pictures and make sophisticated predictions about people’s likely future behaviour.
Facebook looking to humanise the News Feed system
Facebook’s Chief Technology Officer, Mike Schroepfer, said that one of the most important places where deep learning will be put to use is to improve the News Feed experience. Sophisticated algorithms are currently employed by Facebook in order to cull down the potential number of posts one could see from about 1,500 down to 30 or 60 posts. Using deep learning, the selection of posts that hit your News Feed will only end up getting better, with a more personal touch.
“The data set is increasing in size, people are getting more friends, and with the advent of mobile, people are online more frequently,” Schroepfer said “It’s not that I look at my news feed once at the end of the day; I constantly pull out my phone while I’m waiting for my friend, or I’m at the coffee shop. We have five minutes to really delight you.”
Another obvious place the AI can be useful is boosting its ad targeting. The potential this feature has to be able to pluck out users who will genuinely be interested in looking at and clicking your advertisements is huge.
While deep learning is only just kicking off, Srinivas Narayanan, an Engineering Manager at Facebook says that the AI group will be working on both applications hat can help the company’s products and on more general research on the topic that will be made public.