Facebook looking for meaning in user posts with 'deep learning' algorithms

Sep 23, 2013 by Bob Yirka weblog

(Phys.org) —Officials at Facebook have apparently decided to get serious about making sense of posts by its vast user base—according to MIT's Technology Review, officials with the company (specifically Chief Technical Officer Mike Schroepfer) have announced that they have put together a team of eight professionals with the mission of developing what the software industry has begun calling "deep learning." Deep learning is a type of software programming where algorithms are created that allow for building simulated neural networks. Such neural networks are capable of "learning" by analyzing patterns over time. Facebook, TR reports, is hoping to use its algorithms to better target ads, and also to improve its newsfeed.

As anyone who uses Facebook knows, friending people means adding their posts to your personal newsfeed. As the number of friends grows, so too does the number of newsfeed entries. Eventually, a point is reached where it becomes untenable. To deal with this problem, Facebook has created algorithms that are meant to pick out what it believes are the most relevant posts and only send those to the newsfeed, rather than deliver them all. Thus far, this approach has met with mixed reviews from its user community. The company is hoping that giving its algorithms more smarts, will help improve the quality of newsfeeds.

Deep learning isn't something Facebook created, it's an idea that has been around for several years, Microsoft and Google have used it (Google most famously to help identify cats in YouTube videos). IBM uses it too, as part of Watson, the supercomputer that beat Jeopardy champions on television. Netflix is experimenting with similar technology to help improve its movie suggestion algorithms and might serve as a guide for how successful Facebook might expect to be with its initiative (movie suggestions are still not all that relevant). But perhaps, that's not the point at all. Instead, maybe it's the process that is the real news. Big companies are starting to spend a lot of money on with the idea of a big payoff. Surely Facebook would be happy if they could improve their newsfeed, but the real motive, as with any big business, will always be about improving the bottom line. And if deep learning algorithms can figure out a way to coax users into clicking on more ads, than the initiative will most certainly be deemed a success by all concerned.

Explore further: Study examines viral reach of hashtags on Facebook

More information: via TechReview

Related Stories

Study examines viral reach of hashtags on Facebook

Sep 05, 2013

(Phys.org) —Facebook analysts EdgeRank Checker have performed analysis of over 35,000 posts to discover the level of viral impact of Facebook's hashtags, The discovery is, well, zero. "We decided to dig ...

Recommended for you

Simplicity is key to co-operative robots

4 hours ago

A way of making hundreds—or even thousands—of tiny robots cluster to carry out tasks without using any memory or processing power has been developed by engineers at the University of Sheffield, UK.

Freight train industry to miss safety deadline

4 hours ago

The U.S. freight railroad industry says only one-fifth of its track will be equipped with mandatory safety technology to prevent most collisions and derailments by the deadline set by Congress.

IBM posts lower 1Q earnings amid hardware slump

6 hours ago

IBM's first-quarter earnings fell and revenue came in below Wall Street's expectations amid an ongoing decline in its hardware business, one that was exasperated by weaker demand in China and emerging markets.

Microsoft CEO is driving data-culture mindset

6 hours ago

(Phys.org) —Microsoft's future strategy: is all about leveraging data, from different sources, coming together using one cohesive Microsoft architecture. Microsoft CEO Satya Nadella on Tuesday, both in ...

User comments : 0

More news stories

Microsoft CEO is driving data-culture mindset

(Phys.org) —Microsoft's future strategy: is all about leveraging data, from different sources, coming together using one cohesive Microsoft architecture. Microsoft CEO Satya Nadella on Tuesday, both in ...

Floating nuclear plants could ride out tsunamis

When an earthquake and tsunami struck the Fukushima Daiichi nuclear plant complex in 2011, neither the quake nor the inundation caused the ensuing contamination. Rather, it was the aftereffects—specifically, ...

Patent talk: Google sharpens contact lens vision

(Phys.org) —A report from Patent Bolt brings us one step closer to what Google may have in mind in developing smart contact lenses. According to the discussion Google is interested in the concept of contact ...

Quantenna promises 10-gigabit Wi-Fi by next year

(Phys.org) —Quantenna Communications has announced that it has plans for releasing a chipset that will be capable of delivering 10Gbps WiFi to/from routers, bridges and computers by sometime next year. ...