Facebook Now Using Natural Language Processing
An anonymous reader writes "Facebook has added a new type of story to its News Feed today: if more than one of your friends post about the same topic, and it has a Page on the social network, the posts will be grouped under a Posted About story, even if your friends don't explicitly tag the Page. It turns out Facebook is using natural language processing on status updates as well as the headlines of posted links to figure out if a topic mentioned has a corresponding Page, and then searches to see if your other friends have done so as well."
So, now that we can correlate what people are talking about to topics, can we get proper threading? The code doesn't work in some overloaded cases such as categorizing the phrase "Big Apple" under "Apple, Inc.".
I saw a bunch of posts mentioning Google being grouped together, which seemed silly.
Frozen Bubble.
Now if Facebook could only get posters to actually post in some sort of readable "natural language."
can Zuckerberg pull his hands out of his ass for a few minutes and allow image replies to posts? every bbcode html and *chan board has been able to do so since forever
Snowden and Manning are heroes.
Bonus points to whoever registers under the name Fruit Flies and goes on to like A Banana.
Bogtha Bogtha Bogtha
That makes it a keyword search. Natural Language Processing implies that it intelligently parses the text to locate the topic, this is Dumb Processing.
Anyone know why Facebook removed the default https security setting? Logging on is now insecure by default. You have to dig into your configuration options to reactivate it. Whats up with that?
If I and several of my friends have a discussion about LaTeX, I wonder which page will it get associated with. Or vice versa.
May result in some amusing situations.
AccountKiller
When did facebook receive exclusive rights to that name? Just because a movie was made about it under that name does not mean there is no other social network on the planet...
Damn_registrars has no butt-hole. Damn_registrars has no use for a butt-hole.
There are a lot of existing examples
Story: One toddler dead, another critical after house fire.
Ad: Burn, baby, burn!
The inevitability of inappropriateness... can ya feel it?
A lot of people...? Or am I in a different universe?
... until everybody who joined it discovered they need friends.
"I like to lick butts!" by MobileTatsu-NJG (#32700246) (Score:5, Informative)
psst... you're on LiveJournal.
"I like to lick butts!" by MobileTatsu-NJG (#32700246) (Score:5, Informative)
Most Facebook users I know don't use anything resembling natural language.
Daniel Ellsberg, Former State and Defense Dept. Official on Facebook and FBI :)
http://www.youtube.com/watch?v=VYqWabxYhSs
Domestic spying is now "Benign Information Gathering"
Reason to not use Facebook. Where is the assumption that I want to be grouped in with some off the cuff remark made by friend 1871. Who is that guy and why is he on my Facebook anyway? I should do some defriending soon.
It's beyond annoying- it will deter me from posting things I otherwise would, when the content is critical of the subject. Every time I tell my friends about the latest idiotic thing Sarah Palin or Donald Trump say and why only an ignoramus could believe that, it'll link to their pages. I don't want those halfwits to get credited as being relevant just because I share a laugh with my friends, and I certainly don't want to be related to her even if only by hyperlink.
What would be reasonable is If possibly-related-pages were recommended to me, where I could choose to link to a page of my choosing. I'd be glad to have that ability (as well as the option not to use it). This is of course regardless of who is implementing it- I wouldn't want Google forcing unwanted semantics on my content either. Luckily, Google at least recognizes that my content is actually mine, so there's some hope there.
How am I the only person upset about the issue at hand here? I don't want my posts being associated with any person, topic, or page, unless I INTEND for that to happen.
Example: There are a lot of drug groups on Facebook and if I use my account to criticize drugs, what if the NLP accidently groups myself and my friends and tags our posts as "About Citizens for Legal Marijuana" or some such? The point isn't whether the AI makes mistakes, the point is, how does it know my posts ought to be About anything at all? This should be a user choice, not something Facebook bots arbitrarily determine. Now I should worry that mention of a book might become implied endorsement of some big-budget summer movie on my part? This is ridiculous, I don't ever think Google+ will take over social networking, but Facebook keeps this up and sure enough people WILL leave.