The business of fake news
Dragon
58359 cr points
Send Message: Send PM GB Post
37 / M
Online
Posted 13 days ago
So, growing up, I always knew there was fake news out there. The supermarket tabloids got a chuckle with the aliens landing on the White House lawn and such.

But after reading http://www.npr.org/sections/alltechconsidered/2016/11/23/503146770/npr-finds-the-head-of-a-covert-fake-news-operation-in-the-suburbs?utm_campaign=storyshare&utm_source=twitter.com&utm_medium=social, I have to admit, I'm pondering a shift in careers. Fake news has really taken off in a big way!

Even with Facebook trying to crack down on it, and Google pulling their ads from the guy in the article, still seems extremely lucrative. Guy in the article even explains how that works out for him, thanks to all the other ad companies out there.

...

.. honestly, no, I wouldn't do that. This isn't a conservative, liberal, or anything else issue. This is people making money taking advantage of others' fears. It's people making fake memes, making up news stories that sound just possible enough when you already have a low opinion of whom the story is about, and things of that nature. Honestly, I hate that kind of thing.

Buuuuuut... as a computer programmer, I have a separate interest in this, which is what I really want to talk about here. What can be done to detect fake news? A few ideas I had off the top of my head:

A) Human training. Basically, this is similar to how some image detection algorithms can work. Human looks at thing, computer looks at thing, and human corrects the computer when it's incorrect. The programmers develop a heuristic which the program adapts thanks to input, and with thousands or millions of users (thank you, internet), it gets better at detecting what's fake and can shuffle it off to a bin until a human can give the final word before it goes up for folks to get riled up about.

B) Location trustworthiness. From what I've been reading (heh, can I trust that?), much of the fake news originates from specific locations, such as the guy in the linked article. If an article has key phrases or links that can be traced back to such a location, mark it down on the trust scale. I don't like this approach as much - mostly because as soon as it went live, fake news would figure out how to mask their location. Perhaps a modified version of this, using it to capture those key phrases to match against other news to assist in the training from above.. hmm.

C) .. profit? If a business seems to be making a lot of money in news, but isn't actually putting reporters in the field and such, I suppose that seems a bit suspicious, though with blogs that just collect and republish/repackage real news, that may be difficult to accurately detect.

Anyway, what do you think? Is there anything that can be done on the automated assist side of things to help fight fake news? Or is it all up to us puny humans to deal with it?
3889 cr points
Send Message: Send PM GB Post
20 / M
Offline
Posted 13 days ago , edited 13 days ago
I think it's hilarious and sad. I like tricksters who if you see past their tricks you can gain insight, but when they profit from it on both a monetary level and psychic vampiric level I don't find it as appealing. As for the to help fight it using programming I have no clue. It would be best if we dealt with it ourselves, if anything these fake news sites could be used as training to help ourselves look at information better or something like that.

I actually wouldn't mind something like this blowing up to the point where it starts forcing people to take steps to not be deceived.
2080 cr points
Send Message: Send PM GB Post
30 / F
Offline
Posted 13 days ago , edited 13 days ago
Maybe something that could check the website using a list of legitimate sources, but it wont stop the majority of online tabloid and conspiracy theorist from seeking them out. There has always been a lot of cases of using propaganda for vile intentions like the Nazis, or to manipulate others into a protectionist state of mind and manipulate them using fear like early Christianity's use of Apocalyptic preaching to keep their flock from converting to other religions and ritual practices of guilds and secret societies. Fox and AM talk radio are filled to the brim fear mongering and scapegoating against all sorts, and are wildly popular with what was the civil rights generation. Now they are almost the opposite. The older they get and the more they listen too: it becomes more and more likely that I will hear about "the evil gay-agenda" and that "Muslims want to kill us all". The truth may be stranger than fiction, but America has snapped out of it before and will again. People will have to accept facts again, eventually.
5831 cr points
Send Message: Send PM GB Post
Offline
Posted 13 days ago , edited 13 days ago
well ya some of those are fake sources that could be legitimately considered fake i do think that some are at more accurate about some subjects then official news sources and are more journalist/people who are truly trying to give accurate news and when official news organizations start to not tell the truth regularly for political gain or something yet smaller for example YouTube channels or such that are trying to. start becoming considered more trust worthy then them what do you the response would be the official news organizations maybe at least a little jealous and try and stomp out the competition or such. now i try and listen to the news to gain some perspective but a decent amount of the time their does seem to be inaccuracies in how they tell it or avoid subjects that really could use more discussion or such.
24273 cr points
Send Message: Send PM GB Post
27 / M
Online
Posted 13 days ago
If there is a market for it I don't really care.
39056 cr points
Send Message: Send PM GB Post
26 / M / Your friendly nei...
Offline
Posted 13 days ago
Breaking News: I ignore all news
16795 cr points
Send Message: Send PM GB Post
Hoosierville
Offline
Posted 13 days ago , edited 13 days ago
Block CNN, MSNBC nytimes, and washington post to see a 33% decrease in fake news.

Really this isn't an attempt to stop news. Its an attempt at censorship disguised as protecting the people. They want to block news they don't approve of or are from sources they dislike. This whole thing is a direct result of all the corruption of Hillary trending on every site only to be deleted and banned from the trending list.

What's going to happen is they will block news sources they dislike and only allow ones they like so they can control how people think. Trump literally won because of social media and the internet, this is an attempt to plug that hole since they already have control of the MSM



Heck the reddit CEO already admitted to manipulating the voting system to stop pro-trump messages. They modified the algorithm to prevent Trump messages from showing up but we got so big it didn't work properly http://archive.is/ZmULb
2009 cr points
Send Message: Send PM GB Post
Offline
Posted 13 days ago
how do they determine fake news cnn and other news outlets plus all pollsters said trump had no chance at winning and he did so are they considered people who use fake new then
1582 cr points
Send Message: Send PM GB Post
Offline
Posted 13 days ago , edited 13 days ago
fake news are all over the place now a day..

it's best to stick to those independent journalists who will provide facts or videos to back up their claims.. especially ones with no connection to the clinton news network

objective journalism is rare now a day..


huffington post took the cake on this one

but what do you expect from a group of sexist/racist elitists?

look at their board meeting.. no minority and no man-- not to mention there's an age limit or something as well ?

so it's discrimination based on sex/gender, race and age





2631 cr points
Send Message: Send PM GB Post
17 / F / Canada
Offline
Posted 13 days ago
tbh i would still ignore
Dragon
58359 cr points
Send Message: Send PM GB Post
37 / M
Online
Posted 12 days ago , edited 12 days ago

Rujikin wrote:

Block CNN, MSNBC nytimes, and washington post to see a 33% decrease in fake news.

Really this isn't an attempt to stop news. Its an attempt at censorship disguised as protecting the people. They want to block news they don't approve of or are from sources they dislike. This whole thing is a direct result of all the corruption of Hillary trending on every site only to be deleted and banned from the trending list.

What's going to happen is they will block news sources they dislike and only allow ones they like so they can control how people think. Trump literally won because of social media and the internet, this is an attempt to plug that hole since they already have control of the MSM



Heck the reddit CEO already admitted to manipulating the voting system to stop pro-trump messages. They modified the algorithm to prevent Trump messages from showing up but we got so big it didn't work properly http://archive.is/ZmULb


So, an article which has a person straight up admitting to generating fake news in order to make money by having folks who are afraid spread it leads to.. talking up more fear?

I do find it interesting that you started your post talking about blocking news services as a good thing in your opinion, and finished it with talking about others blocking news services as a bad thing. Which, you'll notice, isn't something I proposed in my OP. I was looking for ways of detecting fake news, not just saying "XYZ service should be cut off".

As for reddit, I avoid that place, personally. A wretched hive of scum and villainy. I'd suggest staying away, though merely as a suggestion. It's not news, anyway, even if people post news there occasionally. Kind of like 4chan in my mind.
24273 cr points
Send Message: Send PM GB Post
27 / M
Online
Posted 12 days ago

MakotoKamui wrote:


Rujikin wrote:

Block CNN, MSNBC nytimes, and washington post to see a 33% decrease in fake news.

Really this isn't an attempt to stop news. Its an attempt at censorship disguised as protecting the people. They want to block news they don't approve of or are from sources they dislike. This whole thing is a direct result of all the corruption of Hillary trending on every site only to be deleted and banned from the trending list.

What's going to happen is they will block news sources they dislike and only allow ones they like so they can control how people think. Trump literally won because of social media and the internet, this is an attempt to plug that hole since they already have control of the MSM



Heck the reddit CEO already admitted to manipulating the voting system to stop pro-trump messages. They modified the algorithm to prevent Trump messages from showing up but we got so big it didn't work properly http://archive.is/ZmULb


So, an article which has a person straight up admitting to generating fake news in order to make money by having folks who are afraid spread it leads to.. talking up more fear?

I do find it interesting that you started your post talking about blocking news services as a good thing in your opinion, and finished it with talking about others blocking news services as a bad thing. Which, you'll notice, isn't something I proposed in my OP. I was looking for ways of detecting fake news, not just saying "XYZ service should be cut off".

As for reddit, I avoid that place, personally. A wretched hive of scum and villainy. I'd suggest staying away, though merely as a suggestion. It's not news, anyway, even if people post news there occasionally. Kind of like 4chan in my mind.


I've heard some of the bigger news companies put out exaggerated or sometimes intentionally misleading stories just so they can see the subject of the article deny the allegations. Then redact it a week later in fine print no one will ever see. This fake news boom is that practice's final form. A natural progression from "My sources in the _____ tell me _____".

"Mainstream" media did this to themselves. The current uproar could just be salt that social media has disrupted their monopoly on mindshare. NY Times, WaPo and CNN are just more coy and clever than your average internet troll. For a lot of people perception is reality and their might be a real danger of violence springing up as a result of fake news. However, we have free speech so their is the same danger from the truth.
31145 cr points
Send Message: Send PM GB Post
30 / M
Offline
Posted 12 days ago

MakotoKamui wrote:

A) Human training. Basically, this is similar to how some image detection algorithms can work. Human looks at thing, computer looks at thing, and human corrects the computer when it's incorrect. The programmers develop a heuristic which the program adapts thanks to input, and with thousands or millions of users (thank you, internet), it gets better at detecting what's fake and can shuffle it off to a bin until a human can give the final word before it goes up for folks to get riled up about.

Hmm.. How do we go about doing that though?

I mean, human training for image detection works because the human is teaching the computer how to do things the human brain is good at. The problem with fake news is that a significant number of people aren't good at recognizing it.

So to go with the image detection analogy again - it would be like trying to teach a computer to reliably recognize pictures of flowers when 40% of the humans training it are misidentifying tulips as ducks.

This is before you even get into all the fuzzy areas where the fake news is sort of true, or partially true, or true factually but being misrepresented so as to lead readers to conclusions that aren't true, etc. There's a lot more subjectivity in "truth" than in the contents of a photograph.
Emtro 
1563 cr points
Send Message: Send PM GB Post
Offline
Posted 12 days ago

MakotoKamui wrote:

So, growing up, I always knew there was fake news out there. The supermarket tabloids got a chuckle with the aliens landing on the White House lawn and such.

But after reading http://www.npr.org/sections/alltechconsidered/2016/11/23/503146770/npr-finds-the-head-of-a-covert-fake-news-operation-in-the-suburbs?utm_campaign=storyshare&utm_source=twitter.com&utm_medium=social, I have to admit, I'm pondering a shift in careers. Fake news has really taken off in a big way!

Even with Facebook trying to crack down on it, and Google pulling their ads from the guy in the article, still seems extremely lucrative. Guy in the article even explains how that works out for him, thanks to all the other ad companies out there.

...

.. honestly, no, I wouldn't do that. This isn't a conservative, liberal, or anything else issue. This is people making money taking advantage of others' fears. It's people making fake memes, making up news stories that sound just possible enough when you already have a low opinion of whom the story is about, and things of that nature. Honestly, I hate that kind of thing.

Buuuuuut... as a computer programmer, I have a separate interest in this, which is what I really want to talk about here. What can be done to detect fake news? A few ideas I had off the top of my head:

A) Human training. Basically, this is similar to how some image detection algorithms can work. Human looks at thing, computer looks at thing, and human corrects the computer when it's incorrect. The programmers develop a heuristic which the program adapts thanks to input, and with thousands or millions of users (thank you, internet), it gets better at detecting what's fake and can shuffle it off to a bin until a human can give the final word before it goes up for folks to get riled up about.

B) Location trustworthiness. From what I've been reading (heh, can I trust that?), much of the fake news originates from specific locations, such as the guy in the linked article. If an article has key phrases or links that can be traced back to such a location, mark it down on the trust scale. I don't like this approach as much - mostly because as soon as it went live, fake news would figure out how to mask their location. Perhaps a modified version of this, using it to capture those key phrases to match against other news to assist in the training from above.. hmm.

C) .. profit? If a business seems to be making a lot of money in news, but isn't actually putting reporters in the field and such, I suppose that seems a bit suspicious, though with blogs that just collect and republish/repackage real news, that may be difficult to accurately detect.

Anyway, what do you think? Is there anything that can be done on the automated assist side of things to help fight fake news? Or is it all up to us puny humans to deal with it?


Your questions are interesting but I would stop to examine the article a little closer. Aside form the strange name "Jestin Coler" I noticed something wrong with the story. When they confronted Mr. Coler, after they identified him, and asked him about "Disinfomedia" he said "I don't know what to tell you" and they just gave up. Even though they somehow managed to leave their information with him in case he changed his mind, it doesn't sound plausible that such an in depth investigation would end right there. Further on that point, it doesn't make sense that they would just accept his story that he gave them after an unsaid number of hours had passed. So, I googled Jestin Coler and found this article that proposes a different explanation that of matches the other tricks the DNC used to distract from the Wikileaks exposure.

https://willyloman.wordpress.com/2016/11/24/meet-npr-poster-boy-for-the-fake-news-crisis-jestin-coler-an-obama-administration-cognitive-infiltration-asset/

The author, Scott Creighton, says he was asked to write about the story that the NPR article was based on (the Denverguardian.com murder suicide story) but refused after he looked into it and found it was bogus. The author proposes the bogus story was to distract and discredit this story:

http://www.thegatewaypundit.com/2016/08/wow-breaking-video-julian-assange-suggests-seth-rich-wikileaks-dnc-source-shot-dead-dc/

The article mentions that social media had been circulating a story about him going to meet FBI agents but there is no way to tell if that's true or not. The meat of the story is the part with Julian Assange, which are some rather cryptic hints. He very specifically points to Rich but doesn't confirm if he was a source or not, just that it may mean more of his sources could end up like Rich...

All in all, I would not trust that the source of the NPR story was being honest. Instead I would suggest Coler got instructions on what to tell the NPR reporters and NPR accepted the story as is because of other influences. It's not the first time I've seen NPR blatantly influenced by the current administration.
Dragon
58359 cr points
Send Message: Send PM GB Post
37 / M
Online
Posted 12 days ago

Emtro wrote:


MakotoKamui wrote:

So, growing up, I always knew there was fake news out there. The supermarket tabloids got a chuckle with the aliens landing on the White House lawn and such.

But after reading http://www.npr.org/sections/alltechconsidered/2016/11/23/503146770/npr-finds-the-head-of-a-covert-fake-news-operation-in-the-suburbs?utm_campaign=storyshare&utm_source=twitter.com&utm_medium=social, I have to admit, I'm pondering a shift in careers. Fake news has really taken off in a big way!

Even with Facebook trying to crack down on it, and Google pulling their ads from the guy in the article, still seems extremely lucrative. Guy in the article even explains how that works out for him, thanks to all the other ad companies out there.

...

.. honestly, no, I wouldn't do that. This isn't a conservative, liberal, or anything else issue. This is people making money taking advantage of others' fears. It's people making fake memes, making up news stories that sound just possible enough when you already have a low opinion of whom the story is about, and things of that nature. Honestly, I hate that kind of thing.

Buuuuuut... as a computer programmer, I have a separate interest in this, which is what I really want to talk about here. What can be done to detect fake news? A few ideas I had off the top of my head:

A) Human training. Basically, this is similar to how some image detection algorithms can work. Human looks at thing, computer looks at thing, and human corrects the computer when it's incorrect. The programmers develop a heuristic which the program adapts thanks to input, and with thousands or millions of users (thank you, internet), it gets better at detecting what's fake and can shuffle it off to a bin until a human can give the final word before it goes up for folks to get riled up about.

B) Location trustworthiness. From what I've been reading (heh, can I trust that?), much of the fake news originates from specific locations, such as the guy in the linked article. If an article has key phrases or links that can be traced back to such a location, mark it down on the trust scale. I don't like this approach as much - mostly because as soon as it went live, fake news would figure out how to mask their location. Perhaps a modified version of this, using it to capture those key phrases to match against other news to assist in the training from above.. hmm.

C) .. profit? If a business seems to be making a lot of money in news, but isn't actually putting reporters in the field and such, I suppose that seems a bit suspicious, though with blogs that just collect and republish/repackage real news, that may be difficult to accurately detect.

Anyway, what do you think? Is there anything that can be done on the automated assist side of things to help fight fake news? Or is it all up to us puny humans to deal with it?


Your questions are interesting but I would stop to examine the article a little closer. Aside form the strange name "Jestin Coler" I noticed something wrong with the story. When they confronted Mr. Coler, after they identified him, and asked him about "Disinfomedia" he said "I don't know what to tell you" and they just gave up. Even though they somehow managed to leave their information with him in case he changed his mind, it doesn't sound plausible that such an in depth investigation would end right there. Further on that point, it doesn't make sense that they would just accept his story that he gave them after an unsaid number of hours had passed. So, I googled Jestin Coler and found this article that proposes a different explanation that of matches the other tricks the DNC used to distract from the Wikileaks exposure.

https://willyloman.wordpress.com/2016/11/24/meet-npr-poster-boy-for-the-fake-news-crisis-jestin-coler-an-obama-administration-cognitive-infiltration-asset/

The author, Scott Creighton, says he was asked to write about the story that the NPR article was based on (the Denverguardian.com murder suicide story) but refused after he looked into it and found it was bogus. The author proposes the bogus story was to distract and discredit this story:

http://www.thegatewaypundit.com/2016/08/wow-breaking-video-julian-assange-suggests-seth-rich-wikileaks-dnc-source-shot-dead-dc/

The article mentions that social media had been circulating a story about him going to meet FBI agents but there is no way to tell if that's true or not. The meat of the story is the part with Julian Assange, which are some rather cryptic hints. He very specifically points to Rich but doesn't confirm if he was a source or not, just that it may mean more of his sources could end up like Rich...

All in all, I would not trust that the source of the NPR story was being honest. Instead I would suggest Coler got instructions on what to tell the NPR reporters and NPR accepted the story as is because of other influences. It's not the first time I've seen NPR blatantly influenced by the current administration.


And wordpress is fascinating. I've seen lots of nifty furry sites there. See? It's easy to discredit sources, and has nothing to do with the point of my thread, which was - can we / how can we detect fake news, or is it all up to us puny humans?

At this point, very few responses have been about that. Since it's not generating discussion, I'm closing the thread. I asked the OP, and he/I just said that was cool.
You must be logged in to post.