Researchers Develop Faster Way To Replace Bad Data With Accurate Information (ncsu.edu) 58
sandbagger writes: Researchers from North Carolina State University and the Army Research Office have demonstrated a new model of how competing pieces of information spread in online social networks and the Internet of Things (IoT). The findings could be used to disseminate accurate information more quickly, displacing false information about anything from computer security to public health. "Whether in the IoT or on social networks, there are many circumstances where old information is circulating and could cause problems -- whether it's old security data or a misleading rumor," says Wenye Wang, co-author of a paper on the work and a professor of electrical and computer engineering at NC State. "Our work here includes a new model and related analysis of how new data can displace old data in these networks."
"Ultimately, our work can be used to determine the best places to inject new data into a network so that the old data can be eliminated faster," says Jie Wang, a postdoctoral researcher at NC State and first author of the paper. In their paper, the researchers show that a network's size plays a significant role in how quickly "good" information can displace "bad" information. However, a large network is not necessarily better or worse than a small one. Instead, the speed at which good data travels is primarily affected by the network's structure.
"Ultimately, our work can be used to determine the best places to inject new data into a network so that the old data can be eliminated faster," says Jie Wang, a postdoctoral researcher at NC State and first author of the paper. In their paper, the researchers show that a network's size plays a significant role in how quickly "good" information can displace "bad" information. However, a large network is not necessarily better or worse than a small one. Instead, the speed at which good data travels is primarily affected by the network's structure.
Wont stop human natural stupidity (Score:4, Insightful)
Seems like a reasonable approach rather than censorship, but it wont stop natural stupidity putting out flawed information in the first place.
Re: Wont stop human natural stupidity (Score:4, Insightful)
There really is no difference between "good" and "bad" data. This research cuts both ways, you can use it for propaganda or to fight propaganda, depends on your perspective and goals, bits are bits.
Re: (Score:1)
Sounds like a good way to clog the tubes. Replacing all this "bad" data will take some serious bandwidth.
Re: (Score:2)
Re: (Score:3)
You don't necessarily want old data and information removed simply because it is old.
We've always been at war with eastasia!
Re: (Score:1)
Re: (Score:1)
Now we need Ministry of Truth (Score:4, Funny)
All we need now is the Ministry of Truth.
Once we know the Current Truth[tm], we can disseminate it efficiently.
Re: (Score:2)
They had no concept just how efficient the memory-holes could be.
Re: (Score:3)
You don't need to know what the absolute truth is, merely what is objectively a lie.
"Drinking bleach cures coronavirus."
"Vaccines cause autism."
"Vaccines don't work."
"There is no scientific consensus on climate change."
These are all objectively lies. You don't need to replace them with 'the truth', because of course the truth is more nuanced than any single statement that you can make. But you can disabuse people of these notions, and there are no moral judgements implicit in refuting these statements, beca
Re: Now we need Ministry of Truth (Score:1)
Depends on your perspective and what you consider a lie vs your source of truth.
Bleach does cure a lot of diseases, drink a gallon and you won't ever get sick again. That is a true statement backed by scientific evidence.
Re: (Score:2)
That is a ridiculous statement, and we all know it is. Death is good for what ails you, as the saying goes, but it's always said in jest. Anyone that takes it seriously is not a person worth talking to.
Re: (Score:3)
This last one is not so much a lie, it simply makes no sense — because Climate Science simply is not science, there can be no scientific consensus about it.
Moreover, by throwing it together with the Science of Medicine — during an epidemic — you are endangering people. Rightly distrusting the Climate "Science" alarmism, they begin to distrust the medical information too, because you're tainting it with your
Re: (Score:2)
because Climate Science simply is not science, there can be no scientific consensus about it.
You mean hypotheses that cant be invalidated arent theories no matter how much data you gather?
Sounds about right.
Eisenhower warned us about the "scientific-technological elite" capturing policy in the same speech as the well known "military-industrial complex" warning about that capturing policy. What he didnt realize, I think, is that both can coexist, even so far as bolstering and feeding off each other.
Re: (Score:2)
Yes, the proper term is "(non)-falsifiable".
A statement has to be falsifiable [wikipedia.org] to be scientific, which automatically excludes any and all predictions containing "may" or "can" instead of "will", for example.
Geico's commercials are hysterically unfalsifiable ("... could save you up to 15% or more") — and the predictions we see in the press, /. included, are all from the same category.
Funnily enough, even the "
Some of it is falsifiable (Score:3)
Some of it's falsifiable, just time-consuming to falsify.
For example the UN report saying that if we don't reduce C02 emissions drastically, certain cities in California will be underwater within 30 years. That's a falsifiable claim, if you wait long enough.
It's been about 30 years since they made that claim, so now it's not only falsifiable, but falsified - San Francisco is in fact not underwater.
So some of them are falsifiable "eventually*. And most of those have turned out to be false. Which isn't too
Re: (Score:1)
Yes, earlier predictions were falsifiable, but — as they gradually got falsified — the "scientists" became more careful, and the "may" started replacing the "will"... You'd be hard-pressed to find a prediction today, that's not like the Geico commercial I quoted...
The
Re: (Score:1)
Old geezer that I am came of age in the Population Bomb scare days. The population bomb stuff scared us really bad but turned out to be all full of shit. Someday people are going to quit listening.
Re: (Score:2)
In context, Eisenhower was concerned that the practice science (which he supported) could be corrupted by the government by promoting a "scientific-technological elite" that does the government's bidding.
Akin to, and largely responsible for the sweeping changes in our industrial-military posture, has been the technological revolution during recent decades. In this revolution, research has become central, it also becomes more formalized, complex, and costly. A steadily increasing share is conducted for, by, or at the direction of, the Federal government. ...
The prospect of domination of the nation's scholars by Federal employment, project allocation, and the power of money is ever present and is gravely to be regarded. Yet in holding scientific discovery in respect, as we should, we must also be alert to the equal and opposite danger that public policy could itself become the captive of a scientific-technological elite
TL/DR: Eisenhower was concerned that policy would capture science, not the other way around.
Re: (Score:2)
Oh wait. Damn, forget what I just said. I just realized that you're right. He was concerned about science capturing policy. I should read more carully before I post.
"objectively" (Score:1)
Oh man ... you have NO idea how deep that philosophical rabbit hole is, you held open there.
Otherwise you would never use that word.
After . . .
* the relativity of spacetime,
* the distortion of our senses,
* our brains being literally bias machines for which objectivity is physically impossible,
* and almost all of what we social lifeforms think we know, being mere unverified hearsay (Did you verify if that peer-reviewed six sigma double-blind study is true or the peers are right? With your own observed realit
Re: (Score:1)
"I don't demand that a theory correspond to reality because I don't know what it is. Reality is not a quality you can test with litmus paper. All I'm concerned with is that the theory should predict the results of measurements."
--Stephen Hawking
Re: (Score:2)
Yes, I'm aware of all of this, but it's all functionally irrelevant in day to day life. The only way we can make our way through the universe is by believing that the keyboards that we're typing at are, in fact, real. Frankly, the only internally consistent philosophy is pure solipsism and deciding that nobody else exists since you can't verify the consciousness of other people, but it's not a useful way to live your life. We're not concerned with philosophy here, despite how fun it is to think about (and I
Re: (Score:2)
But you can disabuse people of these notions
I think 1984 referred to that as Room 101.
Re: (Score:2)
You forgot one.
"Incorrect information can always be corrected by consulting Slashdot, which always provides the ultimate version of the truth.
"Except of course when Slashdotters contradict each other and call each other names..."
Re: (Score:2)
So we should choose a different traitor?
This is just not true, or too simplistic... (Score:4, Interesting)
From the linked story:
In their paper, the researchers show that a networkâ(TM)s size plays a significant role in how quickly âoegoodâ information can displace âoebadâ information. However, a large network is not necessarily better or worse than a small one. Instead, the speed at which good data travels is primarily affected by the networkâ(TM)s structure.
Ok, right away consider Twitter. A massive network that can spread information super quickly.
Where the model is too simplistic, is that in Twitter each node in the large network can decide to pass information, and very very often that network will spread "invalid" information rapidly across the globe, while any kind of followup with gather but a small percentage of the same spread.
Twiter at least seems to operate as a large but very gated network, where only information conforming to a certain ideology has access to the full network - otherwise there is significant action to block competing information from spreading, even if it's in fact correct or "good" information.
So that is nothing about the structure of the network, but instead it comes down to behaviors of the nodes. Even if you were not talking about Twitter, in pretty much any network nodes will always end up acting as some kind of gates to decide what to spread further in the network, and thus the same effect could easily be seen in any larger network of any kind...
So where the study falls short is in not realizing that any network will not be similarly reachable for all forms of information.
People spread what they want to be true. (Score:1)
And are quiet when they don't.
*
And nothing in the word is gonna change that.
Least of all, *forcing them*, as every "solution" so far attempts.
Here, the solutiom is to accept that, and step outside of that mind box.
_ _ _ _
* The only exception is when they bounce from one extreme to another extreme, like stongly anti-gay priests being caught in a gay dark room orgy with an arm up their anus and a cock in every hand. But that is uncommon and requires off-the-scale extremes.
Re: (Score:2)
The study talks about network structure and key nodes. It reads a lot like they assume that all nodes will pass information without making decisions or applying filters. But we know that this isn't true. I suspect that what is being left out is the ability to identify and 'take out' nodes that don't meet some standard of approved behavior. In other words, researchers have discovered the cancel culture.
Re: (Score:2)
Twitter doesn't block competing information, it just doesn't show you stuff outside your bubble unless you go looking for it.
The problem is that Twitter only has one method evaluating the "quality" of a post - likes. Well, it does do some text analysis but it's primitive and doesn't understand meaning. Ideally it would show a good counter-argument along with popular tweets.
We are at that awkward stage where computers are too dumb and humans too expensive and slow to do this stuff.
Good Data? Bad Data? (Score:2)
I thought maybe this would be something about error correction algorithms, though the concept of 'good' and 'bad' seems a little nebulous.
Data is not the same thing as information.
Automated Minitru (Score:5, Insightful)
I'm sure this will be used responsibly and without any bias as to what is considered "good" information and what is "bad" information.
1984 is starting to look like a quaint utopia compared to what's becoming reality.
Re: (Score:2)
Indeed. And if we could back to 1984, I'd warn my younger self so he can be better prepared for this horrific future: buy a few thousand bitcoins and don't sell them under 5K$USD each.
What? You think that I alone could prevent all that's happening? Please.
Re: (Score:2)
The price climb between Apple's stock and Bitcoin's value [bitcoinpizzaindex.net] doesn't even compare.
Bitcoin's value at the beginning using the pizza index: 0.0041$USD per Bitcoin. Today's value: 6467.92$USD according to Google. Apple's stock never increased its value by 1577541 times.
Re: (Score:2)
Also, why would you wait until 1997 to buy Apple stock?
"Apple first sold shares to the public on Dec. 12, 1980, at $22 per share. The stock has split four times -- three times at 2-for-1, and one split at 7-for-1.Nov 24, 2019"
So even if you bought Apple's share at the very beginning, they were 22.00$USD per share which is a far cry from Bitcoin's 0.0041$USD. To put it in other terms, for every Apple share you bought, you could have had 5365 Bitcoins. Let's go with a single share of Apple just for fun.
With a
Re: (Score:1)
Problem is the alternative is just as bad. Right now some of that misinformation will get you killed. It can create long term problems like climate change denial or not getting children vaccinated.
Anti-intellectualism is a huge problem. The idea that somehow you are an expert in everything and your opinion is the only one that matters has taken hold so people don't listen to experts any more. I'm not a doctor, I'm not an expert on infectious diseases, I'm not going to look on Facebook for a secret COVID-19
Easy (Score:1)
Re: Easy (Score:1)
Might as well include the mainstream media in that. Once you've been alive long enough, you know things are more nuanced than "lies" vs "truth".
Re: (Score:2)
Old, new, bad, good (Score:3)
This is Great News Comrade (Score:4, Interesting)
Re: (Score:2)
It's just gonna be used to manipulate people faste (Score:1)
Isn't it?
What's the opposite of fake news? :D
Fake news!
Of course, as always, what one agrees with, is accurate information, and what one disagrees with, is bad data. "Fuck evidence, fuck the scientific model, what the biggest stick says is true, is true."
(With "one" being their opinion leader [mistaken by them as their society's norm], for all the passive-thinkers out there.)
And we're sitting here, fucked either way, because everyone is being the Mowgli snake, saying "Trust meeee...", and we can't check and
Re: (Score:2)
Instead of being one more asshole who wants to condescendingly impose himself onto us, making him dodgy and untrustworthy all by itself... driving people even deeper into their comforting self-affirming delusions.
In all honesty....
Each mainstream media outlet targets a demographic and feeds that demographic the fake news that that demographic wants to hear. Each government says one thing to its own citizens and another thing to everyone else, where often neither is true. Each religion has its own truth and its not a static truth but instead an ever changing truth as people and society evolves..
In the past few decades we've added several big tech monopolies such as twitter, facebook, and youtube, to the mix, eac
Likely irrelevant to social networks (Score:3)
The core assumptions behind this work are that the problem is the dissemination of corrected information, not its acceptance and replication, and that once good information arrives it will replace the bad information. The problem is that information on social networks doesn't work that way. Information isn't replicated and disseminated because it's accurate, but because it convinces people to accept and spread it. Social networks create a form of memetic evolution, where ideas and their expressions (memes, in the original sense of the word, not the sense of labeled images) compete to see which convinces people to spread them the most. Those with the strongest emotional appeal, not the most accurate information, are usually the most successful.
You can carefully inject the new information at precisely the right points in a social network, the highly-connected nodes that allow information to spread as fast as possible, but unless it provokes the right reactions it will go nowhere.
Re: (Score:2)
I think you mistyped “think for themselves.”
You actually meant, think like you.
Replace "False News" with "Ministry of Truth" (Score:2)
Fast-spreading tools that replace stored data are very dangerous. What is the "truth" that will replace what "false data" ?
The opposite of a stable genius ... (Score:2)
Define accurate (Score:2)
Because this seems a method to use computers to create floods of âoeinformationâ to drown out other data. Basically a way of shouting down opposing views, with little regard for anything but your view prevailing.
How can this work? (Score:2)
Re: (Score:2)
Screw him. He’s just an asshole being an asshole.
What is needed is drowning out the rest of the assholes - Biden, anyone in the DNC or RNC hierarchy, most state governors and (most of all) anything that comes from CBS, NBC, ABC, Fox etc.