Online Social Networks Can Be Tipped By Less Than 1% of Their Population

An anonymous reader writes "A new algorithm developed by researchers at West Point seems to break new ground for viral marketing practices in online social networks. Assuming a trend or behavior that spreads in an online social network based on the classic 'tipping' model from sociology (based on the work of Thomas Schelling and Mark Granovetter), the new West Point algorithm can find a set of individuals in the network that can initiate a social cascade – a progressive series of 'tipping' incidents — which leads to everyone in the social network adopting the new behavior. The good news for viral marketers is that this set of individuals is often very small – a sample of the Friendster social network can be influenced when only 0.8% of the initial population is seeded. The trick is finding the seed set. The algorithm is described in a paper to be presented later this summer at the prestigious IEEE ASONAM conference."

Comments

Wow, Friendster? All 300 Users?

[eldavojohn]

a sample of the Friendster social network can be influenced when only 0.8% of the initial population is seeded.

Friendster? Wow, you could influence, like, 300 people!

Any chance they're just witnessing C&C nodes transmitting spam orders or pagerank gaming links to the remaining 99.2% of Friendster accounts (all of which are hacked and forgotten)?

It all makes sense

[Sparticus789]

Is this what Occupy means when they say 1% of the country controls everything?

Re:It all makes sense

[jellomizer]

But which 1%, There are a lot of 1% out there.

The Tea Party is controlled by the Oil Companies 1%.
The Occupy is controlled by the Unions 1%
The Favorite Trend is controlled by the Marketers 1%...

It doesn't seem that you have any decisions to make for yourself, There is always someone else telling you what to think.

Re:It all makes sense

[geekmux]

Is this what Occupy means when they say 1% of the country controls everything?

Uh, I tend to read this statistic in the other direction. It means 99.2% of people are nothing more than sheep following the flock, which makes sense considering we're basically talking about Farcebook. All other forms of social networking have pretty much been reduced to a moot point.

Re:It all makes sense

[Baloroth]

That's not really what this statistic is saying. It is using a "tipping" model, where it assumes anyone who has more than 50% of his friends exhibiting the behavior will automatically adopt it. A useful model, but not actually true (like nearly all mathematical models, it is only approximately true in the real world). That means they only have to find a "seed" population to adopt the trend: the model says if all of them adopt it, everyone on the network will. Think of it less like sheep and more like dominoes: you only need to trigger one dominoe to trigger the rest, but that presumes a carefully constructed ideal system. In reality, 99% of people may be sheep, but this study says absolutely nothing about that. It assumes it, rather than proving it.

Re:It all makes sense

[tomhath]

It's the other way around; Occupy is trying to be the seed that starts the social cascade. Their problem is twofold though 1) They represent far, far less than the necessary seed size, and 2) Their attempts to initiate tipping incidents don't result in any cascade because the 99% they claim as sympathizers aren't.

Re:It all makes sense

[tlhIngan]

It's the other way around; Occupy is trying to be the seed that starts the social cascade. Their problem is twofold though 1) They represent far, far less than the necessary seed size, and 2) Their attempts to initiate tipping incidents don't result in any cascade because the 99% they claim as sympathizers aren't.

That's because they use the wrong targets that end up making them look like unemployed hippies.

To "tip" a population properly requires people to reaosnably agree with you - if I headed occupy (metaphoically), you can start with something so simple, so basic, yet everyone is powerless to fix.

An example would be gas - why is it costing just the same as it did before the crash? Oil's down these days (and yes, even though there's about as much relation between gas prices and oil prices as there is between a head of lettuce and oil prices, most people equate oil prices with gas). Tap into that rage and it's much easier to tip.

Trying to convince people that the rich are ruining our lives and enslaving us is a concept that's much too foreign to most people to comprehend. Use more concrete examples and you'll be more successful. Especially if that example has a deep-rooted emotion attached with it.

It also applies to everything - take ACTA for example. Talking about copyright law in general gets you blank stares. So talk about its effects - it can make your iPod illegal (think "they're gonna take your iPod away!").

A concrete example is worth way more to tip someone over in your favor than some wishy-washy concept that no one can relate to. Heck, it can even be seasonal - support for global warming ebbs and flows - it ebbs in the winter and reaches a low in the spring, and flows in the summer. The hotter the summer, the more support grows. The colder the winter, the more support is lost.

Re:It all makes sense

[HeckRuler]

The implied metric of the "we are the 99%" is either wealth or income, which is nearly synonymous. This has been painfully clear to anyone that's been paying attention. I'm going to feel dirty for this, but I also have to defend the TEA party. They're only controlled by the oil companies if you look at it through conspiracy glasses and assume that foxnews is controlled by oil men. They have tangential control at best.
I think you just tacked on "1%" to the end of some stereotypes and thought it was clever.

And of course there's always someone else that wants you on their side. Probably a lot of people. Welcome to a democracy, where they care what you think. It's your civil duty to parse through the marketing bullshit, choose the one you think is best, and either vote for the guy or support the cause.

1% can tip anything

[Toe, The]

Look at, well... anything. In any human social activity, there are a few people who drive all the activity, and the rest are happy to follow along.

Even leadership personalities are followers much of the time. It's not like everyone can be leaders in everything. You can only ever lead in a few small areas. (Though of course, some people lead more than others; while some people lead in nothing at all, I suppose.)

As an indie author

[yoctology]

Trying to be noticed among a million other offerings, this is good news. After doing my best job writing, I can then try to figure out how to reach my own 1% to tip them toward my work, rather than trying to brute-force popularity.

Locke and Demosthenes from Ender's Game

[Anonymous Coward]

Does this remind anyone of Locke and Demosthenes from Ender's Game? Seeding a few carefully worded articles to change the discourse of the network?

Re:Wow, Friendster? All 300 Users?

[CohibaVancouver]

Friendster? Wow, you could influence, like, 300 people

American-centrist much, you insensitive clod?

From http://en.wikipedia.org/wiki/Friendster

Since the relaunch of Friendster as a social gaming platform in June 2011, the number of registered users has reached over 115 million. Over 90% of Friendster's traffic came from Asia. The top 10 countries accessing Friendster, according to Alexa, as of May 7, 2009 are the Philippines, Indonesia, Malaysia, Singapore, Thailand, Pakistan, United Arab Emirates, Sudan, South Korea, Bangladesh and India

Sheeple

[Chemisor]

A much more interesting conclusion of this study is that 99.2% of social network users will do anything their friends would tell them to do.

The trick is not just finding the seed set

[timeOday]

"The trick is finding the seed set." No, you still have to influence the seed set, which might be really hard.

Let's say this model predicts that I can end terrorism by converting 100 radical muslims to buddhism. How does that help me? (Simply sending in drones to remove these nodes from the graph, so to speak, will not have the same effect).

Second example, let's say my novel is almost guaranteed to be successful if it gets a glowing review in the New York Times. Well, how hard can that be? Usually trusted nodes are trusted for some reason - because they're reliable. That means they're hard to influence.

Re:The trick is not just finding the seed set

[utoddl]

No, the trick is to find the 1% of the 1% who will influence the other 99% of the greater 1% who will then get everybody else on board and get them on board.

Re:The trick is not just finding the seed set

[yoctology]

I think you are misusing "hard." For an example, one of my books has a very minor incident in which a Glock 30 is used. I know an influential writer and reviewer is an enthusiastic 2nd Amendment champion. I send her a crafted blurb mentioning the Glock. bingo. We exchange friendly messages about pistols, she buys the book and is now reviewing it for her multi thousand followers.

It wasn't hard to influence her; just took a second of reflection and doing what people do socially all the time without thinking. The overall point is that even authors with extremely limited advertising funds can make up for it with luck and savvy. On the other hand I am unlikely to win a chess tournament no matter how clever or lucky I am.

Re:The trick is not just finding the seed set

[Attila Dimedici]

You are missing the OP's point. He is not saying it is hard to influence a member if that group. What makes it hard is that you have to get all of those influencers and you have to do it in such a way that none of them start feeling like you are manipulating them before you have convinced all of them.

Bellwether

[Nethead]

This brings to mind the Connie Willis novel Bellwether

The main character, Dr. Sandra Foster, studies fads in Boulder, Colorado. Her employer, Hi-Tek, wants to know how to predict fads, in order to take advantage of this knowledge and thus to possibly create one.

A good read, quite enjoyable and funny.

"The trick is finding the seed set."

[Lord Byron Eee PC]

"The trick is finding the seed set." No, it's not. The real trick is finding the seed set of the seed set. On Facebook, you have 900 million users. 1% of that is 9 million, which is too large to influence. But 1% of that 1% is just 90,000, something that a targeted advertising campaign might be able to influence.

Re:Wow, Friendster? All 300 Users?

[plover]

Any chance they're just witnessing C&C nodes transmitting spam orders or pagerank gaming links to the remaining 99.2% of Friendster accounts (all of which are hacked and forgotten)?

It's a comp sci paper that is looking for connected nodes in a network, and they're using copies of data sets of social networks as their starting point. They aren't monitoring networks looking for "who is exerting influence over them", they're looking for nodes that are well connected to other nodes, presuming those represent the most valuable people to convince.

Now, could those "friends and families" in the network data actually be there as part of a botnet controller and its zombie minions? Sure, why not? But each one of those would be a single node in the set of nodes as having the right connections. Doesn't mean that marketing to the botherder or the botnet is going to get you much business, but if you were looking for someone who has influence, it would identify the botherder and not the bots themselves.

90-9-1 rule

[sckienle]

I may have the title of this wrong, but it is a well known rule of thumb in social media tech circles that of 100% of users, 90% pretty much just read, 9% post regularly, and only 1% are really active. So they have simply come up with the algorithm to determine that 1%.

Re:This isn't news, is it?

[0123456]

I remember reading about a US study during the Cold War which found that if three specific people in the military and government colluded together they could start an unauthorised nuclear war. Fortunately the government ensured that it couldn't 'work' by monitoring those three people to ensure they couldn't collude.

This just in...

[Minwee]

Online Social Media Networks Inflate Their Numbers by 5000%.

When only 2% of the registered accounts are active, it's not hard to see that the right 1% can make a big change.

Algorithm is very simple

[mrops]

I looked at it, and it looks like this

public static boolean willTip(User user) {
  if(user.sex == SexType.FEMALE && user.hotness>(Long.MAX_VALUE-100)) {
      return true
  }
  return false;
}

Obviously. . .

[PeanutButterBreath]

. . .all anyone needs to know is what it will take to get Kevin Bacon to change from on social network to another.

Re:Wow, Friendster? All 300 Users?

[bkaul01]

How are you defining a "major religion"? Christianity has around 2 billion adherents, Islam around 1.5 billion, Hinduism around a billion, Buddhism around half a billion ... other than Judaism, what major religions can count less than 115 million people?

Uhhhh... Captain Obvios on line one...

[Panaflex]

I guess they've never heard of George Takei... he tips The Facebook everyday.

Re:Uhhhh... Captain Obvios on line one...

[AugstWest]

I was thinking the same thing.

"I'll take Takei for the tip."

Wait, let me rephrase that....

Don't take this seriously

[WOOFYGOOFY]

Look you can't take claims like this seriously, by which i mean as immutable laws of nature or even as normative of online communities in a longitudinal sense, that is, as an enduring property of online communities.

From the paper:

In this problem we have a social network in the form of a directed graph and thresholds for each individual. Based on this data, the desired output is the smallest possible set of individuals such that, if initially activated, the entire population will adopt the new behavior (a seed set)

What the study shows in not that will happen in real social networks, but rather in their "tipping model" which is a directed graph whose nodes "activate" when they reach a certain threshold.of input given to them by surrounding nodes.

So what they demonstrated was a property of directed graphs and nodes with a certain made-up (ad hoc) set of characteristics. To assume that those characteristics are descriptive of human beings in a real social network is to extrapolate beyond the results of paper.

The authors obviously think that such extrapolation may be possible since they cite two other papers that they characterize as showing that real social networks have exhibited such behavior, but actually, those papers show something much more hypothetical and specific which I won't go into here.

When they say they applied their theory to social networks, (Buzznet Douban Flickr Flixster FourSquare Frienster Last.Fm LiveJournal Livemocha WikiTalk ) they mean they borrowed the physical topology - the interconnectedness of the nodes- of those networks, (which is available to researchers) NOT that they either found examples of nor instigated the real world behaviour of the people in those networks.

Back to reality, that such things CAN happen is not surprising . I am pretty sure Jennifer Aniston represented less than 1% of the group of American females in the mid 90s, and she wielded the power to tip hairstyles ("The Rachel" hairstyle!!! ) enormously in that time.

Similarly in Roman times, the hairstyles of prominent individual women would appear on coins, for instance, the Emperor's wife. This would lead to a frenzy of copycat hairstyles because hairstyle was one way the rich signaled their status.

There's a danger here that graph theory being applied to social networks will play the role of the mythical "perfectly rational actor" has played in economics. That is, a clean model which produces complex results whose ultimate referent is ONLY itself and in many decisive ways emits behaviour which is OPPOSITE of the behaviour of the real world entity which the theory sought to model.

People are irrational in ways that until recently, with the advent of behavioural economics, were not accounted for in economic models. IMO behavioural economics might as well have taken the name "real economics" . The same thing is going on here. How real people really behave in social networking sites is a wide open question. What we know is people hate to be manipulated and will act against their own seeming best interests in a wide variety of circumstances. See Dan Ariely's "Predictably Irrational" for some examples. Also here's the page on irrationality in Wikipedia. http://en.wikipedia.org/wiki/Irrationality

The point here is

1) this is not a study of people's behaviour, it's a study of the behaviour of nodes which have just those properties the researchers elected to give them transferred to a network topologies which were taken from a variety of real social networks.

2) The behaviour of real social networks is not determined by the assumptions of the researchers

3) nor did those assumptions model the actual behavior of real people in those networks.

Real behaviour is vastly more complex than emitting behavior when a threshold "input" from surrounding people is reached.

Finally, it should be noted that p

Re:Wow, Friendster? All 300 Users?

[Eponymous Hero]

like this: christianity is not a religion, it's a classification of religions. catholics are not methodists are not baptists are not 7th day adventists are not episcopalians are not jehovah witnesses etc...

Re:Wow, Friendster? All 300 Users?

[SockPuppetOfTheWeek]

You forgot to mention - when he didn't capitalize Like, he insulted half a dozen Grammar Nazis.

Re:Wow, Friendster? All 300 Users?

[Red Flayer]

all of Christainity worship the same savior and are supposed to follow the same rules.

That's not completely true. While they all have their roots in the same book, the actual religions can be very different.

Roman Catholics, for example, include the worship of demigods (they call them Saints) and obeisance to the Church hierarchy, as well as the rite of confession. Some Protestant religions base their religion on personal understanding of the New and Old Testaments, and the Good Book is the only set of rules to live by. Some Protestant religions include the rite of confession, some don't. Some have clergy, some don't.

To say that all Christian religions are the same except for trappings would be the same as saying that all Abrahamic faiths are the same except for trappings. I mean, sure, Christians have a set of extra books to follow (compared to Jews),and Muslims have another book on top of that. But really, it's the same God they worship, so they're all the same religion, right?