What makes WiFi faster at home than at a coffee shop? How does Google order its search results from the trillions of webpages on the Internet? Why does Verizon charge $15 for every GB of data we use? Is it really true that we are connected in six social steps or less? These are just a few of the many intriguing questions we can ask about the social and technical networks that form integral parts of our daily lives. This course is about exploring the answers, using a language that anyone can understand. We will focus on fundamental principles like “sharing is hard”, “crowds are wise”, and “network of networks” that have guided the design and sustainability of today’s networks, and summarize the theories behind everything from the social connections we make on platforms like Facebook to the technology upon which these websites run. Unlike other networking courses, the mathematics included here are no more complicated than adding and multiplying numbers. While mathematical details are necessary to fully specify the algorithms and systems we investigate, they are not required to understand the main ideas. We use illustrations, analogies, and anecdotes about networks as pedagogical tools in lieu of detailed equations. All the features of this course are available for free. It does not offer a certificate upon completion.
Crowds Are Wise & Crowds Are Not So Wise
Loading...
En provenance du cours de Princeton University
Networks Illustrated: Principles without Calculus
40 notes
Explore Related Videos
At Coursera, you will find the best lectures in the world. Here are some of our personalized recommendations for you
À partir de la leçon
Introduction
An introduction to what this course is about: the fundamentals behind social and technical networks.
Rencontrer les enseignants
Christopher BrintonLecturer
Electrical Engineering
Mung ChiangProfessor
Electrical Engineering
Now let's move onto our third principal that crowds are wise.
The material here will be covered in the course.
So to understand this first think about any time you've gone onto Amazon to look
up a product that you may or may not have ended up buying.
But, but what if influence that decision is the star rating that Amazon attributes
to each of the products. Here that's 3.5 stars.
So, have you ever sat back to think about how this rating is actually generated.
If you look at it they'll tell you the number of people which this is
aggregated. So it could be one person, that one
person came on here and rated it as a 3.5.
Now that's not all that reliable because that's only one person's opinion.
But, as you have more and more people that rate it.
Maybe this person gave it a four, this person gives it a three, this person
gives it a five, this person gives is a two, this person gives it a 2.5, and so
on and so forth. When Amazon does is they take the
mathematical average of those numbers and they display the average star rating.
Now the principle of the wisdom of crowds is that the knowledge of the many is
better than the knowledge of the few, so basically as we have more and more people
this rating, this average rating, will become more and more reliable.
But we have too make a lot of assumptions about how these ratings are generated and
as we will see assumptions are not always going to be valid.Other things that
complicate this are the fact that some people may just have different opinions
about the product. This person may like this TV a lot.
Because this person gave it a 5. This person over here maybe have given it
a 1, may just not like this tv at all, its not saying that either of those
ratings are wrong, its just saying that they're different.
Another thing that complicates it is, some reviewers have higher reputations
scores than others. Specifically Amazon has this hall of fame
of reviewers. Now, another place in which this wisdom
of crowds concept comes into play is on Netflix.
But Netflix were not rating items or rating movies specifically.
So, what Netflix does is Netflix wants to make a prediction of a rating for movies
that you haven't yet watched. So, it'll make a prediction for all these
movies, based on your viewing history, movies you've watched, and also the
movies that your neighbors, which we will define exactly what we mean by neighbor
in this course, have watched. And they will look at the responses, and
how well these movies have been received. And it will try to build.
A predicted rating for each of these movies that you haven't watched.
And then it will take those ratings and use that to give you an ordered list of
movies that you should watch. 1, 2, 3 and so on.
And so, the difference between Amazon's problem is that when Amazon.
They're trying to come up with a single rating that is this way to all the users
but, in Netflix they're trying to come up with different ratings for each user and
predictions in order to make recommendations.
So, on this side this is really a problem of rating, whereas in this one we bring
it a step further and actually make recommendations.
Unfortunately things aren't always so great, which brings us to our 4th
principle that crowds are not so wise, and unfortunately we don't have time to
cover this material in this course. We will allude to it at different times.
But to really illustrate this idea of crowds not being so wise, suppose you are
walking down the street. [SOUND] And you come up to someone who is
looking up at the sky. So, this is you and you've just walked
down the street. Now, if you see this person looking up at
the sky you might decide to look up at the sky too.
See, may, maybe there's something interesting up there.
Now, this person may just have a nosebleed or something, or he may just be
trying to play a prank on you, just to get you to look up at the sky for some
reason. But in any event, you follow suit and
look up at the sky as well. Then, the person behind you comes up and
sees you looking up at the sky and decides to do the same thing, and that
continues, and eventually, we have a whole group of people that are looking up
at the sky for, perhaps, no reason. I mean, there might actually be something
wrong with the sky, but there probably isn't anything wrong with the sky.
This is what we call information cascade. Basically everyone ignores there own
private signal, which is that there is nothing wrong with the sky so there's no
reason to look up at the sky and follows the public action, which started by this
person right here, but they see everyone else doing.
And even though they don't think anything might be wrong with the sky, since
everyone else was doing it, they decided they should do it too.
Now these information cascades are precisely what we mean when we say crowds
are not so wise. They're easily influenced and they tend
to follow crowds even when crowds aren't necessarily correct.
Now these types of cascades tend to happen when people exchange information
with one another about what they're thinking.
And that's one of the assumptions that we made in the wisdom of crowds, is that we
can't have people exchanging information and telling each other what they think.
So this can't happen for the wisdom of Krause to hold, everyone has to make
their own guess. So one common example of this would be if
you're working on a homework assignment, and you have three of your friends here,
and three of your friends are all saying that the answer to one question is b.
So they all say yes and they're working together.
They've come up with the fact that the answer is B.
Now you may come along and think that the answer is A, but you don't want to be the
odd man out. So you decide to put that in your paper,
and you say, well, it's 3 against 1 so I may as well just put B because everyone
else has. Now, it might turn out that A is actually
the correct answer. One person came up with the idea that it
was B, and everyone else followed him, not wanting to be the odd man out.
So, now again this is great, but how does this have to do with networking?
Well, this really brings us to the whole idea behind social networks in which we
have people interacting and influencing one another.
And one example of that is on YouTube. And these types of information cascades
can explain why movies such as Gangnam Style and Charlie Bit My Finger Again
have gone viral. And we've seen them explode to even
billions of views. Gangnam Style has broken 1 billion.
The idea is that as you see more people watching these videos, you will also be
recommended to and you will want to watch them to see what's so great about them.
So we have information cascade here. Now it's also prevalent in platforms like
Facebook or Twitter, in which, you actually have big, large social networks
of people. And in Facebook, they're defined by
friend relationships where you might have, be friends with certain amount of
people and everyone else's friends with other people and so on and so forth.
And then the question is that people would be able to determine who the
intellectual people are in the crowd. So for instances if someone was standing
right here, they happen to be friends with a whole lot of people, happens to be
very influential. Than whatever this person does could have
an impact on what everyone else see's on their news feed's or whatever everyone
else does as well. And we have influence that way.
Explorer notre catalogue
Rejoignez-nous gratuitement et obtenez des recommendations, des mises à jour et des offres personnalisées.
Coursera propose un accès universel à la meilleure formation au monde,
en partenariat avec des universités et des organisations du plus haut niveau, pour proposer des cours en ligne.
© 2019 Coursera Inc. Tous droits réservés.
