Netflix Timeline

Loading...

Do curso por Princeton University

Networks Illustrated: Principles without Calculus

39 classificações

Princeton University

Networks Illustrated: Principles without Calculus

39 classificações

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. Please note that per Princeton University policy, no certificates, credentials or reports are awarded in connection with this course.

Na lição

Movie Recommendation on Netflix

One of the perks of having a Netflix subscription is getting recommendations of movies to watch. Behind the scenes, Netflix uses powerful algorithms to determine which will be suggested to each person specifically. In this lesson, we will take a look at the main ideas behind these algorithms.

Netflix Timeline5:21

Video Streaming5:19

Recommendation is Everywhere3:24

Netflix Recommendation System5:20

Netflix Prize: Logistics7:38

Netflix Prize: The Competition4:33

Our Example3:48

Raw Average7:13

User-movie Interactions7:00

Baseline Predictor6:57

Cosine Similarity9:18

Similarity Values10:58

Leveraging Similarity8:30

Neighborhood Predictor4:44

Performance of Different Methods3:30

Copy of Cosine Similarity9:18

Conheça os instrutores

Christopher Brinton
Lecturer
Electrical Engineering
Mung Chiang
Professor
Electrical Engineering

Hello, welcome back. In this lecture we are going to look at

Netflix and Netflix's recommendation in particular.

And if you ever been to Netflix's site before you've seen how they have

something where if you watch movies or watch whatever, watch TV shows and they

have a recommended section for you. And they'll say, oh we think you should

watch this. You probably would like this.

And so we'll look at some of the main ideas as to how they bill that

recommendation out. We're not going to look at the

mathematical specifics because it would take a course in many mathematical

courses to explain all that. but we'll look at some of the main ideas.

And we'll see some interesting stories behind how it came to be, the way it is.

And so the main over arching principle here, remember in the optional textbook

we base everything in principles is that crowds are wise.

And we didn't look at the chapter before this in the book or the section before

this, rather in the book on Amazon. And it we looked at how Amazon ranked

products on the display page. And there is actually something called

the wisdom of crowds. Where basically it's saying that if

everyone is guessing on the value of something, right?

In this case or in Amazon's case it would be on the, what the ranking of a product

should be on the scale of one to five where everybody submits their view.

I think it's at three stars. I think it's a five star.

I think, you think it's a four star. Someone else thinks it's a two star,

whatever, then they When you average those estimates, you get

something that is closer to the true value, provided that they may make some

assumptions on how the estimates are made.

we didn't have time to go through that here, but the, the main ideas still play

a part in that we're going to rely on what people think about ratings, in order

to make predictions as to what other people's ratings would be.

So let's start off by looking at a brief timeline of Netflix's history.

So you've, if you've ever rented something or seen a box on Netflix, you

know it comes in this red packaging. Typically this is their trademarked red

boxing that you could see in the mail, or anything along those sorts.

And Netflix all started with a movie of Apollo 13, way back when in about 1997.

One of their founders Reed Hestings, had rented Apollo 13 out as a movie.

And he had already been six weeks late on returning it, so he had racked up already

a large sum of late fees. And so, one thing that he thought was,

why should I be charged a late fee for every day that I'm late for this movie?

This is the only movie that I have out right now.

Why, why shouldn't I be able to keep it out as long as I want as long as I don't,

for instance take another movie out. And that was kind of one of his ideas as

he started to get fed up with these, these late fees.

And so in that same year as a matter of fact, in 1997, Netflix was founded and

they started off with their standard DVD rental service.

And they still operated according to that pay per rental.

They didn't move on to the other scheme that I just described yet.

They they were still operating under the assumption that you would rack up late if

fees if you returned a movie late and so on.

And then in 1998, they opened up their online store, which started to add more

convenience to the customers. Because rather than having to drive out

to the store or anything like that, you could just order the DVD's offline and

have them shipped to your house in an envelope that kind of looks like this.

And then in 1999, that's when they really made that switch to rather than pay per

rental to a monthly subscription plan. Where you can keep any movie out as long

as you want and you just, you pay Netflix a monthly subscription, however much each

month I'm not sure what their prices are even now.

But they have a monthly subscription basically.

You can keep a certain number of movies maybe one, two, three, depending on for

how long as you want, but you have a limited number that you can keep out at

one time. Right, so if you had the movie Apollo 13

out and you had it out for six weeks, that's okay but you just can't get a new

one until you return that DVD of Apollo 13.

And this also works for the online store basically until you ship it back, you

can't get a newer DVD back that you want. But you can only have a limited number

out at one time, that's the idea. So rather than paying for each rental

that you have, right, so paying per rental would be that every DVD you take

out, you actually have to pay and have the late fees associated with that.

You just pay a monthly subscription, you can keep each of these out as long as you

want. But you can only have a limited number

amount, at one time. So now in terms of Netflix operation,

there is two things that we should note. They've been really scalable over the

years, and we've looked at scalable in terms of algorithms and things along

those sorts. But, what we really mean by scalable here

is that the cost that it takes them to operate is less as they have more users

basically, so the cost per user is going to be less.

And so basically if you plot this out, right and this would be their cost of

operation in dollars, say. And then this is the number of people or

the number of subscribers that they have. It's going to look at, it's going to be a

curve kind of like the utility curve that we looked at in the last lecture, right.

Kind of like this, right. Where it's going to tip off in the, the

increase is going to get less as the number of people goes higher and higher.

And there are also, in addition to being scalable, they also operate with a lot of

stickiness, and that means customers tend to stay.

Explore nosso catálogo

Registre-se gratuitamente e obtenha recomendações, atualizações e ofertas personalizadas.

O Coursera proporciona acesso universal à melhor educação do mundo fazendo parcerias com as melhores universidades e organizações para oferecer cursos on-line.

© 2018 Coursera Inc. Todos os direitos reservados.

Baixar na App Store

Baixar no Google Play