案例学习：预测房价

Loading...

來自 University of Washington 的課程

机器学习：回归

3806 個評分

案例学习：预测房价

從本節課中

Multiple Regression

The next step in moving beyond simple linear regression is to consider "multiple regression" where multiple features of the data are used to form predictions. <p> More specifically, in this module, you will learn how to build models of more complex relationship between a single variable (e.g., 'square feet') and the observed response (like 'house sales price'). This includes things like fitting a polynomial to your data, or capturing seasonal changes in the response value. You will also learn how to incorporate multiple input variables (e.g., 'square feet', '# bedrooms', '# bathrooms'). You will then be able to describe how all of these models can still be cast within the linear regression framework, but now using multiple "features". Within this multiple regression framework, you will fit models to data, interpret estimated coefficients, and form predictions. <p>Here, you will also implement a gradient descent algorithm for fitting a multiple regression model.

- Emily FoxAmazon Professor of Machine Learning

Statistics - Carlos GuestrinAmazon Professor of Machine Learning

Computer Science and Engineering

[MUSIC]

So we keep talking about this housing application.

And it's really a nice intuitive way to describe the different methods in

regression that we're gonna be talking about, and

that we have been talking about.

But regression, like we've mentioned, is much, much, much more widely applicable.

And this notation of capturing seasonality also

appears in lots of applications beyond just housing.

And so I wanted to spend a little time talking about other places where we see

this seasonality or seasonal effects coming into play.

And one, which makes a lot of sense, is if you're doing weather modeling.

Let's say you're trying to predict the temperature or rainfall.

Well, if you're thinking about temperature,

well there's variation in temperature across a day.

It's hotter during the daytime hours and cooler during the night.

But of course, it's also hotter in the summer and cooler in the winter.

So there's actually this seasonality at different time scales.

So in addition to just having that one sine and

cosine that we showed in that housing model, where in that case,

we're just just looking at this monthly effect repeating every year.

Well, for weather modelling, if you're predicting temperature,

you might wanna add in a sine and

cosine functions at different frequencies to capture the fact that there are these

daily effects as well as these monthly effects, and maybe other effects as well.

Also you might be thinking about Flu monitoring, so you wanna think about

monitoring the incidence rate of flu and, for example here, the picture I'm

showing is in the United States in a whole bunch of different regions.

And if we look at any one of those regions and we look at rate of flu over time,

well of course there are gonna be peaks during flu season and

valleys during the off months.

And so you see this kind of seasonal pattern in flu monitoring and

lots of other types of health monitoring like this.

And you'll also see it in the things like E-commerce, for example, Amazon

is really interested in being able to stock their inventory pretty accurately.

And if you're thinking about selling jackets, you have some warehouse here in

the US and you wanna figure out how many snow jackets or ski jackets to stock.

Well, of course, you're gonna wanna stock more in the winter months than you would

in the summer months because more people are likely to purchase jackets

in the winter than in the summer.

So that's another place where seasonality is really important.

And it appears in so many applications.

Another one that you might not think of is Motion capture,

just trying to model how a person walks over time.

And if you look at the data, if you put sensors over a person's body and

look at how they walk, if you take those recordings, you're gonna get

these kind of up and down, up and down swings, as the person's going through

their different motions, raising their knees or walking or their arms.

And so in this plot here, I'm looking at some center trajectories from a person

wearing a motion capture suit as they're going through different behaviors, and

you clearly see this type of seasonality here as well.

[MUSIC]