Building & exploring a nearest neighbors model for Wikipedia articles

Loading...

来自 University of Washington 的课程

机器学习基础：案例研究

7231 个评分

University of Washington

机器学习基础：案例研究

7231 个评分

课程 1（共 4 门，Specialization Machine Learning）

你是否好奇数据可以告诉你什么？你是否想在关于机器学习促进商业的核心方式上有深层次的理解？你是否想能同专家们讨论关于回归，分类，深度学习以及推荐系统的一切？在这门课上，你将会通过一系列实际案例学习来获取实践经历。在这门课结束的时候，

从本节课中

Clustering and Similarity: Retrieving Documents

A reader is interested in a specific news article and you want to find a similar articles to recommend. What is the right notion of similarity? How do I automatically search over documents to find the one that is most similar? How do I quantitatively represent the documents in the first place?<p>In this third case study, retrieving documents, you will examine various document representations and an algorithm to retrieve the most similar subset. You will also consider structured representations of the documents that automatically group articles by similarity (e.g., document topic).</p>You will actually build an intelligent document retrieval system for Wikipedia entries in an iPython notebook.

Loading & exploring Wikipedia data5:21

Exploring word counts5:53

Computing & exploring TF-IDFs7:09

Computing distances between Wikipedia articles5:38

Building & exploring a nearest neighbors model for Wikipedia articles3:18

Examples of document retrieval in action4:16

与讲师见面

Carlos Guestrin
Amazon Professor of Machine Learning
Computer Science and Engineering
Emily Fox
Amazon Professor of Machine Learning
Statistics

[MUSIC]

In the lectures, Emily talked about nearest neighbor models and

how they can be used for document retrieval.

So today, we're gonna actually do some cool document retrievals

using simple nearest neighbor models.

Next we're going to build

a nearest neighbor model for document retrieval.

And here we go, nearest neighbor model for document retrieval.

Again, you could implement your own, but we have one already implemented

here that you can use, which is great for this first course.

We're gonna use the one from the offline, and it is called nearest_neighbors.

And just like that use the verb create to create a nearest neighbor model.

So we're gonna give it this input.

The people table.

And we have to say what features we're gonna use in the nearest neighbors.

So the features are gonna be the tfidf, tfidf.

Just like we've been using for the cosine distances above.

And I'm going to provide a label column and

this label column is going to be the name.

So when I return my nearest neighbors, it's going to

return the actual label of the nearest neighbor, so the name of the person.

And that will make the print out prettier.

So that's cool that it's there for us.

So it should only take a few seconds.

And we will have, now there we are.

We will have a nearest neighbor model for document retrieval.

So now we can use it, so let's go ahead and use it.

So I'm going to just create another header here and

say applying the nearest-neighbors model for

retrieval, and let's see what we get.

So the first question is, so let's do our first query.

So who is closest to Obama?

So who are the closest people to Obama in the data set.

So if you just take our neighbor model,

we can use the method query, the function query associated with it,

which just queries the model with some input.

So in this case Obama as our input, and gives me back the nearest-neighbor.

So where the nearest-neighbor is to Obama, we'll compute it, and it's very quick.

And I say, okay, the closest person to Obama is Obama himself, which makes sense.

But after that, we have Joe Biden, who is the current US vice president.

So the president is closest to the vice president.

And then we find a few other politicians, including former president Bill Clinton,

which we saw above.

[MUSIC]

探索我们的目录

免费加入并获得个性化推荐、更新和优惠。

Coursera 致力于普及全世界最好的教育，它与全球一流大学和机构合作提供在线课程。

© 2018 Coursera Inc. 保留所有权利。

通过 App Store 下载

通过 Google Play 获取