1
00:00:00,000 --> 00:00:04,796
[MUSIC]

2
00:00:04,796 --> 00:00:08,519
In summary, in this module we
covered two very important and

3
00:00:08,519 --> 00:00:11,740
widely used algorithms,
k-means and MapReduce.

4
00:00:11,740 --> 00:00:15,550
We started the module by
presenting the goal of clustering.

5
00:00:15,550 --> 00:00:19,750
And since then, we've motivated
many applications of clustering.

6
00:00:19,750 --> 00:00:25,240
And then we dug into k-means as a specific
algorithm for performing clustering.

7
00:00:25,240 --> 00:00:29,030
And it really is the mostly
widely used algorithm out there.

8
00:00:29,030 --> 00:00:33,360
And then we took k-means as
an opportunity to present MapReduce,

9
00:00:33,360 --> 00:00:38,170
which is a general framework for
parallelizing algorithms.

10
00:00:38,170 --> 00:00:42,880
And we showed how you can apply
an iterative version of MapReduce

11
00:00:42,880 --> 00:00:45,010
to parallelize k-means.

12
00:00:45,010 --> 00:00:48,620
And like we almost always say,
what you learned in this module

13
00:00:48,620 --> 00:00:52,430
really generalizes to a wide range
of different applications, and

14
00:00:52,430 --> 00:00:54,460
different things you might
be interested in doing.

15
00:00:54,460 --> 00:00:57,365
But it's really, really,
really especially true for this module.

16
00:00:57,365 --> 00:01:02,130
K-means is a very generic tool for
clustering, and clustering is used in so

17
00:01:02,130 --> 00:01:03,560
many different areas.

18
00:01:03,560 --> 00:01:08,810
And MapReduce is a very, very powerful and
widely used framework for

19
00:01:08,810 --> 00:01:13,220
creating distributed implementations or
parallelizing algorithms.

20
00:01:13,220 --> 00:01:14,150
So, in summary,

21
00:01:14,150 --> 00:01:18,493
here are the set of things that you should
be able to do after finishing this module.

22
00:01:24,158 --> 00:01:28,379
[MUSIC]