1
00:00:00,410 --> 00:00:03,634
So, in this next segment of the class, 
we're going to describe an algorithm 

2
00:00:03,634 --> 00:00:07,070
called The Brown et al, Word Clustering 
Algorithm. 

3
00:00:07,070 --> 00:00:12,170
It's a rather remarkable algorithm and 
that it takes in as input, a large 

4
00:00:12,170 --> 00:00:18,739
quantity of unlabeled raw text. 
And as it's output produces very useful 

5
00:00:18,739 --> 00:00:24,584
representations of individual words. 
It's really the first example we'll see 

6
00:00:24,584 --> 00:00:29,348
in this course of a so-called 
unsupervised learning algorithm. 

7
00:00:29,348 --> 00:00:33,360
In that, it doesn't require annotated 
data rather it just requires raw 

8
00:00:33,360 --> 00:00:36,630
unlabeled text as it, as it's input.