﻿1
00:00:00,133 --> 00:00:00,866
Hello my friends.

2
00:00:00,866 --> 00:00:01,633
Welcome back.

3
00:00:01,633 --> 00:00:02,300
Are you ready.

4
00:00:02,300 --> 00:00:04,700
To finish and complete that cleaning of.

5
00:00:04,700 --> 00:00:05,833
All our reviews?

6
00:00:05,833 --> 00:00:06,600
Let's do this.

7
00:00:06,600 --> 00:00:09,033
So just to recap on what we did so far.

8
00:00:09,033 --> 00:00:11,033
We made this for loop to, you know.

9
00:00:11,033 --> 00:00:12,866
Treat each review separately.

10
00:00:12,866 --> 00:00:14,866
And clean each of them. Separately.

11
00:00:14,866 --> 00:00:15,600
And the first thing we.

12
00:00:15,600 --> 00:00:19,700
Did for each of these reviews
was to replace all these. Non.

13
00:00:19,700 --> 00:00:22,366
Letters by spaces and basically all.

14
00:00:22,366 --> 00:00:24,833
The punctuations by. Spaces.

15
00:00:24,833 --> 00:00:27,600
Then we turned all the letters.

16
00:00:27,600 --> 00:00:28,800
Into. Lowercase.

17
00:00:28,800 --> 00:00:31,633
So all these capital letters
were. Turned into.

18
00:00:31,633 --> 00:00:33,400
Lowercase. And finally.

19
00:00:33,400 --> 00:00:35,333
We splitted each review.

20
00:00:35,333 --> 00:00:37,600
Into its different words. Because now.

21
00:00:37,600 --> 00:00:38,900
We were about to proceed with.

22
00:00:38,900 --> 00:00:40,600
Stemming, which consists.

23
00:00:40,600 --> 00:00:41,800
Of. Simplifying.

24
00:00:41,800 --> 00:00:44,433
Each word by the root of the word.

25
00:00:44,433 --> 00:00:44,800
So for.

26
00:00:44,800 --> 00:00:47,800
Example. Loved will be replaced by love.

27
00:00:48,000 --> 00:00:49,800
And I remind that it's very important to.

28
00:00:49,800 --> 00:00:50,833
Do this to.

29
00:00:50,833 --> 00:00:54,533
Optimize
the dimensionality of the sparse matrix.

30
00:00:54,533 --> 00:00:56,466
Which will be created through. The next.

31
00:00:56,466 --> 00:00:58,433
Step with the. Bag of words model.

32
00:00:58,433 --> 00:01:01,000
Indeed, we will create that sparse matrix
where each.

33
00:01:01,000 --> 00:01:03,300
Column is. A. Word among all the.

34
00:01:03,300 --> 00:01:05,333
Different words of our different reviews
after.

35
00:01:05,333 --> 00:01:09,000
The cleaning, and therefore
we need to simplify it to the maximum.

36
00:01:09,000 --> 00:01:10,966
Well,
the different. Words that are going to be.

37
00:01:10,966 --> 00:01:12,233
In this sparse matrix.

38
00:01:12,233 --> 00:01:14,433
So it's very.
Very important to apply stemming.

39
00:01:14,433 --> 00:01:17,266
And that's.
What we're about to do. Right now.

40
00:01:17,266 --> 00:01:17,633
All right.

41
00:01:17,633 --> 00:01:19,800
So remember we imported this.

42
00:01:19,800 --> 00:01:21,000
Porter Stemmer class.

43
00:01:21,000 --> 00:01:23,900
That's exactly
the class that will apply stemming.

44
00:01:23,900 --> 00:01:25,300
And so now the first thing we'll do is.

45
00:01:25,300 --> 00:01:27,166
Actually. Create an object.

46
00:01:27,166 --> 00:01:31,533
Of this class, which we will call p s for,
you know, for a steamer. And.

47
00:01:31,533 --> 00:01:33,800
Forest
Emma is a particular type. Of stem here.

48
00:01:33,800 --> 00:01:36,766
You can check it out online.
But that's. Basically the classic way.

49
00:01:36,766 --> 00:01:38,166
Of applying stemming.

50
00:01:38,166 --> 00:01:39,766
So and now.

51
00:01:39,766 --> 00:01:42,766
We will just call the class Porter.

52
00:01:42,900 --> 00:01:45,366
Stem nerd. Perfect.

53
00:01:45,366 --> 00:01:46,966
Then adding some parentheses.

54
00:01:46,966 --> 00:01:49,266
And we don't have to enter any parameters.

55
00:01:49,266 --> 00:01:51,600
So let me actually. Scroll down a bit.

56
00:01:51,600 --> 00:01:52,233
All right.

57
00:01:52,233 --> 00:01:54,233
This will be easier for you.

58
00:01:54,233 --> 00:01:56,533
So Porter. Stemmer
now we. Have our object.

59
00:01:56,533 --> 00:01:59,666
So basically we have our tool
to apply the stemming.

60
00:01:59,833 --> 00:02:00,933
And therefore well.

61
00:02:00,933 --> 00:02:03,300
The next. Step will be exactly to apply.

62
00:02:03,300 --> 00:02:05,500
Stemming to. This particular review.

63
00:02:05,500 --> 00:02:06,966
We're dealing with right now inside.

64
00:02:06,966 --> 00:02:09,266
This for loop iterating all the reviews.

65
00:02:09,266 --> 00:02:10,800
So the way we're going to do.

66
00:02:10,800 --> 00:02:12,666
This because you know,
we have to apply stemming to.

67
00:02:12,666 --> 00:02:15,400
Each of the words. Inside. This review.

68
00:02:15,400 --> 00:02:20,000
And the words have already been gathered
separately by splitting this review.

69
00:02:20,266 --> 00:02:22,000
And therefore what we're going to do now.

70
00:02:22,000 --> 00:02:24,433
Is first, we'll update our review.

71
00:02:24,433 --> 00:02:27,066
Again to become this time.

72
00:02:27,066 --> 00:02:30,133
A list of all these separate words.

73
00:02:30,166 --> 00:02:31,500
Coming from the. Split.

74
00:02:31,500 --> 00:02:33,033
But stepped, you know.

75
00:02:33,033 --> 00:02:35,633
After we applied. Stemming on them.

76
00:02:35,633 --> 00:02:38,600
And so now the question is,
how are we going to apply stemming to.

77
00:02:38,600 --> 00:02:40,333
Each of the words of the review?

78
00:02:40,333 --> 00:02:44,100
Well, we're going to do that,
of course, with a single row for loop,

79
00:02:44,100 --> 00:02:46,200
as I like to call it,
you know, a full loop.

80
00:02:46,200 --> 00:02:47,600
Inside a same row.

81
00:02:47,600 --> 00:02:50,466
Which is,
you know, usually applied to lists.

82
00:02:50,466 --> 00:02:51,100
And since now.

83
00:02:51,100 --> 00:02:53,700
We actually have a list of.
The words in the reviews.

84
00:02:53,700 --> 00:02:54,400
Thanks to this.

85
00:02:54,400 --> 00:02:56,733
Steps. And here, well,
we can totally apply.

86
00:02:56,733 --> 00:03:00,866
This single row for loop to iterate
through all the words of the review.

87
00:03:00,866 --> 00:03:03,400
And apply stemming to each of them.

88
00:03:03,400 --> 00:03:05,300
So let's actually start with this
for loop.

89
00:03:05,300 --> 00:03:10,366
We're going to start with four which will
iterate through all the word you know.

90
00:03:10,366 --> 00:03:14,033
So word is just a iterated variable
that will be equal to the.

91
00:03:14,033 --> 00:03:15,666
Different words of. The review.

92
00:03:15,666 --> 00:03:17,266
So for word.

93
00:03:17,266 --> 00:03:18,733
Then in.

94
00:03:18,733 --> 00:03:20,100
Review right.

95
00:03:20,100 --> 00:03:22,933
Because review is now
a. List of different words.

96
00:03:22,933 --> 00:03:26,800
So it will iterate through
all the different words of this review.

97
00:03:27,133 --> 00:03:29,400
And then we will add something which.

98
00:03:29,400 --> 00:03:30,933
Is probably.
Something. You will see for the.

99
00:03:30,933 --> 00:03:33,133
First time if you're. New to Python.

100
00:03:33,133 --> 00:03:35,400
But here, after a single row for loop.

101
00:03:35,400 --> 00:03:36,133
We can. Actually.

102
00:03:36,133 --> 00:03:38,533
Add a condition to omit.

103
00:03:38,533 --> 00:03:41,433
Some of the words
we are. Iterating through.

104
00:03:41,433 --> 00:03:43,966
And according. To you, what.
Do we want to emit?

105
00:03:43,966 --> 00:03:46,800
You know, what do we want
not to include in the full loop?

106
00:03:46,800 --> 00:03:49,433
Well, that's of. Course remember them.

107
00:03:49,433 --> 00:03:51,233
The stopwords.

108
00:03:51,233 --> 00:03:53,166
Right. Remember, we want to get rid.

109
00:03:53,166 --> 00:03:54,666
Of all the words that.

110
00:03:54,666 --> 00:03:58,300
Won't help us or, you know,
that won't help the machine learning model

111
00:03:58,333 --> 00:03:59,433
to understand whether.

112
00:03:59,433 --> 00:04:01,700
A review is positive or negative.

113
00:04:01,700 --> 00:04:02,500
And these include.

114
00:04:02,500 --> 00:04:05,766
All the articles like the, Unlike an.

115
00:04:05,766 --> 00:04:07,800
Apple, we would just keep Apple, for.

116
00:04:07,800 --> 00:04:10,266
Example,
and even all the pronouns, you know.

117
00:04:10,266 --> 00:04:12,233
Like I, you, he she.

118
00:04:12,233 --> 00:04:13,800
We, you know, then.

119
00:04:13,800 --> 00:04:15,733
They all these words give us.

120
00:04:15,733 --> 00:04:17,366
Absolutely no hint to.

121
00:04:17,366 --> 00:04:20,000
Figure out
whether review is positive. Or negative.

122
00:04:20,000 --> 00:04:22,900
And therefore we want.
To get rid of all the stopwords.

123
00:04:22,900 --> 00:04:24,166
And that's again for the same.

124
00:04:24,166 --> 00:04:25,966
Purpose as. Stemming.

125
00:04:25,966 --> 00:04:29,833
It is to simplify the final sparse matrix,
which will contain

126
00:04:29,833 --> 00:04:30,900
all the different words of.

127
00:04:30,900 --> 00:04:33,000
All our. Reviews into columns.

128
00:04:33,000 --> 00:04:36,566
So we not only want to simplify
the different versions of words

129
00:04:36,566 --> 00:04:39,366
or, you know, branches of words
by only keeping their root. But.

130
00:04:39,366 --> 00:04:40,400
Also we want to get with.

131
00:04:40,400 --> 00:04:41,366
All the stopwords.

132
00:04:41,366 --> 00:04:44,300
In order
not to include them in the sparse matrix.

133
00:04:44,300 --> 00:04:46,666
And a way to do this,
you know, the way to omit.

134
00:04:46,666 --> 00:04:47,466
These words.

135
00:04:47,466 --> 00:04:50,233
When iterating
through the different words in the review.

136
00:04:50,233 --> 00:04:52,633
Is by just. Adding. If, here.

137
00:04:52,633 --> 00:04:56,033
And then and not, you know
you will understand after I finished this.

138
00:04:56,033 --> 00:04:57,500
So let me just write it.

139
00:04:57,500 --> 00:05:00,366
If not, word in.

140
00:05:00,366 --> 00:05:02,700
The set up.

141
00:05:02,700 --> 00:05:03,233
Then I'm going to.

142
00:05:03,233 --> 00:05:05,700
Call the Stopwords module here.

143
00:05:05,700 --> 00:05:07,300
You know this stopwords here.

144
00:05:07,300 --> 00:05:10,100
From which I'm
going to get all the English.

145
00:05:10,100 --> 00:05:12,566
Stopwords because. You know,
our reviews are in English.

146
00:05:12,566 --> 00:05:15,366
And the way to. Do
this is to add here words.

147
00:05:15,366 --> 00:05:17,966
And then in parentheses
because that's a function.

148
00:05:17,966 --> 00:05:20,966
And then in quotes English okay.

149
00:05:21,166 --> 00:05:23,633
So we clearly. Understand what this means.

150
00:05:23,633 --> 00:05:26,900
It means that. If the word of the review.

151
00:05:27,000 --> 00:05:28,566
You know we're dealing with right now
in this.

152
00:05:28,566 --> 00:05:31,000
For loop, if this word is.

153
00:05:31,000 --> 00:05:34,000
Not in the set, you know, the symbol.

154
00:05:34,033 --> 00:05:37,366
Of all the English stopwords. Well, we.

155
00:05:37,366 --> 00:05:39,033
Will consider it in this for loop.

156
00:05:39,033 --> 00:05:41,400
And then we will. Apply the. Stemming.

157
00:05:41,400 --> 00:05:44,300
However, if this word is. In all.

158
00:05:44,300 --> 00:05:46,200
These stopwords list, well.

159
00:05:46,200 --> 00:05:49,500
We won't include it in this for loop
and therefore we won't apply the stemming

160
00:05:49,500 --> 00:05:49,900
to. It.

161
00:05:49,900 --> 00:05:52,800
Therefore it
won't. Be in the future. Sparse matrix.

162
00:05:52,800 --> 00:05:54,566
So you see the trick. Pretty cool.

163
00:05:54,566 --> 00:05:55,300
But so.

164
00:05:55,300 --> 00:05:57,666
Far we only have the for loop
and we don't know what we're going to.

165
00:05:57,666 --> 00:06:00,600
Do for each of these words.
So we have to specify it now.

166
00:06:00,600 --> 00:06:01,266
And remember we.

167
00:06:01,266 --> 00:06:05,666
Have to specify this before the for loop
because that's a single row for loop.

168
00:06:05,966 --> 00:06:06,300
And so.

169
00:06:06,300 --> 00:06:08,233
What we're going to specify is of course.

170
00:06:08,233 --> 00:06:10,600
That we want. To apply. The stemming.

171
00:06:10,600 --> 00:06:13,100
To each of these. Words in the review,
which are.

172
00:06:13,100 --> 00:06:14,666
Not just upwards.

173
00:06:14,666 --> 00:06:15,766
And so now the question is.

174
00:06:15,766 --> 00:06:17,700
How do we apply. The stemming.

175
00:06:17,700 --> 00:06:19,733
Well remember we. Just created our.

176
00:06:19,733 --> 00:06:21,666
Border stemmer object.
So we're going to start.

177
00:06:21,666 --> 00:06:23,533
With this calling our object.

178
00:06:23,533 --> 00:06:26,366
From which
we're going to call specific method.

179
00:06:26,366 --> 00:06:27,433
Which is exactly the.

180
00:06:27,433 --> 00:06:30,433
Stem method.
That will apply. The stemming.

181
00:06:30,600 --> 00:06:33,566
And since it is method
I'm adding some parentheses.

182
00:06:33,566 --> 00:06:35,366
And now I'm sure you. Guessed perfectly.

183
00:06:35,366 --> 00:06:37,133
What's inside these parentheses.

184
00:06:37,133 --> 00:06:38,600
You know what must be inside.

185
00:06:38,600 --> 00:06:41,033
And that of course our word right.

186
00:06:41,033 --> 00:06:42,566
Our word on which we want to.

187
00:06:42,566 --> 00:06:44,033
Apply. Stemming.

188
00:06:44,033 --> 00:06:45,500
And now there you go, my friends.

189
00:06:45,500 --> 00:06:47,733
All this line of. Code apply.

190
00:06:47,733 --> 00:06:48,800
The stemming to.

191
00:06:48,800 --> 00:06:53,200
All the words in your review
except the stopwords, except all.

192
00:06:53,200 --> 00:06:55,133
The non relevant words that are not.

193
00:06:55,133 --> 00:06:57,500
Helpful to predict
whether a review is positive.

194
00:06:57,500 --> 00:06:59,466
Or negative. Great.

195
00:06:59,466 --> 00:07:00,733
So that's a good step. Done.

196
00:07:00,733 --> 00:07:01,966
Now you know how. To apply.

197
00:07:01,966 --> 00:07:03,833
Stemming to. A review.

198
00:07:03,833 --> 00:07:04,800
And now since.

199
00:07:04,800 --> 00:07:06,333
We actually have our.

200
00:07:06,333 --> 00:07:08,966
Different words
in the review into a. List, we will just.

201
00:07:08,966 --> 00:07:09,766
Join them back.

202
00:07:09,766 --> 00:07:12,666
Together because, you know,
we actually had to do. This in order to.

203
00:07:12,666 --> 00:07:15,100
Easily apply.
The stemming to all. The words.

204
00:07:15,100 --> 00:07:17,400
Thanks to this for loop,
but now that it's done, we will.

205
00:07:17,400 --> 00:07:20,400
Just join these words back together
to get the.

206
00:07:20,400 --> 00:07:23,166
Original format of the review,
you know, as a string.

207
00:07:23,166 --> 00:07:23,833
And therefore the.

208
00:07:23,833 --> 00:07:25,666
Way to do this is to.

209
00:07:25,666 --> 00:07:27,000
Well, you know, update.

210
00:07:27,000 --> 00:07:28,800
Our review variable. Again.

211
00:07:28,800 --> 00:07:31,166
By setting it equal to, well.

212
00:07:31,166 --> 00:07:31,900
The junction.

213
00:07:31,900 --> 00:07:34,466
Of these different words after we applied.

214
00:07:34,466 --> 00:07:35,466
Stemming to them.

215
00:07:35,466 --> 00:07:36,466
And the way to. Do this is.

216
00:07:36,466 --> 00:07:41,033
By calling the join function,
which takes as input.

217
00:07:41,033 --> 00:07:43,800
Exactly.
Well our list. You know, the. Review.

218
00:07:43,800 --> 00:07:47,233
Which contains a list of the different
words after stemming was applied to them.

219
00:07:47,700 --> 00:07:49,866
And however, if we. Join them, this.

220
00:07:49,866 --> 00:07:51,366
Way, this. Will, you know.

221
00:07:51,366 --> 00:07:53,000
Stick the words together.

222
00:07:53,000 --> 00:07:55,666
And we will end up with one word
that makes no sense.

223
00:07:55,666 --> 00:07:56,333
So in order.

224
00:07:56,333 --> 00:07:59,033
To separate these words by a space.

225
00:07:59,033 --> 00:08:03,566
You know, as in a classic string or text,
well, we just need to add here in quotes.

226
00:08:03,833 --> 00:08:06,200
And then inside the quotes
we add. A. Space.

227
00:08:06,200 --> 00:08:07,900
And then separating this.

228
00:08:07,900 --> 00:08:10,500
With the join with you by a dot.

229
00:08:10,500 --> 00:08:13,833
All right.
So this will. Not only join your.

230
00:08:13,833 --> 00:08:18,000
Review but also adding space
between each word of your review.

231
00:08:18,333 --> 00:08:19,266
All right.

232
00:08:19,266 --> 00:08:21,466
So good. Now
you know how to. Do this as well.

233
00:08:21,466 --> 00:08:24,233
And finally we have. A final step here.

234
00:08:24,233 --> 00:08:26,066
I'm sure you would guess what it is.
Otherwise.

235
00:08:26,066 --> 00:08:27,600
Please press pause to just.

236
00:08:27,600 --> 00:08:29,866
Finish. This. Cell implementation.

237
00:08:29,866 --> 00:08:31,200
But that's pretty obvious.

238
00:08:31,200 --> 00:08:32,366
All we did here.

239
00:08:32,366 --> 00:08:35,300
Was a transformation on a single review.

240
00:08:35,300 --> 00:08:36,400
And thanks to this for loop.

241
00:08:36,400 --> 00:08:36,966
Now all.

242
00:08:36,966 --> 00:08:40,133
The different reviews of our data
set will.

243
00:08:40,166 --> 00:08:42,566
Be transformed
and will be cleaned and essentially will.

244
00:08:42,566 --> 00:08:43,666
Be. Simplified.

245
00:08:43,666 --> 00:08:44,966
But now remember.

246
00:08:44,966 --> 00:08:46,533
We have this corpus here that.

247
00:08:46,533 --> 00:08:48,466
Was initialized as an empty list.

248
00:08:48,466 --> 00:08:49,933
And we want to add. All.

249
00:08:49,933 --> 00:08:51,433
These cleaned reviews into.

250
00:08:51,433 --> 00:08:54,433
The corpus,
because that's what will be expected.

251
00:08:54,500 --> 00:08:55,400
By the method of the.

252
00:08:55,400 --> 00:08:57,433
Class that will create the.

253
00:08:57,433 --> 00:08:58,733
Bag of words model.

254
00:08:58,733 --> 00:09:01,000
And therefore we must absolutely now.

255
00:09:01,000 --> 00:09:03,200
Add each review to the corpus.

256
00:09:03,200 --> 00:09:03,533
And the.

257
00:09:03,533 --> 00:09:06,766
Way to do
this is by taking our corpus list.

258
00:09:06,966 --> 00:09:07,833
And then.

259
00:09:07,833 --> 00:09:11,200
Call. The append function to append.

260
00:09:11,200 --> 00:09:14,200
Indeed the review.
And that's it my friends.

261
00:09:14,366 --> 00:09:16,966
We are done
with the cleaning of the texts.

262
00:09:16,966 --> 00:09:18,033
So now we're going to.

263
00:09:18,033 --> 00:09:19,866
Execute. This.
You know I think we deserve it.

264
00:09:19,866 --> 00:09:22,366
That was quite a. Long
cell implementation.

265
00:09:22,366 --> 00:09:23,100
So let's do this.

266
00:09:23,100 --> 00:09:24,300
However, my run time.

267
00:09:24,300 --> 00:09:27,133
Actually stopped so I have to restart it.

268
00:09:27,133 --> 00:09:29,233
By clicking,
you know, runtime and then. Restart.

269
00:09:29,233 --> 00:09:32,400
Runtime.
I will execute the two first cells.

270
00:09:32,400 --> 00:09:33,166
And then this one.

271
00:09:33,166 --> 00:09:36,100
And we'll make sure that everything works
correctly.

272
00:09:36,100 --> 00:09:37,000
All right.

273
00:09:37,000 --> 00:09:37,400
All good.

274
00:09:37,400 --> 00:09:38,766
Here let's do this.

275
00:09:38,766 --> 00:09:41,133
Let's first re-import the. Libraries.

276
00:09:41,133 --> 00:09:42,966
Done. Import the. Data set.

277
00:09:44,533 --> 00:09:45,300
Done as well.

278
00:09:45,300 --> 00:09:46,733
And now let's see.

279
00:09:46,733 --> 00:09:49,333
And let's hope
we. Didn't make any mistake.

280
00:09:49,333 --> 00:09:51,466
Let's play the cell. Cleaning.

281
00:09:51,466 --> 00:09:53,633
All the texts, all the reviews and.

282
00:09:53,633 --> 00:09:55,466
It seems to be all good. And yes.

283
00:09:55,466 --> 00:09:58,033
This is what we were supposed
to get. Perfect.

284
00:09:58,033 --> 00:10:00,066
So here it downloaded the, you know.

285
00:10:00,066 --> 00:10:02,200
Stopwords package from this.

286
00:10:02,200 --> 00:10:04,666
Path containing all the Nltk data.

287
00:10:04,666 --> 00:10:06,500
And also it's unzipped.

288
00:10:06,500 --> 00:10:08,200
This zip file. Containing.

289
00:10:08,200 --> 00:10:09,700
All the stopwords and perfect.

290
00:10:09,700 --> 00:10:12,233
And then of. Course,
it applied all. This cleaning process.

291
00:10:12,233 --> 00:10:13,500
To each of the reviews.

292
00:10:13,500 --> 00:10:14,666
And adding them to.

293
00:10:14,666 --> 00:10:16,300
The final corpus. Perfect.

294
00:10:16,300 --> 00:10:19,466
So now now
it is time for the next essential step.

295
00:10:19,633 --> 00:10:20,800
When doing sentiment.

296
00:10:20,800 --> 00:10:21,600
Analysis,

297
00:10:21,600 --> 00:10:25,633
it will be to create the bag of words
model, which will consist basically of.

298
00:10:25,633 --> 00:10:28,600
Creating this sparse matrix
containing now all.

299
00:10:28,600 --> 00:10:31,266
The words of the reviews
after they. Were cleaned.

300
00:10:31,266 --> 00:10:33,900
So we will take all the different words
of all the reviews.

301
00:10:33,900 --> 00:10:37,533
We will put them in different columns
of the sparse matrix, and then that will.

302
00:10:37,533 --> 00:10:40,700
Be actually our future matrix of features,
which.

303
00:10:40,700 --> 00:10:44,700
We will combine to the dependent variable
vector containing all the binary outcomes

304
00:10:44,700 --> 00:10:48,466
to train our future machine learning
model, which will be the Naive Bayes

305
00:10:48,466 --> 00:10:51,500
model, to learn the text and understand
whether the.

306
00:10:51,500 --> 00:10:53,966
Reviews are positive or negative.

307
00:10:53,966 --> 00:10:56,466
So I can't wait to proceed
to this next step.

308
00:10:56,466 --> 00:10:57,000
Creating the.

309
00:10:57,000 --> 00:10:57,966
Bag of Words model with.

310
00:10:57,966 --> 00:11:00,466
You and until
then, enjoy machine learning.
