1
00:00:00,450 --> 00:00:07,020
Oh, yes, and welcome back to another class of our course about the complete introduction to that science

2
00:00:07,020 --> 00:00:07,770
with Python.

3
00:00:08,520 --> 00:00:13,270
So basically in this course, we are going to talk about the basics of data science.

4
00:00:13,290 --> 00:00:15,450
So this part of the course is really theoretical.

5
00:00:15,840 --> 00:00:21,570
We are going to talk about the theory behind that science and everything that you guys need to know

6
00:00:21,570 --> 00:00:23,380
about this subject.

7
00:00:23,880 --> 00:00:24,890
So let's start.

8
00:00:25,470 --> 00:00:25,780
All right.

9
00:00:25,800 --> 00:00:27,750
So basically, what is data science?

10
00:00:27,750 --> 00:00:33,090
So when we ask someone what is exact data science or how it works or anything else?

11
00:00:33,090 --> 00:00:35,730
Well, it looks very fancy in the word.

12
00:00:35,730 --> 00:00:37,830
Data science looks really, really fancy.

13
00:00:38,610 --> 00:00:42,190
But basically, data science is simply the study of data.

14
00:00:42,540 --> 00:00:48,210
Yes, you have a lot of positions that can exist in data science to talk about them a bit later.

15
00:00:48,540 --> 00:00:56,190
But basically, data science is really the study of data so that science can start as early as the recording

16
00:00:56,190 --> 00:00:56,360
part.

17
00:00:56,400 --> 00:01:01,000
So let's say, for example, you have a project and you want to find out a solution to a certain project.

18
00:01:01,650 --> 00:01:08,640
So the first thing that you want to do is recording the necessary data to be able to answer the problematic

19
00:01:08,640 --> 00:01:09,690
inside of your project.

20
00:01:10,110 --> 00:01:16,260
So basically, let's say, I don't know, you want to know how many will you give a survey to some people

21
00:01:16,710 --> 00:01:19,080
and those people fill up the surveys.

22
00:01:19,080 --> 00:01:24,960
So let's say there are people who fill up the survey and I don't know, 100 percent say they like the

23
00:01:24,960 --> 00:01:25,550
color blue.

24
00:01:25,800 --> 00:01:29,790
And let's say of those 100 persons that like the color blue.

25
00:01:31,470 --> 00:01:34,780
I don't know, 80 percent said that they like hamburgers as well.

26
00:01:34,800 --> 00:01:40,220
So you can link up hamburgers with the color blue and say in majority of cases, people who like the

27
00:01:40,230 --> 00:01:42,440
color blue, like hamburgers.

28
00:01:42,450 --> 00:01:47,460
So you can, for example, create well, this could help you for marketing for a lot of purposes.

29
00:01:47,850 --> 00:01:50,730
But this is basically data science.

30
00:01:50,730 --> 00:01:54,180
Like this is a really basic example of what data science can do.

31
00:01:55,620 --> 00:01:58,680
So the study of that, I can start as early as in the recording phase.

32
00:01:58,680 --> 00:02:05,060
Recording phase could be, for example, giving people well, giving people questionnaires where they,

33
00:02:05,070 --> 00:02:09,260
for example, well, service, I mean, surveys, stories that people can take.

34
00:02:09,270 --> 00:02:12,870
So, for example, do like the color blue or the like the color red.

35
00:02:13,200 --> 00:02:17,350
And once again, those surveys are linked up with your problematics.

36
00:02:17,350 --> 00:02:18,270
So very important.

37
00:02:18,300 --> 00:02:22,230
You at first need to have it problematic that you want to solve.

38
00:02:22,620 --> 00:02:27,360
And for this problematic, you need a certain certain data from certain people and you will give them,

39
00:02:27,360 --> 00:02:28,200
for example, surveys.

40
00:02:28,200 --> 00:02:30,450
It could be surveys, it could be just ask them questions.

41
00:02:31,140 --> 00:02:34,530
There are plenty of ways that exist to be able to record data.

42
00:02:35,670 --> 00:02:41,100
The next step would be storing the data, because once you have the data, you need to store it somewhere.

43
00:02:41,100 --> 00:02:45,840
You don't know if you have the data physically, if, for example, it's on papers and you need to store

44
00:02:45,840 --> 00:02:46,400
this data.

45
00:02:47,160 --> 00:02:52,590
So you need really a storage room if it's going to be stored on the hard drive of your computer, if

46
00:02:52,590 --> 00:02:57,750
it's really huge amounts of data, maybe it's going to be stored somewhere on the server, maybe it's

47
00:02:57,750 --> 00:02:58,890
going to be stored on the cloud.

48
00:02:59,520 --> 00:03:03,780
So data science is really all those elements together.

49
00:03:04,410 --> 00:03:06,770
Next thing would be analyzing this data.

50
00:03:06,780 --> 00:03:11,820
So once again, transport, analyzing all the and the like sports or you want to take the raw data and

51
00:03:11,820 --> 00:03:16,380
you want to transform it into data that you can use to be able to fix your problem.

52
00:03:16,560 --> 00:03:17,520
Problematic, sorry.

53
00:03:18,060 --> 00:03:21,000
So basically this path is really, really important.

54
00:03:21,000 --> 00:03:26,380
And the majority of times people not only analyzing it, but they don't know the storing and recording

55
00:03:26,400 --> 00:03:26,630
part.

56
00:03:26,910 --> 00:03:34,250
But all this is part of data science and of the scale of the data data transformation.

57
00:03:34,260 --> 00:03:39,430
So the second part right here is just, well, another part of data science.

58
00:03:39,440 --> 00:03:42,920
So basically, data transformation could be part of the analyzing part right here.

59
00:03:43,890 --> 00:03:47,310
Basically, data transformation is part of data science.

60
00:03:47,310 --> 00:03:50,750
And in this case, this could be extracting the right information.

61
00:03:51,420 --> 00:03:57,480
So when we're talking about data science, as I said, we work with huge amounts of data so we can have

62
00:03:57,480 --> 00:04:05,090
millions, if not billions, sometimes data of peoples, depending of what type of problematic you want.

63
00:04:05,100 --> 00:04:10,380
If you want to fix, for example, we talking about marketing were a specialized, specialized ads.

64
00:04:10,770 --> 00:04:17,380
For example, when Google once when Google is able to give you like perfect ads for you, it will collect

65
00:04:17,400 --> 00:04:18,230
data on you.

66
00:04:18,240 --> 00:04:24,150
And with those patterns and with all the data that they collected, they took this raw data and they

67
00:04:24,150 --> 00:04:25,080
transformed it.

68
00:04:25,080 --> 00:04:31,290
And after that, they are able to know exactly what type of advertising they will give you.

69
00:04:32,100 --> 00:04:38,070
So really extracting the right information to be able to well after that, transform it, gain insight

70
00:04:38,070 --> 00:04:39,780
into useful clues to run efficiently.

71
00:04:39,780 --> 00:04:41,080
A project once again.

72
00:04:41,100 --> 00:04:43,950
Right now I'm talking about Google, but it could be anything else.

73
00:04:44,940 --> 00:04:50,940
It could be, for example, in the manufacturing field, for example, I don't know if you call it on

74
00:04:50,940 --> 00:04:56,160
the machine, for example, let's say the data is collected on a on real time.

75
00:04:56,160 --> 00:04:58,050
Let's say, for example, data is collected on real time.

76
00:04:58,410 --> 00:04:59,370
Let's say we spot a.

77
00:04:59,960 --> 00:05:06,980
Problematic each time that I don't know, there is I don't know, there is a certain peace that is created

78
00:05:07,280 --> 00:05:12,770
or there is one hundred pieces that are created on this machine and there is a problematic each one

79
00:05:12,770 --> 00:05:13,450
hundred pieces.

80
00:05:13,790 --> 00:05:20,360
Well, with data science, we are able to spot this problematic and say, OK, each time that this machine

81
00:05:20,360 --> 00:05:25,870
creates one hundred pieces of something, we need to make maintenance on this machine.

82
00:05:25,880 --> 00:05:33,120
If we don't want it to break, then that well, data science can also be used to predict certain outcomes.

83
00:05:33,410 --> 00:05:38,900
So once again, data science can be used in the planning of fields and predicting outcomes could be

84
00:05:38,900 --> 00:05:44,510
well, for example, in the banking world, in the financial world, it could be as well in the manufacturing

85
00:05:44,510 --> 00:05:44,930
world.

86
00:05:44,930 --> 00:05:49,940
In marketing, as I talked a bit earlier right now, this could be part of it.

87
00:05:49,970 --> 00:05:56,240
So, for example, giving specialized ads or ads that are perfect for you, this could work as well.

88
00:05:57,170 --> 00:05:59,050
So this is for what is that word?

89
00:05:59,050 --> 00:06:00,050
Data science is used.

90
00:06:01,520 --> 00:06:04,340
Data science is used in many, many, many fields.

91
00:06:05,540 --> 00:06:07,190
Those are just some examples.

92
00:06:07,220 --> 00:06:13,370
So, for example, for online businesses or e-commerce, which is really popular right now, we can

93
00:06:13,370 --> 00:06:16,670
say that a science could be used to recommend products.

94
00:06:17,060 --> 00:06:21,610
So product recommendation, better marketing as well.

95
00:06:21,860 --> 00:06:27,390
If, for example, you need the well, let's say a certain business wants to market the right products

96
00:06:27,390 --> 00:06:30,520
to the right audience, well, it will help them.

97
00:06:30,530 --> 00:06:35,290
So, as I said, if, for example, you go on Amazon right now and you start Googling certain products,

98
00:06:35,300 --> 00:06:37,500
you write down a certain product on Amazon.

99
00:06:37,910 --> 00:06:44,090
Well, Amazon will take this data and will give you products that are recommended based on your best

100
00:06:44,090 --> 00:06:44,630
research.

101
00:06:44,710 --> 00:06:49,940
So this is all data science, health care, for example.

102
00:06:49,940 --> 00:06:56,210
You can use that as science and health care in a lot of ways to, for example, virtual assistant.

103
00:06:56,750 --> 00:07:00,110
Let's say, for example, there are systems that are virtual, not right now.

104
00:07:00,350 --> 00:07:08,270
And this could be an amazing way to well, to use that as science, it could improve medical researches.

105
00:07:08,690 --> 00:07:12,490
Since we have a lot of data, we have a lot of knowledge about a certain subject.

106
00:07:12,500 --> 00:07:19,760
For example, if we see that, well, I don't know this medication, how well this medication tested,

107
00:07:19,940 --> 00:07:26,150
I don't know, 1000 people help them with this in this in this symptom, for example, this could be,

108
00:07:26,150 --> 00:07:31,790
well, pretty much a good thing for image analysis or in this case, picture or picture analysis.

109
00:07:32,310 --> 00:07:38,210
This could be a really good thing as well in the health care in the health care world, sorry, in the

110
00:07:38,210 --> 00:07:41,330
banking industry, for example, for fraud detection.

111
00:07:41,330 --> 00:07:46,370
Well, it's not an employee that is sitting in his office and looks for frauds.

112
00:07:46,700 --> 00:07:51,650
It's complex algorithms that are created and that works for fraud detection.

113
00:07:51,950 --> 00:07:57,440
And it's all about data science and machine learning, all the stuff to create credit scores, for example,

114
00:07:57,440 --> 00:08:01,490
and to give the right product to the consumers.

115
00:08:02,000 --> 00:08:03,980
Yeah, there is a little mistake right here.

116
00:08:04,190 --> 00:08:10,310
So giving the right product to the right consumer, especially in the banking world, sometimes it's

117
00:08:10,640 --> 00:08:17,480
well, it's not really easy because customers, some customers have millions of dollars or other customers

118
00:08:17,480 --> 00:08:18,620
don't have money at all.

119
00:08:18,980 --> 00:08:22,130
So it's really well, it's completely it's it's really different.

120
00:08:22,130 --> 00:08:27,890
And you can't offer a type of product that you offer to the one to the other one for transport, for

121
00:08:27,890 --> 00:08:28,210
example.

122
00:08:28,430 --> 00:08:28,670
Yes.

123
00:08:28,670 --> 00:08:32,480
Optimization in this case, it's going to be let's say, for example, you want to go for place from

124
00:08:32,480 --> 00:08:36,170
place to place B, how to go the fastest way possible.

125
00:08:36,170 --> 00:08:38,390
And they're really Real-Time optimization.

126
00:08:38,660 --> 00:08:40,430
So in this case, it's all about data science.

127
00:08:40,430 --> 00:08:45,590
For example, it will know that there is traffic at this place and it will not suggest you to pass by

128
00:08:45,590 --> 00:08:46,640
this place or this place.

129
00:08:47,240 --> 00:08:53,450
Self-driving cars could also be an example of how data science can be used, because once again, this

130
00:08:53,450 --> 00:09:00,010
is a huge amount of data that is collected to make the car driving alone personalised virtual assistant.

131
00:09:00,020 --> 00:09:03,110
Once again, it's like the health care part for manufacturing.

132
00:09:03,110 --> 00:09:05,420
As I said before, we have problem detection.

133
00:09:05,420 --> 00:09:10,670
That could be really an amazing way and the problem of prevention as well as automation.

134
00:09:10,670 --> 00:09:13,280
So let's say, for example, you want to automate your production.

135
00:09:13,700 --> 00:09:15,410
It's possible with data science.

136
00:09:15,410 --> 00:09:21,590
So this so for example, with all the data that is collected, you are able to hire engineers as well

137
00:09:21,710 --> 00:09:28,820
that are engineers will be able to create machines that will automatically work and will automatically,

138
00:09:28,820 --> 00:09:33,110
let's say, for example, well, work on huge skills.

139
00:09:33,870 --> 00:09:36,470
So let's just take a basic example.

140
00:09:36,890 --> 00:09:42,470
Let's say a huge business that is, I don't know, putting water inside of bottles.

141
00:09:43,070 --> 00:09:44,330
So it's all automated.

142
00:09:44,330 --> 00:09:47,540
There is no one person that is like putting water in each bottle.

143
00:09:47,810 --> 00:09:51,650
It's really a huge machine that is automated because of the use of data science.

144
00:09:51,650 --> 00:09:53,450
Once again in finances.

145
00:09:53,450 --> 00:09:59,780
It's really, really used in algorithms, algo trading, algorithms, algorithm trading and that.

146
00:09:59,860 --> 00:10:06,940
This is all algorithms that are created based on past performances of a certain stock or ETF or whatever,

147
00:10:07,210 --> 00:10:15,550
and it's based on all this data that algo traders are able to create programs that will trade by themselves

148
00:10:15,790 --> 00:10:22,930
and really control the race, creed, all the strategies around the around their trading style, which

149
00:10:22,930 --> 00:10:24,490
is pretty much a very cool thing.

150
00:10:25,060 --> 00:10:30,340
So those are some users of data science, but it can also be used in engineering.

151
00:10:30,550 --> 00:10:33,270
It can also be used in agriculture.

152
00:10:33,520 --> 00:10:35,560
It's used pretty much everywhere.

153
00:10:35,560 --> 00:10:38,260
And it's really part of our lives right now.

154
00:10:39,880 --> 00:10:40,270
All right.

155
00:10:40,330 --> 00:10:43,490
So next thing is about the data scientists.

156
00:10:43,510 --> 00:10:46,130
So basically, what are the skills that you will need to have?

157
00:10:46,930 --> 00:10:50,710
So for me, once again, this is based on my personal experience.

158
00:10:51,200 --> 00:10:53,050
You need the three skills.

159
00:10:53,050 --> 00:10:57,420
Even if you don't necessarily have them now, you can work on them.

160
00:10:57,430 --> 00:11:04,230
So basically logical thinking and mathematical understanding so that science is all about mathematics.

161
00:11:04,240 --> 00:11:06,120
So there is a lot of numbers.

162
00:11:06,130 --> 00:11:09,300
There is a lot of well, you work always with numbers.

163
00:11:09,640 --> 00:11:14,590
So if you're bad in mathematics, well, I'm sorry to say that, but it's going to be a bit hard for

164
00:11:14,590 --> 00:11:14,710
you.

165
00:11:15,490 --> 00:11:21,280
So really having a good mathematical understanding and logical thinking because mathematical understanding

166
00:11:21,280 --> 00:11:22,650
is linked with logical thinking.

167
00:11:23,020 --> 00:11:25,300
So you need to understand basic logical concepts.

168
00:11:25,480 --> 00:11:31,030
And with those basic concepts through that, you can put them on the higher and higher and higher skills

169
00:11:31,300 --> 00:11:33,430
and create something that is pretty cool.

170
00:11:33,430 --> 00:11:35,480
But just understanding the basics sometimes.

171
00:11:35,500 --> 00:11:41,110
Well, the moment you understand the basics, you will be able to go well further and further and further

172
00:11:42,220 --> 00:11:47,560
understanding the business world around us, because there are science is not just a thing that you

173
00:11:47,560 --> 00:11:53,440
play with, that science is really something that you will use to solve real problem issues.

174
00:11:54,970 --> 00:11:56,170
Well, an example.

175
00:11:56,170 --> 00:12:01,660
Let's say, for example, you want well, you the first thing the first step in data science is really

176
00:12:01,660 --> 00:12:03,600
to find out the problematic that you can solve.

177
00:12:03,970 --> 00:12:08,070
You will not just collect data to collect data because this don't make sense at all.

178
00:12:08,080 --> 00:12:12,760
This is this would be raw data that will not be used and doesn't make sense.

179
00:12:12,760 --> 00:12:18,130
Just collecting data for collecting data if you collect data is to improve something is to reach a certain

180
00:12:18,130 --> 00:12:19,720
goal or a certain objective.

181
00:12:20,050 --> 00:12:23,980
This is why you need an understanding of the business that is around this world, the business world

182
00:12:23,980 --> 00:12:29,230
that is around us, because in the majority of times the use of data science would be done inside of

183
00:12:29,230 --> 00:12:29,740
business.

184
00:12:29,740 --> 00:12:34,950
World's interest in technology and programming, data science nowadays.

185
00:12:34,960 --> 00:12:40,450
Well, mathematics is the first word, but programming and technologies are the second part.

186
00:12:40,600 --> 00:12:46,840
As I said, that science is not only in the analysing part where you analyze the numbers, it's well,

187
00:12:46,840 --> 00:12:53,740
it starts from the moment that you guys will give surveys, for example, and you collect data and store

188
00:12:53,740 --> 00:12:54,040
data.

189
00:12:54,040 --> 00:12:57,040
So you need to understand all the technological parts around this.

190
00:12:57,790 --> 00:13:02,950
What what are the things that you can use, for example, to store data, how you can collect your data,

191
00:13:03,130 --> 00:13:07,200
and so that all the techniques that you can use to analyze properly your data?

192
00:13:07,270 --> 00:13:07,810
Yes.

193
00:13:08,260 --> 00:13:13,630
The more you get well, the more you work with that data science and the more you analyze data, the

194
00:13:13,630 --> 00:13:18,970
more you will get you will create your own techniques and you will have your own ways to analyze data.

195
00:13:19,360 --> 00:13:22,900
And you will see it's not something that is, well, that complicated at the end.

196
00:13:23,170 --> 00:13:25,030
It's just logical thinking.

197
00:13:25,030 --> 00:13:29,500
Well, it's 100 percent logical thinking on this thing that assigns jobs.

198
00:13:29,500 --> 00:13:33,970
Well, you can have a lot of data science jobs that are not listed here are just listed.

199
00:13:33,970 --> 00:13:34,780
Three examples.

200
00:13:34,780 --> 00:13:40,510
For example, that analysis database administrator or business analysis, those are the jobs that come

201
00:13:40,510 --> 00:13:43,390
to me like quickly right this right now.

202
00:13:43,390 --> 00:13:49,720
But you could have, for example, data engineer, you can have data scientist, you can have business

203
00:13:49,720 --> 00:13:55,120
analysts, you can have statistics that this statistician, statistician, statistician.

204
00:13:56,830 --> 00:14:02,590
So there are plenty of other jobs that are linked with the data science, which is pretty cool.

205
00:14:02,590 --> 00:14:04,930
So you'll see this field is already pretty cool.

206
00:14:07,180 --> 00:14:08,140
So the basics of data.

207
00:14:08,150 --> 00:14:13,960
So right now we are going to talk a bit more in depth about the basics of data and the data lifecycle.

208
00:14:14,320 --> 00:14:20,980
And basically, in this case, there is six steps that we need to follow for data.

209
00:14:20,980 --> 00:14:23,920
So just for you guys to understand how exactly it works.

210
00:14:24,580 --> 00:14:25,390
So it's pretty simple.

211
00:14:25,390 --> 00:14:29,410
The first step would be, let's say, for example, you are in the business in the first step will be

212
00:14:29,740 --> 00:14:32,490
identifying the problematic that you want to solve.

213
00:14:33,490 --> 00:14:36,180
So let's say, for example, I don't know you.

214
00:14:36,340 --> 00:14:38,380
Let's see, just a really basic example.

215
00:14:38,500 --> 00:14:44,230
Let's take our example of Hamburger's you want to create a new packaging for your hamburger.

216
00:14:44,230 --> 00:14:48,760
So let's say, for example, you working for, I don't know, a hamburger store and you have thousands

217
00:14:48,760 --> 00:14:54,300
and thousands of consumers and you give them I service, let's say it's online surveys and they fill

218
00:14:54,730 --> 00:14:55,750
fill up their servers.

219
00:14:55,750 --> 00:14:59,500
So you have your program at your first step will be identifying.

220
00:14:59,660 --> 00:15:00,480
Your problematic.

221
00:15:00,520 --> 00:15:09,530
So your problem is I want to create another packaging for my hamburgers, then the next step would be

222
00:15:09,530 --> 00:15:17,450
finding a way to collect or acquired data so you can collect yourself through that or you can ask a

223
00:15:17,450 --> 00:15:19,730
business to acquire data for you.

224
00:15:19,760 --> 00:15:21,340
Yes, you need to pay a certain fee.

225
00:15:21,620 --> 00:15:28,040
But once again, if you are a big corporation, you don't really care about paying those fees because

226
00:15:28,280 --> 00:15:33,240
at the end of the day, you will make profit because of all this research around it.

227
00:15:34,190 --> 00:15:38,580
So the second step will be finding a way to collect this data.

228
00:15:39,170 --> 00:15:44,620
So the first step well, the first thing that you need to understand here is what data you need.

229
00:15:44,630 --> 00:15:50,720
So basically, you need to understand the data that you guys will need to collect what it because if,

230
00:15:50,720 --> 00:15:54,950
for example, you want to change your packaging, you will not ask your consumers, do you like it when

231
00:15:54,950 --> 00:15:55,960
it rains outside?

232
00:15:56,120 --> 00:15:56,930
It doesn't make sense.

233
00:15:57,800 --> 00:16:02,720
So, for example, you will ask them, hey, do you prefer blue or red, for example?

234
00:16:03,650 --> 00:16:06,140
So they will answer the question.

235
00:16:07,190 --> 00:16:11,690
Next step would be thinking about where I can find this data.

236
00:16:11,720 --> 00:16:13,040
So once again, you can acquire it.

237
00:16:13,040 --> 00:16:15,020
You can buy it from, let's say, another business.

238
00:16:15,020 --> 00:16:16,910
Already did this research.

239
00:16:17,240 --> 00:16:18,720
You can buy it from the business.

240
00:16:18,740 --> 00:16:22,040
Well, you can buy it from the company that did the research for that business.

241
00:16:22,400 --> 00:16:24,920
Or you can go and collect the data by yourself.

242
00:16:25,910 --> 00:16:31,520
And then the most efficient way to obtain and store this data, because once again, this is part of

243
00:16:31,550 --> 00:16:32,180
the first step.

244
00:16:32,750 --> 00:16:38,510
So you'll think about a good way to tell an efficient way to collect is that if you want to collect

245
00:16:38,510 --> 00:16:41,420
it by yourself, you will think about an efficient way to collect that.

246
00:16:41,420 --> 00:16:46,220
I could be, for example, giving surveys to your consumers where they come when they come to your restaurant,

247
00:16:46,670 --> 00:16:51,380
or it could be sending them emails where you tell them, hey, you get a free hamburger if you, for

248
00:16:51,380 --> 00:16:53,600
example, fill up this survey.

249
00:16:54,980 --> 00:16:57,020
All right, that would be data transformation.

250
00:16:57,020 --> 00:17:03,990
And in this case, you want to put your data in to the right format and you will clean your data.

251
00:17:04,010 --> 00:17:07,550
So let's say, for example, you have surveys that are not filled up to the maximum.

252
00:17:08,240 --> 00:17:11,810
You want to clean this data, all the missing values, you want to take them away.

253
00:17:12,380 --> 00:17:15,070
You want to come compute all the data.

254
00:17:15,080 --> 00:17:20,970
So put everything together of the corrupted data, all the applications you want to take them away.

255
00:17:20,990 --> 00:17:25,370
So let's say, for example, someone to fill that up, the survey like five times because he wanted

256
00:17:25,370 --> 00:17:26,530
50 hamburgers.

257
00:17:27,050 --> 00:17:32,980
Well, you only counted as one person and removing unnecessary data.

258
00:17:33,230 --> 00:17:36,760
So let's say someone just wrote, I don't know anything inside of history.

259
00:17:36,770 --> 00:17:38,440
Well, you just take him away as well.

260
00:17:40,130 --> 00:17:47,810
Next step is when you have well, when you have all your data ready this that you have all your raw

261
00:17:47,810 --> 00:17:48,330
material.

262
00:17:48,350 --> 00:17:50,220
So in this case, you have all your raw data.

263
00:17:50,540 --> 00:17:54,420
Now it's time to understand it and transform it and create the solution to your problem.

264
00:17:54,710 --> 00:17:57,850
So once you have all the raw data, you want to understand it.

265
00:17:57,860 --> 00:18:01,180
So you will make hypotheses.

266
00:18:01,460 --> 00:18:06,250
So say, for example, you will say, hey, that's a thing that we talked about before.

267
00:18:06,440 --> 00:18:11,990
So, hey, one hundred percent said they love the color blue and 80 percent said they like hamburgers.

268
00:18:12,020 --> 00:18:17,960
So basically, if it's if it's this way, this means that everybody eats a hamburger or the majority

269
00:18:17,960 --> 00:18:20,970
of people who are eating a hamburger likes the color blue.

270
00:18:21,080 --> 00:18:26,310
So basically, maybe this could be a good thing to have a blue packaging for my hamburger.

271
00:18:26,660 --> 00:18:33,130
So this would be making a hypothesis and at the same time understanding the data pattern.

272
00:18:33,150 --> 00:18:34,400
So we get our data.

273
00:18:34,400 --> 00:18:38,480
So we have our data and we understand that this brings us to this.

274
00:18:38,480 --> 00:18:40,870
And this is where you really need logical thinking.

275
00:18:40,880 --> 00:18:42,640
Once again, this is a really basic problem.

276
00:18:42,950 --> 00:18:44,780
Sometimes they're way more complicated.

277
00:18:45,560 --> 00:18:48,590
So this would be for the fourth step, really understanding the data.

278
00:18:49,620 --> 00:18:51,770
So hypothesis you will make them there.

279
00:18:51,770 --> 00:18:58,010
And once again, understanding of the patterns around the data that you have a step five will be modeling

280
00:18:58,010 --> 00:18:58,550
your data.

281
00:18:58,580 --> 00:19:01,290
So once again, you will create a model around this data.

282
00:19:01,580 --> 00:19:05,870
So in this basic example, it will be, for example, creating packages.

283
00:19:06,150 --> 00:19:07,990
So let's say you create a blue package.

284
00:19:08,000 --> 00:19:12,430
In this case, you want to evaluate the efficiency of your model.

285
00:19:12,440 --> 00:19:17,450
So let's say, for example, you create those packages and you give them to your consumers.

286
00:19:17,450 --> 00:19:20,270
And if they're happy, well, this would be a good way.

287
00:19:20,300 --> 00:19:21,440
So once again, you will deploy.

288
00:19:21,440 --> 00:19:27,440
But once again, since our example is really simple, let's say we keep we'll with this example, you

289
00:19:27,440 --> 00:19:29,110
will deploy, but only small scale.

290
00:19:29,390 --> 00:19:31,850
So let's say, for example, you will deploy one day a week.

291
00:19:31,850 --> 00:19:37,310
So you will have your new packaging one day a week and you will give them the give it to your new consumers.

292
00:19:38,120 --> 00:19:39,520
But they will be able to see it.

293
00:19:39,530 --> 00:19:40,700
They will be able to use it.

294
00:19:40,700 --> 00:19:42,610
And who knows, maybe it works perfectly.

295
00:19:43,850 --> 00:19:47,950
So when all this works perfectly, so you're in your step five.

296
00:19:47,960 --> 00:19:53,300
If you see that your consumers are happy, your sales are going up because you change your clothes,

297
00:19:53,300 --> 00:19:57,380
you change the packaging of burgers and they're blue now and everybody is happy.

298
00:19:57,560 --> 00:19:59,320
It's time to go to the store.

299
00:19:59,430 --> 00:20:04,270
Six, which is launching the solution or launching your model on a larger scale.

300
00:20:04,590 --> 00:20:09,690
So let's say you have a few restaurants, so instead of launching your model on only one restaurant,

301
00:20:10,020 --> 00:20:15,110
you will launch it on more than one restaurant or inside of all your restaurants.

302
00:20:16,680 --> 00:20:19,320
The best way to do this is really going step by step.

303
00:20:19,620 --> 00:20:25,200
So when you reach the step five, you launched your model or your solution inside of a small scale.

304
00:20:25,950 --> 00:20:28,930
And the next thing would be inside of it larger scale.

305
00:20:29,220 --> 00:20:31,130
So you always want to limit your risk.

306
00:20:31,140 --> 00:20:39,690
So never lunch at a whole scale straight up, because what can happen is that you could have world problems

307
00:20:39,690 --> 00:20:43,290
where anything could happen and your solution will just break.

308
00:20:43,300 --> 00:20:49,890
So you lunch on a small scale and you slowly grew your lunch and everything if everything goes the right

309
00:20:49,890 --> 00:20:50,070
way.

310
00:20:50,850 --> 00:20:53,200
This means that you can just launch it faster.

311
00:20:53,670 --> 00:20:57,180
So those are the six steps inside of that as well.

312
00:20:57,990 --> 00:20:59,260
The basics of data.

313
00:20:59,610 --> 00:21:01,570
Those are the six steps that we need to follow.

314
00:21:01,590 --> 00:21:08,430
So really, from finding the problematic to launching the solution, once again, those are a really

315
00:21:08,430 --> 00:21:09,180
basic example.

316
00:21:10,860 --> 00:21:14,160
So this is really for the basics of that right now.

317
00:21:14,170 --> 00:21:16,440
We will jump inside of the course.

318
00:21:16,440 --> 00:21:20,550
We are going to start talking about the tools that are used inside of the restaurants.

319
00:21:20,820 --> 00:21:25,980
And you'll see basically this course will teach you all the basics about the subject and you will have

320
00:21:26,010 --> 00:21:32,370
a lot more knowledge when this course will be done so that first class guys in C all in our next class.