1 00:00:00,360 --> 00:00:05,820 Hello and welcome to this new section which is data analysis using Python. 2 00:00:06,030 --> 00:00:10,320 So in this section we will learn how to perform data analysis on a set of data. 3 00:00:10,440 --> 00:00:16,440 So this is an intellectual when we will learn what is data analysis and why Python is actually used 4 00:00:16,440 --> 00:00:18,190 to perform data analysis. 5 00:00:19,920 --> 00:00:24,080 So let's first go ahead and understand what actually data analysis is. 6 00:00:24,210 --> 00:00:30,960 So the simple definition of data analysis is analyzing data from multiple sources to extract meaningful 7 00:00:30,960 --> 00:00:32,280 information. 8 00:00:32,280 --> 00:00:37,920 So in data analysis what we do is that we don't mind some data source which we have to analyze and that 9 00:00:37,920 --> 00:00:43,290 is let's say you have a website or an e-commerce Web site which is going to generate tons of data on 10 00:00:43,290 --> 00:00:44,640 a daily basis. 11 00:00:44,670 --> 00:00:50,370 So if you want to analyze that data you simply extract data from that source you perform data analysis 12 00:00:50,370 --> 00:00:54,550 on it using some programming using some language like Python. 13 00:00:54,660 --> 00:01:00,210 And then after doing a data analysis you get some meaningful information which you could use to make 14 00:01:00,210 --> 00:01:01,740 better decisions. 15 00:01:01,770 --> 00:01:05,480 So that is the basic and the most simplest definition of data analysis. 16 00:01:05,489 --> 00:01:08,250 And obviously we could go into more complicated stuff. 17 00:01:08,340 --> 00:01:12,020 But for now this is what the data analysis is. 18 00:01:12,300 --> 00:01:16,230 Now let's learn why didn't Alliss is required. 19 00:01:16,260 --> 00:01:20,950 So the first and foremost reason is that it helps us make better decisions. 20 00:01:20,970 --> 00:01:27,810 So a lot of organizations like let's say the e-commerce Web site Amazon has to make business decisions 21 00:01:27,840 --> 00:01:29,210 on a daily basis. 22 00:01:29,280 --> 00:01:33,720 So these organizations actually rely upon the data which they get. 23 00:01:33,720 --> 00:01:39,570 So for example if you have a e-commerce Web site like Amazon there are going to be a ton of users visiting 24 00:01:39,570 --> 00:01:44,550 your Web site on a daily basis and what they are going to do is that they are going to generate a large 25 00:01:44,550 --> 00:01:46,200 amount of raw data. 26 00:01:46,200 --> 00:01:52,590 Now this can actually be analyzed and information could be drawn from that data and that helps Amazon 27 00:01:52,620 --> 00:01:57,660 or any e-commerce website for that matter to make better decisions. 28 00:01:57,660 --> 00:02:01,350 And the next most important thing is best use of available data. 29 00:02:01,430 --> 00:02:07,740 Now not only Amazon but any data intensive organization is going to produce a large number of data in 30 00:02:07,740 --> 00:02:12,300 the backend and throwing away that data does not make any sense. 31 00:02:12,330 --> 00:02:18,360 And thats why this data is actually analyzed so that some conclusions could be drawn from that data 32 00:02:18,390 --> 00:02:21,240 which could be more valuable to the organization. 33 00:02:21,240 --> 00:02:25,840 So this is the main purpose of data analysis and there would be a lot more. 34 00:02:25,860 --> 00:02:32,730 But the basic purpose of a data analysis or the basic significance of data analysis is to help make 35 00:02:32,730 --> 00:02:38,640 better decisions for organizations as well as to make the best use of the available data. 36 00:02:39,150 --> 00:02:41,220 So why data analysis is the future. 37 00:02:41,340 --> 00:02:46,730 So what Ive done here is that I've laid out the basic set up of an economist company or you could see 38 00:02:46,840 --> 00:02:51,840 e-commerce Web site and we will learn how data is used by this organization. 39 00:02:51,840 --> 00:02:56,520 So as you could see we have the e-commerce Web site at the center and on the left hand side you could 40 00:02:56,520 --> 00:02:57,520 see user. 41 00:02:57,600 --> 00:03:03,660 So what this user actually does is that he visits the e-commerce Web site and he basically browse through 42 00:03:03,690 --> 00:03:09,810 or goes through the commerce Web site he might view certain products he might buy certain products and 43 00:03:09,840 --> 00:03:13,790 he might carry out a bunch of activities on the e-commerce Web site. 44 00:03:13,800 --> 00:03:19,470 So what the e-commerce Web site actually does is that it generates a real user data that is nothing 45 00:03:19,470 --> 00:03:21,500 but the user's browsing history. 46 00:03:21,510 --> 00:03:24,580 It also stores the products viewed by the user. 47 00:03:24,600 --> 00:03:29,880 It also stores a bunch of things like which particular product they use bought and everything like that. 48 00:03:29,880 --> 00:03:35,400 So every time a user visits on an e-commerce Web site or our data is generated at the back end. 49 00:03:35,460 --> 00:03:39,510 And just imagine this thing with a big site like Amazon. 50 00:03:39,510 --> 00:03:45,710 So there are going to be a million users visiting per day on Amazon who generate raw data. 51 00:03:45,840 --> 00:03:51,900 And with this what we could do is that we could perform data analysis on the raw data and we would find 52 00:03:51,900 --> 00:03:57,560 certain things like which product is being sold and which category or actually the most viewed products 53 00:03:57,570 --> 00:04:03,000 by the user which are the trending products which products they use are not buying and everything like 54 00:04:03,000 --> 00:04:03,550 that. 55 00:04:03,570 --> 00:04:10,320 So you could use that information that is let's assume that from the data analysis we came to know that 56 00:04:10,350 --> 00:04:17,089 Apple products are being sold most in the mobile category so this information can be used to make decisions 57 00:04:17,100 --> 00:04:21,339 like OK as the users are buying Apple products in the mobile category. 58 00:04:21,450 --> 00:04:26,760 Let's place that product at the top and let's place the other products at the bottom. 59 00:04:26,760 --> 00:04:34,290 So in such a way that raw data can be converted into useful information using data analysis and later 60 00:04:34,380 --> 00:04:38,400 that information can be used to make logical conclusions. 61 00:04:38,400 --> 00:04:43,210 And depending upon those conclusions we could optimize e-commerce website. 62 00:04:43,290 --> 00:04:49,230 Now this was one of the most simplest example of data analysis and obviously Ive always simplified things 63 00:04:49,230 --> 00:04:49,570 here. 64 00:04:49,620 --> 00:04:55,410 A lot of them things actually happen in the back end but this is what generally happens when you perform 65 00:04:55,410 --> 00:04:57,370 data analysis on the raw data. 66 00:04:57,540 --> 00:05:02,400 So now the next thing which we will learn is that why Python for data analysis. 67 00:05:02,430 --> 00:05:07,920 So actually there are a lot of options to perform data analysis you could use tons of tools or tons 68 00:05:07,920 --> 00:05:10,140 of other languages as well. 69 00:05:10,320 --> 00:05:16,310 But the reason for using Python for the analyst is that its faster as compared to other alternatives. 70 00:05:16,320 --> 00:05:22,430 Now when you talk about data one obvious program comes into mind is Excel and Excel is actually good 71 00:05:22,430 --> 00:05:28,010 at analyzing data but Excel is only suited when that data is ready less. 72 00:05:28,310 --> 00:05:33,920 So let's see if you have hundred rows of data that's ok you could analyze that with Excel but then you 73 00:05:33,920 --> 00:05:39,360 have a million user data like Amazon or any other e-commerce Web site for that matter. 74 00:05:39,380 --> 00:05:45,970 Then using axil makes no sense because it's obviously too slow and there are not a lot of tools in Excel 75 00:05:46,090 --> 00:05:49,450 which you could use to perform meaningful data analysis. 76 00:05:49,460 --> 00:05:53,860 So Python is comparatively much faster as compared to excel or the Internet. 77 00:05:53,900 --> 00:05:58,580 And that's why Python is actually used to perform data analysis. 78 00:05:58,580 --> 00:06:03,550 The next most important thing about Python is that it has a simple syntax. 79 00:06:03,620 --> 00:06:09,560 No you won't if someone is not a programmer or if he's not from the computer science background even 80 00:06:09,560 --> 00:06:13,670 then he could learn Python because it has a simpler syntax. 81 00:06:13,670 --> 00:06:19,810 And that's another reason why most people prefer Python for performing data analysis and the most important 82 00:06:19,820 --> 00:06:26,820 and the last reason for using Python for the Donnellys is that it has multiple libraries for data analysis. 83 00:06:26,840 --> 00:06:33,620 So Python actually has a bunch of tools or a bunch of options for the Daniels's like by Bandar's and 84 00:06:33,620 --> 00:06:34,520 everything like that. 85 00:06:34,520 --> 00:06:39,650 So we're actually going to learn about non-plan pandas and the upcoming lectures as well. 86 00:06:39,650 --> 00:06:44,370 So this is the main reason why we choose Python mostly for data analysis. 87 00:06:44,660 --> 00:06:50,660 So that's it for this lecture and hopefully you guys will do understand what the time analysis is how 88 00:06:50,660 --> 00:06:57,650 data analysis is performed and what is the significance of a data analysis and also why Python is chosen 89 00:06:57,650 --> 00:07:00,920 as the best language for analyzing data. 90 00:07:00,920 --> 00:07:05,010 So thank you very much for watching and I'll see you guys next time. 91 00:07:05,150 --> 00:07:05,650 Thank you.