1 00:00:00,120 --> 00:00:06,570 Hello, all, if you want to become a data scientist or engineer, then you have to showcase your practical 2 00:00:06,570 --> 00:00:14,610 skills and the best way to showcase your practical skills by solving some real world use cases, by 3 00:00:14,610 --> 00:00:21,630 solving some real life challenges, because understanding machine learning, understanding data science 4 00:00:21,630 --> 00:00:29,070 is never going to help out until unless you don't have your technical skills under Ullas, you don't 5 00:00:29,070 --> 00:00:31,290 have your own projects on your resume. 6 00:00:31,560 --> 00:00:38,640 That's why in the schools we all have come up with these three real world challenges, three real world 7 00:00:38,640 --> 00:00:46,080 projects for you that will definitely help you and that will increase your chance of being recruited 8 00:00:46,080 --> 00:00:54,690 in some interviews of top notch based companies like Amazon, Google, Netflix, Facebook and in some 9 00:00:54,690 --> 00:00:56,370 top MNC as well. 10 00:00:56,940 --> 00:01:04,290 The very first project of machine learning that you guys are going to consider in this practical course 11 00:01:04,440 --> 00:01:07,490 is all about hotel booking prediction. 12 00:01:07,680 --> 00:01:13,880 So here we have to protect rather the particular booking is going to cancel or not. 13 00:01:13,920 --> 00:01:20,940 So in this entire project, this is exactly our main strategy, that if we are going to collect some 14 00:01:20,940 --> 00:01:26,310 data after it, what we are going to do, we are going to pre-process that data. 15 00:01:26,340 --> 00:01:32,280 We are going to perform some data cleaning on the data because in real world, what our data we will 16 00:01:32,280 --> 00:01:40,020 get, we will always get our raw data and we have to make that data ready for our machine learning purposes 17 00:01:40,020 --> 00:01:47,100 after we have to perform something known as exploratory data analysis to understand what data, what 18 00:01:47,100 --> 00:01:48,570 exactly is going inside. 19 00:01:48,570 --> 00:01:55,890 In my data, after what we have to do, we have to apply various techniques of feature engineering and 20 00:01:55,980 --> 00:02:02,280 see if we have to encode our categorical data, which is exactly known as future encoding operate. 21 00:02:02,280 --> 00:02:06,000 We have to do something known as our glide detection operator. 22 00:02:06,000 --> 00:02:13,110 We have to select some important features for our model that contribute most to the dependent features 23 00:02:13,110 --> 00:02:15,750 by doing some statistical techniques. 24 00:02:15,750 --> 00:02:22,260 And after doing all these systems that we are going to do prediction using Savrin machine learning algorithm, 25 00:02:22,260 --> 00:02:28,620 depending upon what use case we have, that we have classification, whether we have regression or whether 26 00:02:28,620 --> 00:02:29,830 we have a cluster in use. 27 00:02:30,630 --> 00:02:36,390 So this is going to be your entire life cycle of the Machine Learning Project. 28 00:02:36,420 --> 00:02:45,300 So a second use case is exactly my healthcare use case in which we are going to consider data of those 29 00:02:45,300 --> 00:02:48,680 patient that have kidney disease or not. 30 00:02:49,050 --> 00:02:57,270 So we have to predict whether a particular patient or a particular person having all these different 31 00:02:57,270 --> 00:03:03,720 different properties can have a chronic kidney disease or not using some classification algorithms. 32 00:03:03,750 --> 00:03:07,050 And what we have to do, we have to understand our data as well. 33 00:03:07,140 --> 00:03:14,120 And from this huge chunk of data, we have to analyze it and extract some meaningful insights from it. 34 00:03:14,130 --> 00:03:21,630 And the third, you will see this is exactly most demanding use case, which is exactly my airline industry 35 00:03:21,630 --> 00:03:22,390 use case. 36 00:03:22,410 --> 00:03:26,850 So in this case, we have to consider data of flights. 37 00:03:27,060 --> 00:03:34,020 So we have to understand this data, how we can understand by analyzing this data, by extracting some 38 00:03:34,020 --> 00:03:41,700 amazing insights and add that we have to build such a machine learning model that can predict what exactly 39 00:03:41,700 --> 00:03:48,330 it can with a fear of the takers of airlines on different different routes, having all these different 40 00:03:48,330 --> 00:03:49,690 different features. 41 00:03:49,920 --> 00:03:56,660 So we have to build such a cool and fancy machine learning model for this use case as well. 42 00:03:56,820 --> 00:04:03,750 So for what you guys are waiting for, just with me and have a phone for all the upcoming sessions.