1 00:00:01,090 --> 00:00:08,640 Here are the big words that are taking the industry by storm machine learning now. 2 00:00:08,760 --> 00:00:14,190 Throughout the course we're going to explore this topic as well as data science and how it all fits 3 00:00:14,550 --> 00:00:16,480 into the tech space. 4 00:00:17,220 --> 00:00:22,980 But first things first you need to be able to explain this topic to your co-workers and not look like 5 00:00:23,010 --> 00:00:24,650 you don't know what you're talking about. 6 00:00:24,690 --> 00:00:26,380 So let's start with this. 7 00:00:26,400 --> 00:00:30,240 You see machines or computers. 8 00:00:30,240 --> 00:00:35,220 In this case are really really good at certain things right. 9 00:00:35,220 --> 00:00:42,900 Machines can perform tasks really really fast and we can control these machines and get them to do tasks 10 00:00:42,930 --> 00:00:48,480 for us which is what programming is we programmed computers. 11 00:00:48,570 --> 00:00:53,200 We give them instructions to do tasks and they do it for us. 12 00:00:53,220 --> 00:00:58,590 You see computers came into this world because it allowed humans to do certain tasks really fast. 13 00:00:58,650 --> 00:01:02,380 That may have required hundreds of workers before. 14 00:01:02,580 --> 00:01:09,000 I mean computers used to actually mean people who do tasks that compute. 15 00:01:09,150 --> 00:01:11,740 They computed things. 16 00:01:11,750 --> 00:01:19,900 So for example let's say I wanted to find out how to get to Danielle's house using Google maps. 17 00:01:19,920 --> 00:01:27,720 Well one option is I could pull out a map and with a ruler measure each of the routes that I could take 18 00:01:27,720 --> 00:01:38,250 to Daniel's house and then do some math do some addition and find the shortest path or I could ask a 19 00:01:38,250 --> 00:01:39,730 computer to do this. 20 00:01:39,780 --> 00:01:46,620 Imagine we had ten different routes instead of me measuring each route one by one I could simply program 21 00:01:46,830 --> 00:01:52,450 and ask a computer Hey can you tell me how to get to Daniel's house. 22 00:01:52,860 --> 00:01:58,830 And through programming we tell the computer can you calculate really quickly these 10 routes and find 23 00:01:58,830 --> 00:02:00,640 the shortest one. 24 00:02:00,690 --> 00:02:08,260 That's what programming is instead of me taking 10 minutes to figure out how to get to Daniel's house 25 00:02:08,290 --> 00:02:14,900 and the fastest route I just click a button and the computer tells me what to do. 26 00:02:14,920 --> 00:02:20,920 Obviously we have to program it and give it instructions but once we give it a set of instructions it 27 00:02:21,640 --> 00:02:28,780 just like that finds the solution for us it computes solutions and you know why computers became so 28 00:02:28,780 --> 00:02:36,550 popular because computers saved companies lots of money instead of hiring lots of workers. 29 00:02:36,550 --> 00:02:41,270 Well let's just buy a computer that does the task of one hundred workers. 30 00:02:41,290 --> 00:02:42,300 And there you go. 31 00:02:42,310 --> 00:02:48,160 We saved a lot of money and computers don't complain they'll work 24/7 for you. 32 00:02:48,160 --> 00:02:52,530 Now these machines or computers are really good things that we can describe. 33 00:02:52,540 --> 00:02:53,100 Right. 34 00:02:53,230 --> 00:03:00,620 That we can write in code if else blocks to do something by the way if you're new to programming don't 35 00:03:00,620 --> 00:03:00,850 worry. 36 00:03:00,890 --> 00:03:06,600 We're actually going to start you off from scratch as well with Python so don't worry. 37 00:03:06,650 --> 00:03:09,290 This part is just theoretical. 38 00:03:09,350 --> 00:03:14,430 So for example let's say we talk about roots. 39 00:03:14,510 --> 00:03:21,380 I can say to the computer hey if Route 1 is shorter than route 2 and if Route 1 a shorter than Route 40 00:03:21,380 --> 00:03:29,870 3 and if Route 1 is shorter then Route 4 and also Route 5 well then pick Route 1 as the best option 41 00:03:30,410 --> 00:03:36,890 I can describe that through programming to computers and computers are really good at working on tasks 42 00:03:36,920 --> 00:03:44,860 that have these defined rules all the way up to a game of chess when you play a game of chess against 43 00:03:44,860 --> 00:03:45,800 a computer. 44 00:03:45,800 --> 00:03:53,750 We could technically have these A if else if then blocks lots of them to see how to move each piece 45 00:03:53,810 --> 00:03:56,000 and what the pros and cons are. 46 00:03:56,120 --> 00:04:00,880 And because computers are fast we can just do a ton of these calculations. 47 00:04:01,130 --> 00:04:04,260 But then there's a problem. 48 00:04:04,420 --> 00:04:11,830 What if instead of just trying to get to Daniel's house we have to ask is this person angry let's say 49 00:04:11,830 --> 00:04:18,490 somebody left a review on Amazon or we're building a product that detects human emotion. 50 00:04:18,490 --> 00:04:23,300 How can I describe to a computer what angry means can you do that. 51 00:04:23,340 --> 00:04:25,250 If an if else block. 52 00:04:25,540 --> 00:04:27,600 What about this. 53 00:04:27,700 --> 00:04:28,920 What is a cat. 54 00:04:28,930 --> 00:04:32,050 How do we tell a computer what a cat is. 55 00:04:32,110 --> 00:04:39,550 Let's say I ask you to program a computer to detect if this picture is a cat or not. 56 00:04:39,620 --> 00:04:41,260 Can you program that. 57 00:04:41,450 --> 00:04:44,450 I mean sure you can technically say a yes. 58 00:04:44,450 --> 00:04:46,400 This cat has fur. 59 00:04:46,400 --> 00:04:48,620 This cat has whiskers. 60 00:04:48,710 --> 00:04:50,630 This cat meows. 61 00:04:50,630 --> 00:04:57,590 But then the computers comes back to you and asks what is a meow or what are whiskers. 62 00:04:57,590 --> 00:05:00,140 What is Hair. 63 00:05:00,190 --> 00:05:06,320 You see the harder things become to describe the harder it is for us to tell machines what to do. 64 00:05:07,770 --> 00:05:11,690 So we hire humans to do these things that are harder for us. 65 00:05:11,730 --> 00:05:17,460 So we let machines take care of the easier part of which is you know things that we can describe and 66 00:05:17,550 --> 00:05:22,350 the harder things that are hard to just give instructions to. 67 00:05:22,620 --> 00:05:27,450 Well we just let humans do it like being a salesperson or being an artist. 68 00:05:27,450 --> 00:05:28,830 Computers aren't good. 69 00:05:28,860 --> 00:05:30,930 So we hire humans. 70 00:05:30,930 --> 00:05:36,330 But then there's this new idea of machine learning and you've definitely heard of it because it is a 71 00:05:36,330 --> 00:05:43,570 big buzzword right now in our industry because machine learning has a lot of applications. 72 00:05:43,620 --> 00:05:51,360 We could have self-driving cars robots vision processing language processing recommendation engines 73 00:05:51,450 --> 00:05:54,750 translation services stock price predictions. 74 00:05:54,750 --> 00:06:01,330 There's so many applications and this is all because of machine learning you see things that computers 75 00:06:01,330 --> 00:06:07,530 couldn't do before and only humans can can now be done by computers with machine learning. 76 00:06:07,600 --> 00:06:11,050 Well kind of sometimes it gets it right sometimes it doesn't. 77 00:06:11,050 --> 00:06:14,230 And we'll get to that but the idea is this. 78 00:06:14,400 --> 00:06:20,850 The goal of machine learning is to make machines act more and more like humans because the smarter they 79 00:06:20,850 --> 00:06:25,440 get the more they help us humans accomplish our goals. 80 00:06:25,470 --> 00:06:30,480 Now in this section we're going to go over some theory to get us familiar with this topic to understand 81 00:06:30,480 --> 00:06:34,230 this topic a little bit more but don't get worried. 82 00:06:34,230 --> 00:06:35,490 Don't get intimidated. 83 00:06:35,520 --> 00:06:38,810 We're going to try and simplify things and have fun along the way. 84 00:06:39,000 --> 00:06:44,550 And in later sections we're actually going to code build our own machine learning models and do that 85 00:06:44,550 --> 00:06:45,360 exciting stuff. 86 00:06:45,390 --> 00:06:48,120 But we have to learn the foundations first. 87 00:06:48,120 --> 00:06:51,580 So let's take a break and I'll see you in the next video by.