0 1 00:00:01,140 --> 00:00:07,800 This is start this session with a very important question, what is classification? Here, 1 2 00:00:07,810 --> 00:00:14,130 according to dictionary classification is an action or a process of classifying something according 2 3 00:00:14,130 --> 00:00:17,680 to shared qualities or characteristics. 3 4 00:00:18,090 --> 00:00:19,710 But why is it important? 4 5 00:00:20,490 --> 00:00:25,740 The first thing that we all learn in this school was classifying in a math class, 5 6 00:00:25,740 --> 00:00:29,970 We learned how to classify different shapes into a different groups. 6 7 00:00:30,390 --> 00:00:36,830 If a shape has four right angles and fout straight lines, then we can call it square. 7 8 00:00:37,500 --> 00:00:42,260 Basically, we are learning how to classify different things. In science 8 9 00:00:42,270 --> 00:00:47,160 We also learn how to classify different plants, animals, and we gave them a name. 9 10 00:00:47,820 --> 00:00:55,890 This is also very important science for machine learning, as well as. A manager who wants to decide 10 11 00:00:55,890 --> 00:00:58,160 about some particular situation. 11 12 00:00:58,680 --> 00:01:03,840 He will compared the existing situation with one of the sets that he has in his mind. 12 13 00:01:03,990 --> 00:01:05,840 Finally, he will make a decision. 13 14 00:01:06,810 --> 00:01:08,730 Let's take a look at this example. 14 15 00:01:09,640 --> 00:01:14,010 Each of these orange circles are representing a patient. 15 16 00:01:14,490 --> 00:01:23,880 Let's say they are the result of tests of some patients and a doctor wants to decide if they are diabetic 16 17 00:01:23,880 --> 00:01:24,480 or not. 17 18 00:01:25,080 --> 00:01:32,700 What the doctor is eventually doing is asking them some questions, maybe conducting some tests and gathering 18 19 00:01:32,700 --> 00:01:33,570 some data. 19 20 00:01:33,990 --> 00:01:42,180 And then he or she will compare this set of data to one of the sets and groups that he has in his mind. 20 21 00:01:42,630 --> 00:01:44,580 And then he/she (the doctor) can make some decision. 21 22 00:01:44,580 --> 00:01:51,930 He can say, for example, this patient is a diabetic person and this person is a very healthy one 22 23 00:01:52,050 --> 00:01:54,050 and he doesn't have any problem. 23 24 00:01:55,110 --> 00:02:03,300 But sometimes there are situations that these two patients, they have all similar symptoms, but only 24 25 00:02:03,300 --> 00:02:10,620 different is in one sign or one symptoms. A doctor who can compare them and who can make a better decision, 25 26 00:02:10,620 --> 00:02:12,450 Is the experienced, doctor! 26 27 00:02:12,740 --> 00:02:19,360 That's why they are usually looking for finding a very expert doctor in critical situations because 27 28 00:02:19,360 --> 00:02:20,850 they have more experience. 28 29 00:02:20,850 --> 00:02:22,430 But what are these experiences? 29 30 00:02:22,710 --> 00:02:29,580 It means they saw more samples, had more patients and they have a better data set on their mind. 30 31 00:02:29,700 --> 00:02:36,960 Suppose these are the patients with diabetic or high blood pressure and these are the healthy patients. 31 32 00:02:37,440 --> 00:02:44,610 Doctor can separate these two sets of their patients and later on, maybe they can group them and prescribe 32 33 00:02:44,610 --> 00:02:47,280 a specific treatment plan for this set. 33 34 00:02:47,280 --> 00:02:55,050 And for another set here, prescribe another medicine or treatment plans. The doctor is classifying 34 35 00:02:55,050 --> 00:02:56,970 patients into different groups. 35 36 00:02:57,060 --> 00:02:59,430 And they are very healthy patients. 36 37 00:03:00,240 --> 00:03:03,740 But let's take a look at mathematics here. 37 38 00:03:03,810 --> 00:03:12,980 I call this one X, and here Y we can see that this is a line with a mathematical function (equation). 38 39 00:03:13,500 --> 00:03:18,610 This is ax plus by plus C equals to zero. 39 40 00:03:18,850 --> 00:03:30,240 We can say that here for these patients, if ax plus by plus c is greater than zero, then these patients 40 41 00:03:30,240 --> 00:03:32,460 are diagnosed with a diabetic. 41 42 00:03:32,790 --> 00:03:39,210 And for these patients, if ax plus by plus c 42 43 00:03:39,570 --> 00:03:46,470 Suppose this is disorder information for this patient is less than zero, then this patient is healthy. 43 44 00:03:46,700 --> 00:03:51,540 We already drove a mathematical model for our patients 44 45 00:03:52,740 --> 00:03:54,630 What is the doctor doing? 45 46 00:03:54,630 --> 00:03:57,710 Actually, first thing is classification. 46 47 00:03:58,200 --> 00:04:02,640 Next thing that he or she does is recognition and identification. 47 48 00:04:03,870 --> 00:04:09,410 They will do the recognition and identification based on the results of the test. 48 49 00:04:09,420 --> 00:04:12,150 And finally, they will do some assessment. 49 50 00:04:12,750 --> 00:04:16,530 And they also have forecasting and a prediction. 50 51 00:04:17,010 --> 00:04:23,700 By looking at the test result of this patient, they can easily say that based on these itemd in your 51 52 00:04:23,700 --> 00:04:24,780 test result, 52 53 00:04:24,780 --> 00:04:31,680 If you continue eating Sweet things(foods), then in one month or in two months, you will be here and you 53 54 00:04:31,680 --> 00:04:32,700 will be diabetic. 54 55 00:04:33,300 --> 00:04:34,680 So you need to be careful. 55 56 00:04:34,680 --> 00:04:40,110 You need to reduce eating Sweets and some other diets. 56 57 00:04:40,560 --> 00:04:45,030 But for this patient, they can say, OK, your situation is pretty stable. 57 58 00:04:45,240 --> 00:04:46,500 You don't have any problem. 58 59 00:04:46,500 --> 00:04:50,250 You can eat as much as sweet as you want and 59 60 00:04:50,250 --> 00:04:56,760 There won't be any problem for you, at least in the next few months. Here, Doctors can predict 60 61 00:04:56,760 --> 00:04:59,850 and they can forecast based on the data that.