1 00:00:00,600 --> 00:00:03,833 And here is our hierarchical clustering results obtained in R. 2 00:00:04,200 --> 00:00:07,500 So let's have a closer look at these clusters and understand what they are. 3 00:00:08,133 --> 00:00:11,400 Our first cluster is cluster one which is the blue cluster right here. 4 00:00:11,833 --> 00:00:15,900 And it contains the customers that have high income and high spinning score. 5 00:00:16,466 --> 00:00:19,600 So that's an interesting cluster for the mall because customers 6 00:00:19,600 --> 00:00:22,866 in this cluster have a high income and spend a lot of money in the mall. 7 00:00:23,100 --> 00:00:26,666 So this cluster would be a good target of the mall marketing campaigns. 8 00:00:26,933 --> 00:00:29,933 And so we can call this cluster the target one. 9 00:00:30,066 --> 00:00:30,433 Okay. 10 00:00:30,433 --> 00:00:33,433 Now cluster two close to two contains the customers 11 00:00:33,433 --> 00:00:36,433 that have high income and low spending score. 12 00:00:36,733 --> 00:00:39,800 And so we can call these customers to careful customers. 13 00:00:40,533 --> 00:00:42,100 Now cluster three 14 00:00:42,100 --> 00:00:45,900 customers in this cluster have average income and average spending score. 15 00:00:46,266 --> 00:00:49,100 So we call this cluster the standard cluster. 16 00:00:49,100 --> 00:00:51,800 Now Christopher cluster four has the customers 17 00:00:51,800 --> 00:00:54,800 that have low income and low spending score. 18 00:00:55,066 --> 00:00:59,633 So that's basically sensible customers who pay attention to the money 19 00:00:59,633 --> 00:01:02,633 they spend by paying attention to the money they earn. 20 00:01:02,933 --> 00:01:05,666 And eventually cluster five contains 21 00:01:05,666 --> 00:01:09,000 the customers that have low income but high spinning score. 22 00:01:09,000 --> 00:01:13,766 So these are customers who are rather careless, especially that 23 00:01:13,766 --> 00:01:18,833 199 customer here who earns low income but spends a lot of money. 24 00:01:19,266 --> 00:01:21,800 So this one should particularly be careful. 25 00:01:21,800 --> 00:01:25,200 And actually, this cluster five here could be a very interesting cluster 26 00:01:25,200 --> 00:01:28,800 for the company if the latter has a social responsibility, 27 00:01:28,800 --> 00:01:31,100 as it is the case for more and more companies today. 28 00:01:32,200 --> 00:01:32,666 All right. 29 00:01:32,666 --> 00:01:35,633 So we are done with hierarchical clustering and R 30 00:01:35,633 --> 00:01:39,300 if you want to reuse this code for your business problem it's very easy. 31 00:01:39,533 --> 00:01:42,366 You just have to change the name of the data set here. 32 00:01:42,366 --> 00:01:45,666 Then change the indexes of your columns of interest just below. 33 00:01:46,066 --> 00:01:48,900 However, if you do clustering in more than two dimensions, 34 00:01:48,900 --> 00:01:50,400 don't execute the last section. 35 00:01:50,400 --> 00:01:53,400 Visualizing the clusters because it's only for two dimensions. 36 00:01:53,766 --> 00:01:56,333 But don't delete this section and leave it as comments, 37 00:01:56,333 --> 00:01:59,533 because later in this course we will learn a super powerful technique called 38 00:01:59,533 --> 00:02:02,966 dimensionality reduction that will reduce the dimensions 39 00:02:02,966 --> 00:02:07,233 of your data to perhaps two dimensions, so that you can use this final section, 40 00:02:07,233 --> 00:02:10,666 visualizing the clusters to see your clusters in two dimensions. 41 00:02:11,666 --> 00:02:12,033 All right. 42 00:02:12,033 --> 00:02:14,666 So that's the end of hierarchical clustering in R. 43 00:02:14,666 --> 00:02:17,600 And that's also the end of hierarchical clustering. 44 00:02:17,600 --> 00:02:21,300 And in the next section we will recap everything we've learned in clustering. 45 00:02:22,500 --> 00:02:23,966 Thank you for watching this tutorial. 46 00:02:23,966 --> 00:02:27,933 I hope you are now more comfortable with doing clustering in R or Python. 47 00:02:28,366 --> 00:02:31,166 I look forward to seeing you in the next tutorials 48 00:02:31,166 --> 00:02:32,866 and until then, happy clustering!