CLUSTERING USING WHOLESALE CUSTOMERS DATASET
Lavanya M, Jenifer PK & Reyne Beatrice
19th March 2016
Introduction:
Wholesale Customer dataset refers to clients of a wholesale distributor. It includes the annual spending in monetary units (m.u.) on diverse product categories. Number of Instances: 440 Number of Attributes: 8 Data Set Characteristics: Multivariate Attribute Characteristics: Integer
Objective:
The major aim is to perform clustering analysis using algorithms like hClust,PAM,kMeans,mclust and provide inferences accordingly.
Attribute Information:
- FRESH: annual spending (m.u.) on fresh products (Continuous)
- MILK: annual spending (m.u.) on milk products (Continuous)
- GROCERY: annual spending (m.u.)on grocery products (Continuous)
- FROZEN: annual spending (m.u.)on frozen products (Continuous)
- DETERGENTS_PAPER: annual spending (m.u.) on detergents and paper products (Continuous)
- DELICATESSEN: annual spending (m.u.)on and delicatessen products (Continuous)
- CHANNEL: customersale Channel - Horeca (Hotel/Restaurant/Cafe) or Retail channel (Nominal)
- REGION: customersale Region - Lisnon, Oporto or Other (Nominal)
Descriptive Statistics:
(Minimum, Maximum, Mean, Std. Deviation) FRESH ( 3, 112151, 12000.30, 12647.329) MILK (55, 73498, 5796.27, 7380.377) GROCERY (3, 92780, 7951.28, 9503.163) FROZEN (25, 60869, 3071.93, 4854.673) DETERGENTS_PAPER (3, 40827, 2881.49, 4767.854) DELICATESSEN (3, 47943, 1524.87, 2820.106)
REGION Frequency Lisbon 77 Oporto 47 Other Region 316 Total 440
CHANNEL Frequency Horeca 298 Retail 142 Total 440
Relevant Papers:
Cardoso, Margarida G.M.S. (2013). Logical discriminant models “Chapter 8 in Quantitative Modeling in Marketing and Management Edited by Luiz Moutinho and Kun-Huang Huarng. World Scientific. p. 223-253. ISBN 978-9814407717
Jean-Patrick Baudry, Margarida Cardoso, Gilles Celeux, Maria Josa Amorim, Ana Sousa Ferreira (2012). Enhancing the selection of a model-based clustering with external qualitative variables. RESEARCH REPORT 8124, October 2012, Project-Team SELECT. INRIA Saclay - AZle-de-France, Projet select, University of Paris-Sud 11 ###Loading the data and finding the summary:
## [1] "C:/Users/admin/Documents"## Channel Region Fresh Milk
## Min. :1.000 Min. :1.000 Min. : 3 Min. : 55
## 1st Qu.:1.000 1st Qu.:2.000 1st Qu.: 3128 1st Qu.: 1533
## Median :1.000 Median :3.000 Median : 8504 Median : 3627
## Mean :1.323 Mean :2.543 Mean : 12000 Mean : 5796
## 3rd Qu.:2.000 3rd Qu.:3.000 3rd Qu.: 16934 3rd Qu.: 7190
## Max. :2.000 Max. :3.000 Max. :112151 Max. :73498
## Grocery Frozen Detergents_Paper Delicassen
## Min. : 3 Min. : 25.0 Min. : 3.0 Min. : 3.0
## 1st Qu.: 2153 1st Qu.: 742.2 1st Qu.: 256.8 1st Qu.: 408.2
## Median : 4756 Median : 1526.0 Median : 816.5 Median : 965.5
## Mean : 7951 Mean : 3071.9 Mean : 2881.5 Mean : 1524.9
## 3rd Qu.:10656 3rd Qu.: 3554.2 3rd Qu.: 3922.0 3rd Qu.: 1820.2
## Max. :92780 Max. :60869.0 Max. :40827.0 Max. :47943.0## 'data.frame': 440 obs. of 8 variables:
## $ Channel : int 2 2 2 1 2 2 2 2 1 2 ...
## $ Region : int 3 3 3 3 3 3 3 3 3 3 ...
## $ Fresh : int 12669 7057 6353 13265 22615 9413 12126 7579 5963 6006 ...
## $ Milk : int 9656 9810 8808 1196 5410 8259 3199 4956 3648 11093 ...
## $ Grocery : int 7561 9568 7684 4221 7198 5126 6975 9426 6192 18881 ...
## $ Frozen : int 214 1762 2405 6404 3915 666 480 1669 425 1159 ...
## $ Detergents_Paper: int 2674 3293 3516 507 1777 1795 3140 3321 1716 7425 ...
## $ Delicassen : int 1338 1776 7844 1788 5185 1451 545 2566 750 2098 ...Algorithm1:hClust
## 'dendrogram' with 2 branches and 440 members total, at height 128968.4
Algorithm2:kMeans
## Channel Region Fresh Milk
## Min. :1.000 Min. :1.000 Min. : 3 Min. : 55
## 1st Qu.:1.000 1st Qu.:2.000 1st Qu.: 3128 1st Qu.: 1533
## Median :1.000 Median :3.000 Median : 8504 Median : 3627
## Mean :1.323 Mean :2.543 Mean : 12000 Mean : 5796
## 3rd Qu.:2.000 3rd Qu.:3.000 3rd Qu.: 16934 3rd Qu.: 7190
## Max. :2.000 Max. :3.000 Max. :112151 Max. :73498
## Grocery Frozen Detergents_Paper Delicassen
## Min. : 3 Min. : 25.0 Min. : 3.0 Min. : 3.0
## 1st Qu.: 2153 1st Qu.: 742.2 1st Qu.: 256.8 1st Qu.: 408.2
## Median : 4756 Median : 1526.0 Median : 816.5 Median : 965.5
## Mean : 7951 Mean : 3071.9 Mean : 2881.5 Mean : 1524.9
## 3rd Qu.:10656 3rd Qu.: 3554.2 3rd Qu.: 3922.0 3rd Qu.: 1820.2
## Max. :92780 Max. :60869.0 Max. :40827.0 Max. :47943.0There’s obviously a big difference for the top customers in each category here for example, Fresh goes from a min of 3 to a max of 112,151. Normalizing / scaling the data won’t necessarily remove those outliers.We could also remove those customers completely. From a business perspective, you don’t really need a clustering algorithm to identify what your top customers are buying. You usually need clustering and segmentation for your middle 50%.We try removing the top 5 customers from each category.
## [1] 19## Channel Region Fresh Milk Grocery Frozen Detergents_Paper Delicassen
## 182 1 3 112151 29627 18148 16745 4948 8550
## 126 1 3 76237 3473 7102 16538 778 918
## 285 1 3 68951 4411 12609 8692 751 2406
## 40 1 3 56159 555 902 10002 212 2916
## 259 1 1 56083 4563 2124 6422 730 3321
## 87 2 3 22925 73498 32114 987 20070 903
## 48 2 3 44466 54259 55571 7782 24171 6465
## 86 2 3 16117 46197 92780 1026 40827 2944
## 184 1 3 36847 43950 20170 36534 239 47943
## 62 2 3 35942 38369 59598 3254 26701 2017
## 334 2 2 8565 4980 67298 131 38102 1215
## 66 2 3 85 20959 45828 36 24231 1423
## 326 1 2 32717 16784 13626 60869 1272 5609
## 94 1 3 11314 3090 2062 35009 71 2698
## 197 1 1 30624 7209 4897 18711 763 2876
## 104 1 3 56082 3504 8906 18028 1480 2498
## 24 2 3 26373 36423 22019 5154 4337 16523
## 72 1 3 18291 1266 21042 5373 4173 14472
## 88 1 3 43265 5025 8117 6312 1579 14351We need to drop the Channel and Region variables. These are two ID fields and are not useful in clustering.
## Fresh Milk Grocery Frozen Detergents_Paper Delicassen
## 1 4189.747 7645.639 11015.277 1335.145 4750.4819 1387.1205
## 2 16470.870 3026.491 4264.741 3217.306 996.5556 1319.7593
## 3 33120.163 4896.977 5579.860 3823.372 945.4651 1620.1860
## 4 5830.214 15295.048 23449.167 1936.452 10361.6429 1912.7381
## 5 5043.434 2329.683 2786.138 2689.814 652.8276 849.8414##
## 1 2 3 4 5
## 83 108 43 42 145Interpretation of the results:
Cluster 1 looks to be a heavy Grocery and above average Detergents_Paper but low Fresh foods. Cluster 3 is dominant in the Fresh category. Cluster 5 might be either the “junk drawer” catch-all cluster or it might represent the small customers.
Plotting Cluster Solutions:
It is always a good idea to look at the cluster results. 
This plot doesn’t show a very strong elbow.Somewhere around K = 5 we start losing dramatic gains. So we are satisfied with 5 clusters.
Algorithm3:PAM
Algortithm4:EMCluster
Algorithm5:mclust
## 'Mclust' model object:
## best model: ellipsoidal, equal volume, shape and orientation (EEE) with 4 components
