Import customer file
customer = read.csv("V:/acedemics/MB BA/sem 3/Predictive Analytics/Customer.csv")
Head Customer
head(customer)
## ID Visit.Time Average.Expense Sex Age
## 1 1 3 5.7 0 10
## 2 2 5 14.5 0 27
## 3 3 16 33.5 0 32
## 4 4 5 15.9 0 30
## 5 5 16 24.9 0 23
## 6 6 3 12.0 0 15
normalize the customer data
scustomer <- scale(customer[,-1])
agglomratic clustering
hc = hclust(dist(customer, method='euclidean'), method = "ward.D2")
plot function to plot dendogram
plot(hc, hang = -0.5, cex = 0.6)
hcsingle = hclust(dist(customer, method='euclidean'), method = "single")
plot(hcsingle, hang = -0.5, cex = 0.5)
hcsingle = hclust(dist(customer, method='euclidean'))
plot(hcsingle, hang = -0.5, cex = 0.5)
heirarchial clustering cut tree example
d <- dist(customer, method='euclidean')
fit <- hclust(d, method='ward.D2')
plot(fit,hang = -0.5, cex = 0.5)
groups <- cutree(fit, k=5)
rect.hclust(fit,k=5,border='blue')
placing a rect around a particular cluster
plot(hc, hang=-0.01, cex=0.5)
rect.hclust(hc, k=4, which=3, border='blue')
library(cluster)
dv=diana(customer, metric='euclidean')
summary(dv)
## Merge:
## [,1] [,2]
## [1,] -39 -40
## [2,] -23 -26
## [3,] -2 -4
## [4,] -54 -58
## [5,] -10 -11
## [6,] -57 -60
## [7,] -48 -49
## [8,] -8 -12
## [9,] -30 -35
## [10,] 1 -43
## [11,] -42 -44
## [12,] -32 -34
## [13,] -17 -19
## [14,] -25 -28
## [15,] -7 -13
## [16,] 2 -27
## [17,] -18 -21
## [18,] -56 -59
## [19,] -36 -41
## [20,] -33 -37
## [21,] -14 -15
## [22,] -5 8
## [23,] -50 -53
## [24,] 4 -55
## [25,] -45 -46
## [26,] -6 5
## [27,] -51 18
## [28,] 16 -29
## [29,] 10 11
## [30,] -38 29
## [31,] 13 -24
## [32,] -1 26
## [33,] 9 19
## [34,] -16 17
## [35,] -3 15
## [36,] 7 24
## [37,] 23 -52
## [38,] 25 37
## [39,] -31 30
## [40,] 3 22
## [41,] -20 14
## [42,] 31 -22
## [43,] 34 28
## [44,] 12 20
## [45,] -47 36
## [46,] 32 21
## [47,] 45 27
## [48,] 35 -9
## [49,] 40 48
## [50,] 33 39
## [51,] 44 38
## [52,] 42 41
## [53,] 46 43
## [54,] 49 52
## [55,] 50 47
## [56,] 51 6
## [57,] 53 54
## [58,] 55 56
## [59,] 57 58
## Order of objects:
## [1] 1 6 10 11 14 15 16 18 21 23 26 27 29 2 4 5 8 12 3 7 13 9 17
## [24] 19 24 22 20 25 28 30 35 36 41 31 38 39 40 43 42 44 47 48 49 54 58 55
## [47] 51 56 59 32 34 33 37 45 46 50 53 52 57 60
## Height:
## [1] 10.881636 8.768694 4.109745 18.300000 7.790379 29.295904 11.819052
## [8] 6.324555 16.192900 3.104835 6.184658 9.282241 43.848489 3.867816
## [15] 15.942396 8.497647 4.983974 22.538855 11.916375 6.000000 22.538855
## [22] 30.832450 5.855766 10.678951 16.117072 26.474327 15.942396 5.861740
## [29] 72.078846 5.000000 11.808472 7.020684 23.499149 14.425325 9.529428
## [36] 2.256103 5.728001 9.486833 5.766281 32.790852 17.811513 4.586938
## [43] 12.122706 4.060788 8.560958 20.651634 9.046546 6.618912 53.331417
## [50] 5.838664 17.298844 7.376313 24.598577 8.616844 14.208800 8.558037
## [57] 12.806639 33.313811 4.267318
## Divisive coefficient:
## [1] 0.8968104
##
## 1770 dissimilarities, summarized :
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 2.256 18.406 27.409 28.336 36.969 72.079
## Metric : euclidean
## Number of objects : 60
##
## Available components:
## [1] "order" "height" "dc" "merge" "diss" "call" "data"