Beruflich Dokumente
Kultur Dokumente
In normal way we can say that cluster is a technique in which we make the cluster on the basic of similarity , there is (homogenous behavior) within the a cluster but heterogeneous behavior with other cluster.
The simplest way to explain the technique is to understand that it is simply measure the distance between object(respondent) on the basic of multiple variable(question) and looks for similarity as a function of distance , i.E. The shorter the distance between two object , the more similar they are.
For ex.
Suppose we divide the class room on the basic of religion( Hindu , Muslim , skih , christen)so those who are Hindu at one cluster and those who are Muslim at other cluster and so on. There is similarity with in the cluster and different to other cluster.
Cluster analysis has widespread applicability in all branches of social science and management. In management science, it most valuable contribution is in the area of marketing , basically market segmentation.
1. 2. 3. 4.
market segmentation industries/sector segmentation career planning and training analysis segmenting financial sector/instruments
1) Market segmentation
Market segmentation is the process of splitting customer /potential customers , within a market into different group/segment with the help of cluster where customer has same /similar requirement satisfied by distinct marketing mix. Then after cluster we divide the customer on the basics of different factor like age, income , education ,culture). Then we make effective strategy for potential as well as present customer. For ex. After cluster our product is more prefer by adult so we make or change the market strategy is like that its satisfied present and attract potential customer.
CONTENT
DETAILS
2)sector/industries segmentation
The researcher could also go about grouping the product or sectors(agriculture and sugar industries) into one group have some common traits.
This make easier for both the organization and policy make while planning or evaluating the performance of the group.
3) Segmenting market
Cities or region with some common traits like population mix ,culture ,climatic could be cluster together For ex: if one city Kerala and another Andhra Pradesh has same climate condition has in one cluster then the organization is able to plan and execute a similar business planning for both the area.
In the area of (HR) the technique can be used to group people into cluster on the basic of their education , Qualification, experience , aptitude.. This help hr to effective manage employee and provide training a/c to requirement.
For obtaining a cluster solution under this method distance between two or more variable can be calculated
DA,B=
(XB1-XA1)2 +(XB2-XA2)2
In this variable a b and c were placed on a 10-point parameter scale (1=very unimportant and 10 = important). then the value selected by the person a b and c is.
Person
nutrition
energy
D a,b= (5-1)2+(2-2)2
=4
D b,c= (6-5)2+(2-2)2
=1
Then the distance between a and b will be 5 and b and c will be 1 so Cluster technique used to group of similar into one cluster So lower the distance , greater the similarity . B and c in one cluster.
A jeweler designer who wish to know the population of young teenage girls 13-19 preference toward jewellery. Assumption There will be five question in questioner and this is based on Five point likert scale ranging from 1= strongly agree to 5=strongly disagree.
1. 2.
Respondent x1 number
1 1
x2
3
x3
5
x4
4
x5
3
2
3 4 5 6
2
3 5 4 2
3
2 5 4 2
4
3 1 2 4
5
3 2 2 3
2
3 4 3 2
7
8 9 10
3
2 4 5
3
1 4 4
4
3 2 1
4
3 2 1
3
2 3 3
1.
D a,b= (xb1-xa1)2+(xb2-xa2)2
D 2,1 = (2-1)2+ (3-3)2+ (4-5)2+ (5-4)2+ (3-2)2
= 2+ 2+ 2+ 2+ 2 (1) (0) (1) (1) (1) =4
1 1 2 3
000
2
4.000 000
3
10.000 8.000 000
4
41.000 35.000 19.000 000
5
23.000 19.000 7.000 4.000 000
6
5.000 5.000 3.000 32.000 14.000 000
7
5.000 3.000 3.000 22.000 10.000 4.000 000
8
11.000 9.000 3.000 34.000 16.000 2.000 8.000
9
23.000 19.000 7.000 4.000 000 14.000 10.000
10
42.000 36.000 16.000 3.000 3.000 27.000 23.000
4
5 6 7
8
9 10
000
16.000
000
27.000
3.000 000
Take the least value that is 0(cluster1: 5,9) Then second least value is 2(cluster2: 6,8) Next value is 3 then Cluster 1:5,9,4,10 Cluster 2: 6, 8,3,2,7
It will dependent on two scale that is nominal and ordinal scale in which response may be denoted in binary number (0 ,1) .
For ex. Marital status , sex etc.
For example we take the 3 respondent on the basic of nominal and ordinal scale(0,1), to know the behavior toward lunch.
BREAKFAST OPTION
PERSON TOAST PARAN THA IDILI POHA DHOKLA PATTIES BAGELS JUICE MILK CHAPA TI
0 0 1
0 0 1
1 1 1
0 0 0
1 1 0
0 0 1
0 0 1
1 1 1
1 1 1
1 0 1
BREAKFAST GROUPING
RAVI-AMIT POSTIVE MATCH-(P) 4 (--) MATCH(N) MISMATCH-(M) 5 1 RAVI -ANKIT 4 1 5 AMIT-ANKIT 3 1 6
COEFFICIENT MEASURE
simple matching coefficient P ------------(P+M+N) jaccard coefficient p -------------(p + m)
CASE PAIR
RAVI-AMIT RAVI-ANKIT AMIT-ANKIT
VALUE
0.4 0.4 0.3
CLUSTER VARIABLE
ASSUMPTION (metric and non-metric)
Distance measures Clustering algorithm Is a hierarchical or non or combination of two Nonhierarchical method
Matching concepts
Hierarchical method
Combined cluster
Number of cluster
Types of method Single linkage Complete linkage Average linkage Wards methods Centriod method
3)Combination
4)two-step cluster
details It is based on minimum distance . The first two most similar case are put in first cluster and then the next closest pair join and then move to every stage . The two cluster show shortest distance as the shortest distance between two closest point. This is opposite of the single linkage method. Rather than minimum distance ,the cluster is based on maximum distance between two element. The cluster criterion here is the average distance from all the element in one cluster with the other entire cluster Here the distance between two cluster is the sum of square between the two cluster across all the cluster variable. in this case cluster variance is reduced to a minimum. Cluster Centroid are calculated as the mean value for the clustering variable.
4)Wards method
5) Centroid method
detail The method goes to one cluster seed to the next in the sequential manner . The first cluster seed is selected and all the case that lie in the states distance are included then one goes to the next seed and the next . this process is continued till all the case are clustered. Here several cluster seed are selected at one go , are parallel and then categorized into cluster ,whose distance is minimum in different cluster. This method allows for a realignment of cases . then allotting cases to the cluster based on the threshold distance. Those cluster have seem to belong to the other cluster , these cluster is moved to the other cluster for optimum solution .
3) Optimizing procedure
method
detail
It has the advantage of being compatible with both continuous and categories data.
The technique first determine the optimal number of cluster automatically by comparing the value across different clustering solution. This method can be used to validate the result obtain by other previous two method.
detail There are different schools of thought on which is betterhierarchal and non- hierarchal method . In practices researcher use them in combination. That Is ,one uses hierarchical method to establish how many /cluster would be ideal and then carries out a non hierarchical method with pre specified number of cluster. This output , then , is used to interpret the cluster solution.
First select the number of variable used and also select the parameter scale used to get the solution, A) the scale may be nominal scale in which answer will be in yes or no. Or in interval or ratio 5-rating likert scale.
A study conducted of 25 two wheeler owner in NCR region to assess there purchase intension to maruti 800.
Since the objective of cluster analysis is to classify object that are similar in composition , the second step is to select the statistical technique applicable for the selected level of measurement.
id
1 2 3
1a
5 3 1
1b
5 3 1
1c
3 5 1
1d
2 4 2
1e
3 4 1
1f
3 5 2
1g
4 4 1
1h
1 1 4
1i
1 1 4
2
1 0 0
3
2 2 2
4
4 2 1
5
2 1 3
6
1 2 1
7
3 3 3
8
3 2 1
9
1 1 1
4
5 6 7
5
2 2 3
5
2 2 3
4
4 1 2
2
5 2 1
3
4 1 1
4
5 1 1
3
4 1 1
2
2 5 5
2
2 5 4
1
0 1 0
2
2 2 3
4
4 4 2
2
3 2 1
1
2 1 2
3
3 3 4
3
2 1 1
1
1 1 3
8
9 10 11
1
4 1 2
1
5 1 2
1
3 4 1
2
3 4 2
1
3 3 1
2
3 4 1
1
4 4 1
4
1 2 5
4
1 2 5
0
1 0 1
2
2 2 2
1
4 1 4
3
2 2 2
2
1 2 1
3
3 3 3
2
3 3 1
1
1 1 1
12
13 14 15
5
3 5 3
4
3 5 2
3
2 2 5
4
1 2 5
3
1 2 5
2
1 3 5
2
1 1 4
2
5 1 2
2
4 1 1
0
0 1 0
1
3 2 2
2
2 3 1
3
1 2 3
2
2 1 2
3
4 3 3
3
1 2 3
3
3 1 2
id
16 17 18 19 20 21 22 23 24 25
1a 1b 1c 1d 1e 1f
4 2 2 4 4 2 2 4 4 2 5 1 3 5 4 2 1 4 5 3 2 5 2 3 2 1 5 2 3 2 2 5 2 3 1 2 5 2 2 2 3 5 2 3 3 1 5 2 3 1 1 4 1 2 2 1 5 3 3 1
11 g
1 5 1 2 1 1 4 4 4 1
1h
1 1 5 1 1 5 1 1 1 5
1i
1 1 4 1 2 5 1 2 1 4
2
1 0 1 1 0 1 0 1 0 1
3
2 3 2 2 2 2 2 2 1 2
4
3 2 3 3 3 4 2 4 4 4
5
2 2 2 2 3 1 3 3 3 2
6
1 2 1 1 2 1 2 2 1 1
7
3 3 3 3 3 3 3 3 3 3
8
3 2 2 3 3 1 2 3 2 1
9
1 1 1 1 1 1 3 1 1 1
Discuss the hierarchical , non-hierarchal, combination method s for obtaining a cluster analyst
Through this above method we combine the same nature at one cluster mean that after collected data and plotted them in the table then we make the cluster of similar nature through different technique.
stage 1 2 3
Next stage 9 4 9
4
5 6 7
6
3 9 1
11
8 24 9
.000
.000 1.000 1.500
0
0 0 0
2
0 0 6
12
20 7 13
8
9 10 11
17
7 16 15
22
18 20 17
2.000
2.000 4.000 4.000
0
3 0 0
0
1 0 8
11
12 16 18
12
13 14 15 16
6
1 12 5 12
7
23 19 10 16
4.000
4.667 5.000 5.000 6.000
4
7 0 0 14
9
0 0 0 10
20
19 16 21 17
stage
Coefficient
Next stage
17
18 19 20 21 22 23
12
2 1 3 2 1 1
14
15 4 6 5 12 2
6.250
6.667 7.000 7.857 8.500 11.800 40.667
16
0 13 5 18 19 22
0
11 0 12 15 17 21
22
21 22 24 23 23 24
24
59.222
23
20
Case 1 2 3 4 5
6 clusters 1 2 3 1 4
5 cluster 1 2 3 1 4
4 cluster 1 2 3 1 2
3 cluster 1 2 3 1 2
2 cluster 1 1 2 1 1
6
7 8 9 10 11 12
5
5 3 1 4 5 6
3
3 3 1 4 3 5
3
3 3 1 2 3 4
3
3 3 1 2 3 1
2
2 2 1 1 2 1
13
14 15 16
5
6 2 6
3
5 2 5
3
4 2 4
3
1 2 1
2
1 1 1
Case 17 18 19 20 21 22 23
6 clusters 2 5 6 6 5 2 1
5 cluster 2 3 5 5 3 2 1
4 cluster 2 3 4 4 3 2 1
3 cluster 2 3 1 1 3 2 1
2 cluster 1 2 1 1 2 1 1
24
25
1
5
1
3
1
3
1
3
1
2
question I thing in India we have been able to achieve technological slandered of high order I prefer to buy thinks in India I usually buy thing which provide value to money Convenience is more important than style. I dont like wasteful expenditure
50.579
23.468 164.22 96.749
.000
.000 .000 .000
question 1
cluster 2 3
I thing in India we have been able to achieve technological slandered of high order
I prefer to buy thinks in India I usually buy thing which provide value to money Convenience is more important than style. I dont like wasteful expenditure When it comes to safety I believe there is no compromise I am saver rather then the spender I like to try new and different things. I always want to be a part of changing world.
2.17
1.67 4.67 4.67 4.33 4.67 4.17 1.50 1.33
2.00
2.22 1.44 1.78 1.00 1.22 1.00 4.78 4.33
4.40
4.70 2.70 2.10 2.80 2.60 2.60 1.20 1.40
Cluster .1 (caution consumer) Cluster .2 (innovative consumer ) Cluster .3 (patriotic consumer) Valid Missing