Cluster Analysis: Concept, Analysis of Tables, Steps, Interpretation and Conclusion

This article explains the concept of Cluster Analysis, analysis of tables, steps involved in the process, interpretation of results and conclusion. It also provides references for further reading. The article helps in understanding the technique used to classify objects or cases.

CLUSTER ANALYSIS 1
CLUSTER ANALYSIS

CLUSTER ANALYSIS 2
Table of Contents
1.Concept:.............................................................................................................................................3
2. Analysis of the table:.........................................................................................................................3
In the table 1:-.......................................................................................................................................3
In the table 2:-.......................................................................................................................................4
3. Steps of Cluster Analisis:...................................................................................................................5
4. Interpretation:....................................................................................................................................5
4. Conclusion.........................................................................................................................................6
References.............................................................................................................................................7

CLUSTER ANALYSIS 3
1. Concept:
It is a technique which is used to classify object or cases. It is basically analysis the case of
relatively homogeneous groups (Bamman et al., 2017). Every object is similar to each other
in each cluster but the other objects are dissimilar in another cluster. Always we will worry
about clustering procedures. Cluster analysis and discriminate analysis both are separate
(Silverman 2018). In this case, objects are including developing the classification rule.
2. Analysis of the table:
In table 1:
 The value singularity is 0.28 for which the value orientation was the only positive
influence, the quality at any price 0.64 segment which had the only negative centred
for the value orientation and the value of reputation or recreation is 0.15.
 The value singularity is negative 0.32 for which the quality orientation was the only
negative influence, the quality at any price 1.01 segments which had the only negative
centred for the quality orientation and the value of reputation or recreation is negative
0.33.
 The value of singularity is negative 0.62 for which the recreational was the only
negative influence, the quality at any price 0.12 segment which had the only negative
centred for the recreational and the value of reputation or recreation is 0.46.
 The value of singularity is negative 0.64 for which the reputation recreation was the
not only negative influence, the quality at any price negative 0.12 segment which had
the negative centred for the reputation recreation and the value of reputation or
recreation is negative 0.62.
 Now we can see that out of the three segments had a positive cluster in only one
factor. This is 0.28.
 After that, we can see that quality with convincing, recreational and reputation are as
(-0.32), (-0.62), and (-0.64). All the values of singularity are negative.
 This suggests that a segment of shoppers which is the positive motivation by the only
issue related to the value. Its a little attention paid to the other consideration.

CLUSTER ANALYSIS 4
In table 2:
 In the past years, which people spend in online divided into some stages? Among
them, less than 20,000 peoples value of singularity is 41, their quality at any price is
34, and reputation or recreation is 35. In between 20,000-40,000 people their value of
singularity is 40, quality at any price is 35 and reputation is 31 and lastly, 40,000 and
above peoples value of singularity is 19, quality at any price is 31 and recreation is 34.
 Those people who are purchase online in the past year are separate also some stages
among them less than 4 people value of singularity is 45, quality at any price is 36 and
reputation is 41. In between 4 to 6, the value singularity is 33, quality at any price is
34 and recreation is 31. And lastly, 7 and above, the value of singularity is 22, quality
at any price is 29 and reputation is 29.
 The growth of like hood of the online purchaser in next year, separate in three stages.
In the low stage the value singularity is 7, quality at any price is 8 and recreation is 6.
In medium value, singularity is 61, quality at any price is 40 and reputation is 47.
Lastly, in the high stage the value singularity is 31, quality at any price 52 and
reputation is 47.
 The people who have experience in internet. Less than 1year, value singularity is 16,
quality at any price is 140, and reputation is 8. Between 2-4 year, value singularity is
42, quantity at any price is also 33, and reputation is 29. And lastly 5 or more year
people, value singularity are 42, quality at any price 57, and reputation 63.
 The people on age basis. Less than 30 years, value singularity is 38, quality at any
price is 53, and reputation is 49. Between "30-39", value singularity is 36, quantity at
any price is also 32, and reputation is 37. And lastly 40 years and above people, value
singularity is 26, quality at any price 15, and reputation 14.
 The people who educated. High school or lower people, value singularity are 4,
quality at any price is 3, and reputation is 5. Vocational school and two year college,
value singularity is 61, quantity at any price is also 54, and reputation is 48. And
lastly four year college and above people, value singularity is 34, quality at any price
43, and reputation 47.
 The people who is in occupation. Student or house wife, value singularity is 15,
quality at any price is 12, and reputation is 19. Clerical/ factory/customer, value
singularity is 34, quantity at any price is also 34, and reputation is 44. And lastly

CLUSTER ANALYSIS 5
management/ professional and self employed people, value singularity is 52, quality
at any price 54, and reputation is 37.
 The people whose income is annual. Less than 3,500,000, value singularity is 17,
quality at any price is 12, and the reputation is 10. Between 3,500,000-6,500,000,
value singularity is 68, quantity at any price is also 73, and reputation is 73. And
lastly 6,500,000 people, value singularity are 16, quality at any price 14, and
reputation 16.
3. Steps of Cluster Analysis:
First of all, we need to formulate the problem. Then measure a distance among the problems.
The distance are Euclidean distance (distance(x,y) = {∑i (xi - yi)2 }½), squared Euclidean
distance (distance(x,y) = ∑i (xi - yi)2), city-block distance (distance(x,y) =∑ i |xi - yi|),
chebychev distance (distance(x,y) = Maximum|xi - yi|), power distance (distance(x,y) = (∑i |xi
- yi|p)1/r), Percent disagreement (distance(x,y) = (Number of xi ≠ yi)/ i), then select a procedure
in which way the cluster is happen, after that Decide on the Numbering of cluster, then
interpret and profile the cluster and lastly assess the validity of clustering.
4. Interpretation:
Centroids mean the average value of the objects which is come from the cluster for each of
the variable (Ahlqvist et al., 2018). Through the discriminate analysis, it may be obtained that
information clustering does not print.
4. Conclusion:
After analysis the table it is seen that there are many types of people who are using the
internet. Some of who are not using internet. Other persons whose annual income is high and
some of them those are mostly spent time on the internet. This analysis is based on
classification of this people in groups.

CLUSTER ANALYSIS 6
References:
Ahlqvist, E., Storm, P., Käräjämäki, A., Martinell, M., Dorkhan, M., Carlsson, A., ... &
Wessman, Y. (2018). Novel subgroups of adult-onset diabetes and their association
with outcomes: a data-driven cluster analysis of six variables. The Lancet Diabetes &
Endocrinology, 6(5), 361-369.
Bamman, M. M., Petrella, J. K., Kim, J. S., Mayhew, D. L., & Cross, J. M. (2017). Cluster
analysis tests the importance of myogenic gene expression during myofiber
hypertrophy in humans. Journal of Applied Physiology.
Morton, C., Anable, J., & Nelson, J. D. (2017). Consumer structure in the emerging market
for electric vehicles: Identifying market segments using cluster analysis. International
Journal of Sustainable Transportation, 11(6), 443-459.
Rennard, S. I., Locantore, N., Delafont, B., Tal-Singer, R., Silverman, E. K., Vestbo, J., ... &
Coxson, H. (2015). Identification of five chronic obstructive pulmonary disease
subgroups with different prognoses in the ECLIPSE cohort using cluster
analysis. Annals of the American Thoracic Society, 12(3), 303-312.
Silverman, B. W. (2018). Density estimation for statistics and data analysis. Routledge.
Ye, L., Pien, G. W., Ratcliffe, S. J., Björnsdottir, E., Arnardottir, E. S., Pack, A. I., ... &
Gislason, T. (2014). The different clinical faces of obstructive sleep apnoea: a cluster
analysis. European Respiratory Journal, erj00323-2014.