Your project work is ready, see attached document for a 12 paged report as instructed. Let me know if you need anything else. Thank you 😃

ABSTRACT
Clustering can be defined as the process of dividing data points or population into groups, that
is, data points in similar groups are very much alike to those in other groups. Clustering aims
to segregate groups with similar traits as well as assign clusters.
This report will highlight the various ways of carrying out clustering. There is no doubt that
clustering implementation is quite easy, it is important to first handle some very essential
aspects. Such as treating outliers in data and ensuring cluster possesses population, which is
sufficient. This report commenced with a definition of clustering, which is responsible for data
grouping of various objects, which is very much based on the information obtained from the
data, which in turn describes the objects alongside its relationship. The report further covers
the various application of clusters as well as its numerous types, accompanied by a pictorial
view of the clustering types.

Page | 1

TABLE OF CONTENT

Abstract ...................................................................................................................................... 1
1.0 Advantages and Disadvantages of Treating Clustering as an Optimization Problem. ........ 3
1.1 Clustering Types .................................................................................................................. 4
1.2 Clustering Application ......................................................................................................... 4
1.

Hierarchy Clustering Type .............................................................................................. 5

2. Well-Separated Clustering .............................................................................................. 5
3.

Centre-Based Cluster ...................................................................................................... 5

4.

Contiguous Clustering .................................................................................................... 6

5.

Density-Based Clustering ............................................................................................... 6

6.

Conceptual Clustering ..................................................................................................... 6

2.1 Pros, Cons Hierarchial, Density-Based, And Partition Clustering. ..................................... 7
Hierarchical Clustering .......................................................................................................... 7
Density-Based Clusters .......................................................................................................... 7
Partitioning Clustering ........................................................................................................... 7
3.0 Treating Clustering as an Optimization Problem................................................................. 8
4.0 Conclusion ........................................................................................................................... 9
References ................................................................................................................................ 10

Page | 2

1.0 THE ADVANTAGES AND DISADVANTAGES OF TREATING CLUSTERING
AS AN OPTIMIZATION PROBLEM.
What is cluster analysis?
Cluster analysis can be defined as that which is responsible for data grouping of various objects,
which is very much based on the information obtained from the data, which in turn describes
the objects alongside its relationship. The primary idea is that these objects embedded in the
group remain similar or possibly related to each other and also entirely dissimilar or in this case
unrelated to the other group’s objects. The higher the homogeneity or similarities, within a
group, the better clustering distinct. The higher the dissimilarities within a group the higher the
chance of clustering distinct.
The ideology behind the word “clustering” is not properly defined, in numerous applications.
To properly grasp the struggle of what exactly makes up a cluster, see the figure below, this
figure depicts twenty points as well as three (3) dissimilar ways of allotting them into clusters.

Figure 1.0: Image representing numerous ways of clustering similar points.
The markers shape depicts the members of the clusters. As displayed in Figure (b) and (d), the
data is further distributed into two (2) and six (6) parts, in that order. From the image above
(figure 1.0), we can easily conclude that the cluster definition is very much imprecise, also the
superlative definition only depends on the data nature as well as the results desired.
Cluster analysis can be tied to other forms of techniques, which are utilized to divide objects
data into groups. Clustering can be observed as a form of classification, that is, it generates
objects labelling with cluster or class labels.

Page | 3

1.1 CLUSTERING TYPES
It is important to note that, a collection of clusters is regarded as clustering. This section of the
report will focus on the different types of clusters. Here the various clustering types will
properly be distinguished. These clustering types include:
1. Hierarchy clustering type.
2. Partitioning clustering type.
3. Fuzzy clustering type.
4. Density-based clustering type.
5. Well-separated clustering type.

Figure 1.1: Clustering algorithm types
1.2 CLUSTERING APPLICATION
The application of clustering is broad, find highlighted below some of the clustering
application.
•

It is utilized in the recognition of patter.

•

Processing of image.

•

Classification of the document.

•

Economic science most especially in market research.

•

Analysis of spatial data.
Page | 4

1. Hierarchy...